Professional Documents
Culture Documents
CHILD and
ADOLESCENT
PERSONALITY
and BEHAVIOR
Paul J. Frick
Christopher T. Barry
Randy W. Kamphaus
Fourth Edition
Clinical Assessment of Child and
Adolescent Personality and Behavior
Paul J. Frick • Christopher T. Barry
Randy W. Kamphaus
Clinical Assessment of
Child and Adolescent
Personality and
Behavior
Fourth Edition
Paul J. Frick Christopher T. Barry
Department of Psychology Washington State University
Louisiana State University Pullman, WA, USA
Baton Rouge, LA, USA
Randy W. Kamphaus
College of Education
University of Oregon
Eugene, OR, USA
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To the memory of my parents, Jane and John Frick, who were
my intellectual inspiration and my biggest fans (Paul J. Frick).
Psychologists offer a wide array of services to the public. Among these ser-
vices, psychological assessment of personality and behavior continues to be
a central activity. One main reason is that other mental health professionals
often do not possess a high level of competence in this area. When one con-
siders psychologists who serve children and adolescents, psychological
assessment seems to take on an even greater role. It follows, then, that com-
prehensive and enlightened graduate-level instruction in assessment should
be a high priority for educators of psychologists who are destined to work
with youth.
This book is an outgrowth of our efforts to improve our own instruction of
child and adolescent assessment skills. We found that existing textbooks were
not serving us well in our training of future clinical assessors. Most of them
were encyclopedic edited volumes that were (1) uneven in the quality across
chapters and/or (2) not geared either in format or level of presentation for
beginning graduate instruction. The few single-authored or coauthored vol-
umes available tended to lack the breadth of coverage we deemed necessary.
Some focused largely on theoretical issues related to psychological testing,
with minimal discussion of practical applications and use of specific tests.
Others focused solely on summaries of individual tests, without reviewing
the theoretical or empirical context within which to use the tests appropri-
ately. Hence, this volume reflects our desire to provide a more helpful tool for
instruction—one that provides a scientific context within which to understand
psychological testing with children and adolescents and that translates this
scientific context into practical guidelines for using individual tests in clinical
practice.
Among our specific objectives for this text are the following:
vii
viii Preface
• Update the content to reflect new research that can inform clinical
assessments
• Update the content to reflect revisions and updates of the measures
reviewed
• Update the content to reflect changes in major systems for making diagno-
ses of children and adolescents and to reflect changes in the standards for
practice
• Enhance the readability of the text based on comments from students and
colleagues
• Add a new chapter on mental health screening, a topic that has become an
increasingly important focus of assessments to aid in early identification
and prevention of serious mental health problems in children and
adolescents
With such changes, we hope to retain the readability and usefulness of previ-
ous editions, while still keeping the content reflective of the most current
research.
xi
xii Contents
Hyperactivity�������������������������������������������������������������������������������� 181
Depression������������������������������������������������������������������������������������ 182
Other Peer-Referenced Techniques�������������������������������������������������� 182
Conclusions�������������������������������������������������������������������������������������� 183
10 Projective Techniques���������������������������������������������������������������������� 185
The Controversy Surrounding Projective Techniques��������������������� 185
Clinical Technique or Psychometric Test?���������������������������������� 186
Projection of Behavioral Sample?����������������������������������������������� 189
Summary�������������������������������������������������������������������������������������� 190
Inkblot Techniques�������������������������������������������������������������������������� 190
The Exner Comprehensive System for Rorschach
Interpretation�������������������������������������������������������������������������������� 191
Thematic (Storytelling) Techniques������������������������������������������������ 195
General Interpretation of Thematic Techniques�������������������������� 196
Roberts Apperception Technique for Children���������������������������� 197
Sentence Completion Techniques���������������������������������������������������� 200
Features of SCTs�������������������������������������������������������������������������� 200
Drawing Techniques������������������������������������������������������������������������ 203
Draw-a-Person Technique������������������������������������������������������������ 203
House-Tree-Person���������������������������������������������������������������������� 205
Kinetic Family Drawing�������������������������������������������������������������� 205
Psychometric Cautions for Drawing Techniques������������������������ 205
Conclusions�������������������������������������������������������������������������������������� 206
11 Structured Diagnostic Interviews �������������������������������������������������� 209
History���������������������������������������������������������������������������������������������� 209
Overview������������������������������������������������������������������������������������������ 210
Commonalities Across Interview Schedules�������������������������������� 210
Major Sources of Variation Across Interview Schedules ������������ 212
Evaluation of Diagnostic Interviews������������������������������������������������ 213
Advantages���������������������������������������������������������������������������������� 213
Disadvantages������������������������������������������������������������������������������ 214
Recommendations for Use of Structured Interviews ���������������������� 215
Focus on Schedule for Affective Disorders
and Schizophrenia for School-Age Children
(KSADS-PL; Kaufman et al. 2016) ������������������������������������������������ 217
Development�������������������������������������������������������������������������������� 217
Structure and Content������������������������������������������������������������������ 217
Administration ���������������������������������������������������������������������������� 217
Conclusions�������������������������������������������������������������������������������������� 218
12 Assessing Family Context���������������������������������������������������������������� 221
Why Is Assessing Family Context Important?�������������������������������� 221
The Causal Role of Familial Influences�������������������������������������� 221
Family History and Differential Diagnosis���������������������������������� 223
Interpreting Information Provided by the Parent ������������������������ 224
Assessing Family Functioning�������������������������������������������������������� 226
Parenting Styles and Practices �������������������������������������������������������� 229
xvi Contents
xxi
Historical Trends
1
described by her parents as either shy or extra- Table 1.1 Early descriptions of the Big Five personality
dimensions (Goldberg, 1992)
verted. The shy (introverted in psychological
terms) child may tend to cope with stressful situ- Factor I: Surgency (or introversion-extraversion)
ations by withdrawing from social contact, Unenergetic vs. energetic
Silent vs. talkative
whereas the extravert readily approaches social
Timid vs. bold
situations. For parents and psychologists alike,
Factor II: Agreeableness (or pleasantness)
these traits are often thought to have value for Cold vs. warm
predicting human behavior, because of the pre- Unkind vs. kind
sumption of trait stability across time and, in Uncooperative vs. cooperative
many cases, environments. For example, because Factor III: Conscientiousness (or dependability)
of presumed trait stability, parents may take spe- Disorganized vs. organized
cial precautions to ensure that the shy child Irresponsible vs. responsible
adapts well to the social aspects of attending a Negligent vs. conscientious
new school, by asking one of their child’s friends Factor IV: Emotional stability (versus neuroticism)
Tense vs. relaxed
who attends the same school to accompany the
Nervous vs. at ease
child on the first day. Similarly, a stable tendency Factor V: Culture, intellect, openness, or
to be shy or introverted should manifest itself in sophistication
numerous social situations such as interactions in Unintelligent vs. intelligent
the neighborhood, at church, and in ballet class. Unanalytical vs. analytical
Personality traits, then, are characterized by lon- Unreflective vs. reflective
gitudinal and situational stability, not unlike
other enduring characteristics of person such as
intelligence, height, and activity level. as personality descriptors. This item format is in
direct contrast to the more commonplace true-
false item format that is typical of many psycho-
he Big Five Personality Traits
T logical tests.
(Factors)
infants and young children. In this conceptualiza- Second, behavioral assessment methods are
tion, personality may be viewed as being super- distinguished from medical models of assess-
imposed on a person’s temperamental foundation. ment (Kaplan & Saccuzzo, 2018). The medical
This distinction between temperament and per- model assumes that symptoms are caused by
sonality, however, is not universally agreed upon underlying conditions, and it is the medical con-
(Goldsmith et al., 1987). dition that must be measured, diagnosed, and
treated in order to remove the symptoms (see
Chap. 3 for a more extended discussion of the
Behavior medical model approach to classification and
assessment). In direct contrast, behavioral assess-
In contrast to temperament and personality trait ment emphasizes the measurement and treatment
assessment, the assessment of behavior focuses of the symptoms or behavior itself and makes no
on the measurement of observable behaviors, assumptions regarding underlying cause.
although the definition has been broadened to Third, behavioral assessment places a pre-
include cognitions as a type of behavior. For most mium on the assessment of discrete behaviors.
purposes, Martin (1988) provides a useful defini- For example, behavioral assessment may empha-
tion of behavior as it applies to current child psy- size the measurement of finger tapping while
chological assessment. completing seatwork in school, as opposed to
When applied psychologists speak of behavior, aggregating several behaviors to form a test or
they are usually referring to that range of human scale with several items that measure “motor
responses that are observable with the naked eye activity” in the classroom. This situation began to
during a relatively brief period of time in a particu- change with the work of Thomas Achenbach,
lar environment. This conception of behavior rules
out biochemical and neurological events, for Keith Conners, Cecil Reynolds, and others (see
example, because they are not observable by the Chaps. 6 and 7), all of whom began combining
unaided eye. Behavior is differentiated from traits behaviors into dimensions or “syndromes” of
or dispositions because the latter may only be seen behavior on rating scales. Such dimensions begin
if behavior is aggregated over relatively long peri-
ods of time and in a number of environmental con- to blur the distinction between behavioral and
texts. Classical examples of observed behaviors of trait-based assessment methods.
interest to child psychologists include tantrum As we suggested in the preface, we think that it
behavior among young children, aggressive inter- is premature to reify any of these approaches as the
actions with peers, attempts at conversation initia-
tion, and so forth. (p. 13) ultimate or sufficient method for assessing chil-
dren’s psychological adjustment. We merely seek
There are, therefore, several distinguishing progress in our methods and theories. This cau-
features of behavioral assessment methods that tious approach seems warranted as the distinctions
differentiate them from trait assessment mea- between the methods have become blurred as the
sures. First, behavioral assessment methods science of assessment emerges. Furthermore, we
have a different theoretical foundation and think that each approach may be more or less help-
associated set of premises. Behavioral assess- ful for answering particular assessment questions.
ment methods draw heavily on the theory and Clearly, some of the questions directed at psychol-
research tradition of operant conditioning as ogists are trait-based, whereas others require the
exemplified by the work of B. F. Skinner measurement of distinct behaviors. For example, a
(1963). This research tradition also emanates parent who asks, “Will my child ever become
primarily from laboratory research on how ani- more outgoing like his sister?” is asking for trait
mals and people learn, as opposed to clinical assessment, but the parent who queries, “How can
practice; thus, it is often considered to be more I get him to stop wetting the bed?” may require
empirically based. assessment with a behavioral focus.
4 1 Historical Trends
World War I applications of psychological testing occurs when the ego, faced with unacceptable
wishes or ideas, thrusts them out onto the external
showed that psychology could make practical world as a means of defense. In projection the indi-
contributions to society by identifying, accu- vidual attributes his or her own thoughts and
rately and time-efficiently, those in need of men- actions to someone else. Thus, if one’s own faults
tal health services. This work was continued after or feelings are unacceptable to the ego, they may
be seen as belonging to someone else; in the pro-
World War II, when the mental health needs of cess, the material may become distorted or remain
citizens, veterans in particular, were the focus of partially repressed. From such a perspective, pro-
greater attention. In the postwar years, the US jective material would not be seen as direct repre-
Veteran’s Administration began to hire psycholo- sentation of aspects of the personality, certainly
not with the sort of one-to-one correspondence that
gists in large numbers to diagnose and treat veter- the first meaning of projection implies. (p. 57)
ans suffering from significant emotional
disturbance. Psychologists brought their psycho- Given that test stimuli (e.g., inkblots, pictures,
metric expertise to bear again by contributing words) on projective tests were by nature ambig-
new methods to the diagnostic process. The uous and could be interpreted in multiple differ-
increased need for postwar mental health ser- ent ways and not clearly linked to known
vices, therefore, created the fertile ground in personality traits per se, projective testing
which personality testing flourished. As depended heavily on the explicit or implicit per-
Kleinmuntz (1967) noted, “The most popular sonality theory that was favored by the test devel-
personality tests of the past 30 years grew out of oper. Theory was necessary to determine the
the need to diagnose or detect individuals whose underlying nature or cause of the projected
behavior patterns were psychopathological” thoughts, emotions, or behaviors. The most pop-
(p. 10). The use of personality tests after the first ular theory of the post-World War I era was psy-
and second world wars expanded beyond diagno- chodynamic personality theory as espoused by
sis and screening into many areas including Sigmund Freud (1936) and others. Psychody
counseling, personnel selection, and personality namic personality theory provided a useful theo-
research (Kleinmuntz, 1967). retical framework for the development of
projective assessment measures due to the con-
cepts of repression, projection, and other con-
Projective Techniques structs that are entirely consistent with the use of
atypical (ambiguous) test stimuli for identifying
One type of test that emerged early in the history personality traits.
of personality assessment is the projective test.
The central assumption of projective techniques
is the “projective hypothesis,” which is described Association Techniques
in more detail in Chap. 10. The projective hypoth-
esis rests on the assumption that, as individuals One of the earliest forms of projective testing
try to describe ambiguous stimuli, they will used association techniques, such as word asso-
reveal some aspects of their personality or ciation methods. Such methods can be traced as
motives through their perceptions of the stimuli far back as the work of Aristotle (DuBois, 1970).
(Butcher, Mineka, & Hooley, 2007). Chandler In more recent history, Sir Francis Galton began
(1990) elucidates the nature of the projective studying association techniques as early as 1879.
hypothesis as follows: Galton’s contribution to the study of association
Projection, in common usage, means to cast for- was his introduction of scientific rigor to the
ward. In this sense, projection implies a direct enterprise. He used experimental methods to
extension of psychological characteristics onto the study association methods, including quantitative
outer world. But projection also has a specific scaling of the results (DuBois, 1970).
meaning within psychoanalytic theory. Freud
(1936) used the term to refer to the process that Subsequently, Kraepelin, Wundt, Cattell, Kent,
and Rosanoff studied the associations of patients
6 1 Historical Trends
and research participants to word lists, recording which used visual stimuli (i.e., pictures and ink-
such variables as response time and type of asso- blots, respectively) in lieu of word lists to elicit
ciation. They developed a list of 100 stimulus associations.
words and systematically recorded the associa-
tions of 1000 normal subjects (DuBois, 1970).
This effort represents an important initial attempt Thematic Apperception Test
at developing norms to which researchers and cli-
nicians could compare the responses of clinical The Thematic Apperception Test (TAT) of Henry
subjects. A. Murray constitutes a prototypical example of
The renowned psychoanalyst Carl Jung made a projective device. Charles E. Thompson (1949)
extensive use of association techniques for the summarized the central tenet of the projective
study of personality. In an address at Clark approach in the following quote taken from his
University in 1909, he described his research 1949 adaptation of Murray’s TAT.
efforts in detail and provided some insight into If the pictures are presented as a test of imagina-
the types of interpretations commonly made of tion, the subject’s interest, together with his need
these measures. Jung described his association for approval, can be so involved in the task that he
word list as a formulary. His list consisted of 54 forgets his sensitive self and the necessity of
defending it against the probing of the examiner,
words including head, to dance, ink, new, foolish, and, before he knows it, he has said things about an
and white. According to Jung (1910), normality invented character that apply to himself, things
could be distinguished from psychopathology which he would have been reluctant to confess in
with this formulary using variables such as reac- response to a direct question. (p. 5)
tion time and response content. In his speech, he
provided a transcript of the responses of a “nor- The TAT is unique among projective measures in
mal” individual and of a “hysteric.” A sampling that it has traditionally been interpreted qualita-
of their associations to the formulary follows: tively, even though quantitative scoring methods
are available (Kleinmuntz, 1967). Murray’s orig-
Stimulus Normal Hysterical
inal approach to TAT scoring was entirely quali-
To sin Much This is totally unfamiliar to
me; I do not recognize it
tative and psychoanalytically based. He proposed
To pay Bills Money the following categories for analyzing the charac-
Bread Good To eat teristics of the stories given by the sub-
Window Room Big ject (Thompson, 1949).
Rich Nice Good, convenient
Friendly Children A man 1. The Hero. This is the person with whom the
subject seems to identify. The hero may share
As noted previously for other measures, reaction characteristics such as age, gender, occupa-
time to the stimulus words was also interpreted tion, or other features with the subject that aid
by Jung. He gave a glimpse of one such interpre- identification. The hero’s traits should be eval-
tation in the following quote: uated to determine the self-perceptions of the
The test person waives any reaction; for the subject including superiority, intelligence,
moment he totally fails to obey the original instruc- leadership, belongingness, solitariness, and
tions, and shows himself incapable of adapting quarrelsomeness.
himself to the experimenter. If this phenomenon 2. Needs of the Hero. Needs may include those
occurs frequently in an experiment, it signifies a
higher degree of disturbance in adjustment. (p. 27) for order, achievement, and nurturance.
3. Environmental Forces. Factors that affect the
These early word association methods set the hero are also referred to as press. An example
stage for the development of other association would involve scoring “aggression”, if the
(projective) techniques, such as the Thematic hero’s property or possessions were destroyed
Apperception Technique and Rorschach’s test, in a story.
Projective Techniques 7
4. Outcomes. The success of the hero and the Table 1.2 Rorschach’s original research sample
hero’s competencies are assessed by evaluat- Sample N
ing the outcomes of stories. Normal, educated 55
5. Themas. Themas assess the interplay of needs Normal, uneducated 62
and presses, and they reveal the primary con- Psychopathic personality 20
Alcoholic cases 8
cerns of the hero.
Morons, imbeciles 12
6. Interests, Sentiments, and Relationships. For
Schizophrenics 188
this aspect of scoring, the examiner records Manic depressives 14
the hero’s preferences for specific topics. Epileptics 20
Paretics 8
Murray’s qualitative scoring system for the TAT Senile dements 10
is a classic example of systems that dominated Arteriosclerotic dements 5
the early interpretation of projective devices. Korsakoff and similar states 3
Numerous quantitative scoring systems followed Source: Adapted from Rorschach (1951)
Murray’s original work, as exemplified by scor-
ing systems eventually developed for Rorschach folded in half. Rorschach was not, however,
test. interested in the content of the subject’s response
to the inkblots. Rather, he was interested in the
form of the response (or its function). Some func-
The Rorschach tions of interest included the number of responses,
perception of color or movement, and perception
Hermann Rorschach was a major figure in Swiss of the whole versus the parts. These and other
psychiatric research who began his work as a characteristics of Rorschach responses continue
physician in 1910. He married a Russian col- as part of modern scoring systems (see Chap. 10).
league who became his collaborator Rorschach first offered his method as an
(Morgenthaler, 1954). He served as a physician experiment. His original sample is described in
in a hospital in Herisau until his death from com- Table 1.2. He expressed a desire for larger sample
plications of appendicitis in 1922. His death was sizes but noted that the number of experiments
described as a critical blow to Swiss psychiatry. was limited because the stimulus figures were
In a 1954 eulogy to Dr. Rorschach, published in damaged by passing through hundreds of hands.
the English translation of his original work, Rorschach’s legacy, his original inkblots, and
Morgenthaler attempted to describe Rorschach many of the associated scoring criteria remain
for future generations. influential as the test continues to enjoy popular-
Flexibility of character, rapid adaptability, fine ity. Several scoring systems have been offered for
acumen, and a sense for the practical were com- the Rorschach (e.g., Rorschach Performance
bined in Hermann Rorschach with a talent for Assessment System; Viglione, Blume-Marcovici,
introspection and synthesis. It was this combina- Miller, Giromini, & Meyer, 2012) with the Exner
tion which made him outstanding. In addition to
this rare nature, which tempered personal emo- Comprehensive System (Exner & Weiner, 1982)
tional experience with practical knowledge, he contributing most to the continuing usage of the
possessed sound traits of character most valuable instrument.
in a psychiatrist. Most important of these were an
unerring tendency to search for the truth, a strict
critical faculty which he did not hesitate to apply to
himself, and a warmth of feeling and kindness. Sentence Completion Techniques
(p. 9)
Sentence completion techniques are venerable
Rorschach’s approach to personality assessment personality assessment methods of the associa-
was novel in many respects. The test stimuli used tion tradition that can trace their roots to Payne
were inkblots placed on paper that was then (1928). The sentence completion method, however,
8 1 Historical Trends
obtained a substantial boost in popularity because content that was the focus of such interpretations
of its use by the US Office of Strategic Services included:
(OSS), the forerunner of the Central Intelligence
Agency (CIA). Henry Murray was the coordina- Size of the person depicted
tor of a sophisticated OSS assessment effort. Placement on the page (bottom, top, corner, etc.)
About 60 military assessment stations, staffed by Stance (vertical, horizontal, balanced, etc.)
American psychologists, were situated in the Line quality (heavy, light, etc.)
United States and abroad to screen recruits for Shading
sensitive and dangerous assignments. Some of Erasures
the methods used in this ambitious program are Omissions (missing body parts)
described in the following quote from DuBois Distortion (poor proportion of body parts)
(1970).
In one of the stations near Washington, recruits in Various interpretations have been associated with
fatigue uniforms assumed a false identity and these and other content variables over the years
developed a cover story, which the staff members (see Chap. 10). Heavy lines, for example, have
during the three-day stay endeavored to break. The been associated with assertive and aggressive
procedures described in a comprehensive report
(OSS, Assessment Staff, 1948) were varied: casual individuals, and light lines have been viewed as
conversations, searching interviews, the sentence being indicative of passive individuals (Koppitz,
completion test, questionnaires about health and 1968).
working conditions and personal history, conven- Apperception or storytelling were also adapted
tional aptitude tests such as map memory and
mechanical comprehension, and a number of situ- for use with children and adolescents. For exam-
ational tests. (p. 111) ple, one of the most well-known apperception
tests used to assess adult personality, the TAT,
After World War II, the sentence completion was adapted for use with children using the
technique continued to enjoy some favor among Children’s Apperception Test (CAT; Bellak &
psychologists. The well-known Rotter Incomplete Bellak, 1949) and designed for ages 3–10. The
Sentences Blank was published in 1950 (Rotter CAT consists of ten pictures with animals as
& Rafferty, 1950). Sentence completion methods stimuli in contrast to the TAT’s depictions pri-
have a lengthy history of use with children and marily of people. The Rorschach was also used
adolescents as well. They have also enjoyed with children, and several compendiums of child
worldwide use, including outside of the context responses were published to aid interpretation
of clinical assessments (e.g., Kujala, Walsh, (e.g., Ames, Metraux, Rodell, & Walker, 1974).
Nurkka, & Crisan, 2014).
Objective Tests
Projective Techniques for Children
As will be discussed greater detail in Chap. 10
The use of projective techniques with children while projective tests dominated early personal-
dates back to the early part of the twentieth cen- ity assessment of both adults and children and are
tury when Florence Goodenough began to study still commonly used in clinical assessments
children’s human figure drawings (DuBois, today, there emerged in the 1950s and 1960s a
1970). Goodenough noted, as did others, that growing dissatisfaction with the poor reliability
children’s drawings were affected by their emo- associated with many methods of scoring project
tionality. The typical paradigm for drawing tech- techniques. This dissatisfaction is nicely illus-
niques has been to have the child draw a picture trated by the comments of Paul Meehl provided
of a person. The content of the drawings was in Box 1.1. Further, many of the scores from
interpreted as a measure of child adjustment and these techniques, even if assessed reliably, did
personality (Swensen, 1968), and some typical not correlate well with emotions and behaviors
Objective Tests 9
The MMPI (Hathaway & McKinley, 1942) cards were separated by the patient into three cat-
differed from its predecessors (such as the egories: true, false, and cannot say. The ten clini-
Personal Data Sheet) in at least one fundamental cal scales of the original version included are
way. It was one of the first tests to use an empiri- provided in Table 1.3. The MMPI has undergone
cal approach to objective personality test devel- many changes since its inception, with the most
opment. Most tests of the day used a priori or recent edition entitled the MMPI-2. In fact, some
rational-theoretical approaches (Martin, 1988). of the scale names (e.g., psychasthenia) had
Rational approaches, as the name implies, depend already fallen into disuse at about the time of the
heavily on the test author’s theory of personality original publication in 1942 (Kleinmuntz, 1967).
for many aspects of test construction, including The MMPI not only profoundly influenced
item development and scoring methods. adult personality assessment but also influenced
Empirical approaches, on the other hand, make child assessment practice including being the
greater use of empirical data to make such deci- stimulus for the development of the Personality
sions (see Chap. 2 for a more detailed discussion Inventory for Children (PIC) in the 1950s. The
of this distinction). PIC was based on a pool of 600 items; hence, it
The MMPI used an item selection method was comparable in length to the MMPI. Also like
called empirical criterion keying (Anastasi, the MMPI, items were placed on scales based on
1988). Simply stated, this method involved empirical methods rather than based on any the-
selecting items that meet an empirical criterion. ory of personality. A central difference between
In the case of the MMPI, items were selected if the MMPI and the PIC was the informant. The
they were able to routinely differentiate clinical PIC was not a self-report measure. Instead, a par-
groups from samples of “normal” subjects and ent rated the child’s behavior. Lachar (1990) gave
distinguish clinical groups from one another. For the following rationale for this decision.
example, items for the psychasthenia scale (a Selection of the parents as the source of PIC test
scale designed to assess anxiety-related problems responses helps overcome two of the major obstacles
such as obsessions and fears) were selected based posed by requesting the referred child or adolescent
on a clinical group of 20 cases, the results for to respond to numerous self-report descriptions in
order to obtain a multiple-scale objective evaluation.
which were compared to “normals” and other The majority of children seen by mental health pro-
clinical groups in order to identify items that best fessionals in a variety of settings appear for such an
differentiated the target clinical group from the evaluation because of their noncompliant behaviors
others. Given that these items did not have to cor-
respond to any known theory of personality, only Table 1.3 The original scales from the MMPI
to provide important for distinguishing normal Clinical scales
from clinical groups, items often did not have 1. Hypochondriasis
“face validity.” This led to some derision as illus- 2. Depression
trated in the 1965 issue of American Psychologist 3. Hysteria
(p. 990), in which a number of questions were 4. Psychopathic deviate
published to poke fun at this method of personal- 5. Masculinity-femininity
ity assessment. They included: 6. Paranoia
7. Psychasthenia
8. Schizophrenia
When I was younger I used to tease vegetables.
9. Hypomania
I think beavers work too hard. 10. Social introversion
I use shoe polish to excess. Validity scales
When I was a child I was an imaginary 1. Question scale
playmate. 2. Lie scale
3. F scale
The original version of the MMPI consisted of 4. Correction scale
550 statements printed on separate cards. The Source: Kleinmuntz (1967)
Objective Tests 11
and/or documented problems in academic achieve- ment (Kleinmuntz, 1967). Reviewers of the day
ment, most notably in the development of reading
skills. Therefore, it seems unlikely that a technique
found many reasons to recommend the
requiring such children to read and respond to a large Wittenborn, including a thorough research base
set of self-descriptions will find broad acceptance in (Eysenck, 1965) and easy administration and
routine clinical practice. (p. 299) scoring (Lorr, 1965). There was considerable
concern, however, about overlapping scales. The
hebephrenic schizophrenic and schizophrenic
Rating Scales excitement scales correlated at 0.88, and the par-
anoid condition and paranoid schizophrenic
As this statement of Lachar illustrates, the use of scales correlated at 0.79. Based on these data,
informants to rate children’s emotions and behav- Eysenck (1965) and Lorr (1965) recommended
iors was largely based on a concern about a chil- that these scales be combined to reflect this
dren’s ability to provide accurate information. overlap.
However, parent and teacher ratings of children’s Other rating scales of adult psychopathology for
behavior and emotions can also trace their roots to use in inpatient settings included the Hospital
the assessment of adult psychopathology in hospi- Adjustment Scale (McReynolds, Ballachey, &
tal settings. Conceptualized as one type of obser- Ferguson, 1952) and the Inpatient Multi
vational method, rating scales were developed in dimensional Psychiatric Scale (Lorr, 1965), a rating
the 1950s for use by nurses and other caretakers of symptomatology completed by the clinician after
who worked closely with adult patients for a diagnostic interview. Such measures probably fell
extended periods of time. Again, the development into decline for many reasons, one of the most prom-
of these scales was prompted by concerns that inent being the deinstitutionalization movement of
some patients may not be capable for providing the 1970s. These instruments did, however, clearly
accurate information on their symptoms. demonstrate the utility of ratings of behavior as prac-
One of the first such measures was the tical and useful assessment tools. This set the stage
Wittenborn Psychiatric Rating Scales (1955). for the development of parent and teacher rating
According to Lorr (1965), the scales were scales of child behavior.
designed for recording currently observable With the increasing use of objective measures
behavior and symptoms in hospitalized mental of children’s personality, often in the form of
patients. The Wittenborn could be completed by informant ratings, this allowed for researchers to
a social worker, psychologist, psychiatrist, nurse, collect large samples of children and to test
attendant, or other individual familiar with the whether there were common dimensions in how
patient’s day-to-day behavior. The original scale children were rated. In a 1978 article in
consisted of 52 symptoms that were combined to Psychological Bulletin, Thomas Achenbach and
yield 9 scores for acute anxiety, conversion hys- Craig Edelbrock introduced many clinicians to
teria, manic state, depressed state, schizophrenic the terms internalizing and externalizing psycho-
excitement, paranoid condition, paranoid schizo- logical disorders of childhood. These dimensions
phrenic, hebephrenic schizophrenic, and phobic were based on an extensive empirical analysis
compulsive. An item assessing withdrawal (typically using factor analysis) of parent and
included the following options: teacher behavior problem rating scales across
many large samples. Children experiencing
No evidence of social withdrawal adjustment difficulties of the internalizing variety
Does not appear to seek out the company of other have also been described as overcontrolled, with
people problems of inhibition, anxiety, and, perhaps,
Definitely avoids people shyness (Edelbrock, 1979). On the other hand,
children with externalizing problems have been
The Wittenborn was used for diagnostic purposes described as undercontrolled with difficulties
as well as for the design and evaluation of treat- such as aggression, conduct problems, and
12 1 Historical Trends
acting-
out behavior (Edelbrock, 1979). The which is still a contentious issue in research
veracity of the broad internalizing and external- (Achenbach, Rescorla, & Ivanova, 2012).
izing categorizations of child psychopathology is
supported by many factor-analytic investigations
of both parent and teacher rating scales alike and iagnostic and Statistical Manual
D
has withstood the test of time, with numerous of Mental Disorders
independent research studies supporting this dis-
tinction on samples throughout the world The two most common systems for classifying
(Gomez, Vance, & Gomez, 2014; Ivanova et al., mental health disorders in both adults and chil-
2007). As result, these two broad dimensions of dren are the International Classification of
child adjustment are still the basis for many cur- Disease (ICD) published by the World Health
rent models for classifying childhood disorders Organization, which is currently on its 11th edi-
(Lahey, Krueger, Rathouz, Waldman, & Zald, tion (ICD-11; World Health Organization, 2018)
2017). and the Diagnostic and Statistical Manual of
Mental Disorders published by the American
Psychiatric Association, which is currently on its
Assessment for Diagnosis fifth edition (DSM-5; American Psychiatric
Association, 2013). The American Psychiatric
As mental health problems became recognized as Association first began its efforts to standardize
conditions worthy of professional treatment, diagnostic procedures in 1917 as the American
rather than moral failure, the need for diagnostic Medico-Psychological Association. However, the
systems became more pressing. Consistent (i.e., first edition of the DSM (Diagnostic and
reliable) diagnosis was necessary for communica- Statistical Manual) appeared in 1952. It has been
tion among clinicians and for the conduct of epi- periodically updated since then, as summarized
demiological research and other scientific in Table 1.4. In Chap. 3, we summarize some of
investigations. Further, and perhaps most impor- the major changes in this classification system,
tantly, if mental health problems were to be con- especially the most recent revisions, and the
sidered for coverage by health insurance, a system impact of these changes on the clinical assess-
for systematically documenting the need for treat- ment of children and adolescents. However, in
ment was needed. This led to a very different the context of this historical discussion, its major
approach to clinical assessment for both adults
and children. That is, rather than designing assess-
ments to assess personality based on some theory Table 1.4 Chronology of diagnostic systems developed
of personality dynamics (e.g., projection) or based under the auspices of the American Psychiatric Association
on some empirical method (e.g., factor analysis), 1917 Classification of Mental Disease
the focus became on finding the best methods for 1933 Standard Classified Nomenclature of Disease
determining who needed treatment and what treat- 1952 Diagnostic and Statistical Manual of Mental
Disorders I (DSM-I)
ment should be used. This resulted in a number of 1968 Diagnostic and Statistical Manual of Mental
important issues in the classification and assess- Disorders II (DSM-II)
ment of children and adolescents that are dis- 1980 Diagnostic and Statistical Manual of Mental
cussed in more detail in Chap. 3. Of particularly Disorders III (DSM-III)
note was that this focus resulted in the need for 1987 Diagnostic and Statistical Manual of Mental
Disorders III-Revised (DSM-III-R)
assessments to answer the question “Does this
1994 Diagnostic and Statistical Manual of Mental
child need treatment?” As a result, it promoted a Disorders IV (DSM-IV)
categorical (i.e., is the child disordered or not) 2000 Diagnostic and Statistical Manual of Mental
rather than dimensional (i.e., how high is the child Disorders IV-Text Revision (DSM-IV-TR)
on the externalizing and internalizing dimension 2013 Diagnostic and Statistical Manual of Mental
of psychopathology) approach to assessment, Disorders 5 (DSM-5)
Assessment for Diagnosis 13
influence on the clinical assessment of children or health factors; (b) an inability to build or main-
tain satisfactory relationships with peers and teach-
and adolescent is the emphasis on having mea- ers; (c) inappropriate types of behavior or feelings
sures that allow the assessor to determine if a under normal circumstances; (d) a general perva-
child or adolescent meets criteria for a mental sive mood of unhappiness or depression; or (e) a
disorder and, if so, what one best describes the tendency to develop physical symptoms or fears
associated with personal or school problems.
child or adolescent in a way that leads to effective (Federal Register, 42, 474, 1977)
treatment .
Also like the DSM, the requirement of the IDEA
to make categorical decisions on the presence of
IDEA and Special Education an emotional disturbance (the word severe has
now been removed) has long been the target of
While the DSM has guided the diagnosis of criticisms that it is invalid, restrictive, or other-
children in mental health settings in the United wise flawed (Forness & Kritzer, 1992). Bower
States, the 1974 Education of Handicapped (1981), the recognized developer of the concep-
Children’s Act, better known as Public Law tual basis of the IDEA diagnostic catego-
94–142 (IDEA), and its reauthorization, the ries, raised these concerns about the system.
Individuals with Disabilities Education “Section ii [which excludes the socially mal-
Improvement Act (IDEIA) in 2004, has been adjusted from the IDEA act] is, one would guess,
the primary guide for diagnosis in schools a codicil to reassure traditional psychopatholo-
across the country. This landmark legislation gists and budget personnel that schizophrenia
mandated special education and related ser- and autism are indeed serious emotional distur-
vices for children classified as having an emo- bances on the one hand, and that just plain bad
tional disturbance. With its passage in the boys and girls, predelinquents, and sociopaths
1970s, this law effectively mandated public will not skyrocket the costs on the other hand. It
schools in the United States to identify and is clear what these modifications and additions
serve children with behavioral or emotional were intended to do. It is perhaps not clear what
problems, many of which had previously been such public policy and fiscal modifications do to
educated in a variety of settings, including res- the conceptual integrity of the definition and the
idential treatment programs or state mental nature and design of its goals” (p. 56). Despite
hospitals. Consequently, these laws expanded such controversy, the IDEA/IDEIA classification
school-based diagnostic practices to include system remains as the “gold standard” nosology
evaluation for the presence of emotional dis- when determining child eligibility for oft-costly
turbance. Like the DSM, it required some special education and related services.
method for determining if the child required
treatment (i.e., was eligible for educational
accommodations). While the DSM defined this Evidence-Based Assessment
as determining if a child had a “disorder,”
IDEA defined this as determining if a child had A final issue in this historical look at the evolu-
a “severe emotional disturbance.” tion of clinical assessments of children and ado-
Under IDEA, and the subsequent IDEIA with lescents is the recent emphasis on evidence-based
few substantive changes, the classification of assessment. Such an approach is an extension of
severe emotional disturbance was defined as the focus on finding evidence-based treatments
follows: for children. A special issue of Journal of Clinical
The term means a condition exhibiting one or more Child and Adolescent Psychology (2005) served,
of the following characteristics over a long period in many ways, as the catalyst to current work in
of time and to a marked degree which adversely this area. Mash and Hunsley (2005), the editors
affects school performance: (a) an inability to learn of this special issue, defined evidence-based
which cannot be explained by intellectual, sensory,
assessment as the use of :
14 1 Historical Trends
assessment methods and processes that are based • To provide current research findings that
on empirical evidence in terms of both their reli-
ability and validity as well as their clinical useful-
enable students to draw heavily on science as
ness for prescribed populations and purposes. the basis for their clinical practice
(p. 364) • To help in the translation of research into prac-
tice by providing specific and practical guide-
They noted that key challenges in evidence-based lines for clinical practice
assessment include the vast number of assessment
tools available and the need to account for differ-
ent purposes of assessment (e.g., diagnostic deci- Chapter Summary
sions, treatment progress monitoring, risk for
harm to self or others). They described a move- 1. Personality is typically considered to be
ment away from rigid, set test batteries toward composed of traits, a more enduring set of
more problem-specific approaches that are flexible characteristics of the individual.
in the tools that are used so that the best method 2. Formal personality measures emerged as a
can be selected based on specific purpose and the logical outgrowth of other efforts to measure
available research. Further, more practical issues individual differences, most notably the
such as the factors that facilitate or impede par- experimental methods of Wundt, Galton, and
ents’ following assessment recommendations have others.
begun to be empirically investigated and could 3. The Woodworth Personal Data Sheet was
have important implications for guiding assess- published in 1918 as a result of the surge of
ment practice (Mucka et al., 2017). interest in testing potential soldiers.
Our writing of this text was highly influenced 4. The needs for diagnosis created by World War
by this approach to assessment. As noted in our I and World War II provided considerable
introduction, we feel that clinical assessments impetus for the development of personality
need to move away from being highly “test tests.
focused,” which this historical overview has 5. A major assumption underlying projective
shown has been the dominant orientation testing is the projective hypothesis, which
through much of the history of clinical assess- rests on the assumption that using stimuli
ment. In the past, the choice of methods used in that allow for a variety of interpretations will
the practice of psychological assessment has encourage clients to reveal information that
often been driven more by an allegiance to a they otherwise would not share in response
theoretical orientation that underlies a particular to direct questioning.
method of assessment or, even more problem- 6. The Rorschach test stimuli were originally
atic, by an allegiance to a particular assessment inkblots placed on paper that was then folded
technique (Frick, 2000). Instead, we view the in half.
goal of psychological assessment as the mea- 7. The use of projective techniques with chil-
surement of psychological constructs and, for dren dates back to the early part of this cen-
clinical practice, measurement of psychological tury, when Florence Goodenough began to
constructs that have important clinical implica- study children’s human figure drawings.
tions, such as documenting the need for treat- 8. The MMPI was one of the first objective per-
ment or the type of intervention that is most sonality tests to use an empirical approach
appropriate. In keeping with the evidence-based for personality test development and used an
approach to assessment, how best to measure item selection method called empirical crite-
these constructs should be based on the most rion keying.
current research. As a result, our most important 9. The use of informant rating scales for the
objectives for this text were: assessment of child psychopathology traces
Assessment for Diagnosis 15
its roots to the assessment of adult psychopa- 13. Need for treatment (i.e., accommodations) in
thology in hospital settings. schools is largely guided by the 1974
10. In a 1978 article in Psychological Bulletin, Education for All Handicapped Children
Achenbach and Edelbrock introduced the Act, better known as Public Law 94–142,
terms internalizing and externalizing when and its reauthorization, the Individuals with
referring to psychological disorders of child- Disabilities Education Improvement Act
hood. This distinction still has a dominant role (IDEIA), which mandated special education
in how emotional and behavioral problems and related services for children classified as
are classified in children and adolescents. emotionally disturbed.
11. A major issue in guiding clinical assess-
14. A current emphasis in the training of practi-
ments is the need to make decisions on tioners of child and adolescent assessment is
whether a child or adolescent needs treat- on evidence-based approaches which are
ment (i.e., has a disorder). directly informed by developmental science
12. The most common way that disorders are and that translate to recommendations that
diagnosed in mental health settings is the DSM emphasize evidence-based approaches to
(Diagnostic and Statistical Manual of Mental intervention.
Disorders), which first appeared in 1952.
Measurement Issues
2
rather broad definition provides a useful starting intended to represent (Mash & Hunsley, 2005)
point for conceptualizing the great variety of or not reflective of a client for which the clini-
measures available. cian wants to draw conclusions. These factors
The central characteristic of this definition is should always be considered in selecting tests
the notion of standardization of behavioral sam- and interpreting results.
pling. Standardization has at least two meanings:
standardization in the sense of collecting a sam-
ple for the purpose of norm-referencing and stan- Types of Tests
dardization as administration of the measure
according to a consistent set of rules. Most of the How does one identify an instrument that assesses
measures discussed in this volume fit the first personality or behavioral, emotional, and social
notion of standardization in that they are norm- functioning?
referenced. That is, these measures use norm Personality tests have traditionally attempted
groups for gauging a child’s performance in com- to assess personality traits such as introversion,
parison with some standard, typically defined agreeableness, and neuroticism. Traits are usu-
based on the child’s age/developmental level. ally considered to be relatively stable characteris-
Furthermore, the principle of administration tics of the individual (Furr, 2018). For children
structure or consistency applies to all of the mea- and adolescents, such characteristics may be
sures in this text. For example, respondents similarly conceptualized under the term person-
should complete the measure in an environment ality traits or, typically for younger children,
free of distractions and should clearly compre- temperament. Research has clearly indicated that
hend the item content, response-format (e.g., individual differences in a variety of personality
true-false, frequency ratings, etc.), and time domains in youth are measurable (see Shiner &
frame (e.g., the last 6 months) referenced by the Caspi, 2003), relatively stable (e.g., Elkins,
test and respond accordingly. Standardized pro- Kassenboehmer, & Schurer, 2017), and predict
cedure emanates from experimental psychology, behavior (e.g., Lemery, Essex, & Smider, 2002).
where laboratory control is central to obtaining Rating scales, a particularly popular child
reliable and valid results (Kamphaus, 2001). assessment method, may fall into yet another cat-
Similarly, in the case of personality assessment, egory of test called diagnostic schedules
standardized administration procedure is neces- (Kamphaus, Morgan, Cox, & Powell, 1995).
sary to produce reliable and valid measurements Kamphaus et al. (1995) define a diagnostic
of behavior and to be able to compare someone’s schedule as a specialized psychometric method
scores to those from the normative sample that that provides a structured procedure for collect-
used the exact same procedures. ing and categorizing behavioral data that corre-
Psychological tests are designed to provide spond to diagnostic categories or systems. A
an indicator or behavioral sample by collecting diagnostic schedule, then, is not designed to
that information in a systematic way. The infor- assess a trait but rather to diagnose a syndrome.
mation from tests can be used to compare indi- How does one identify a diagnostic schedule?
viduals on the construct of interest or compare One clue is the source of the item pool. For exam-
scores on the measured construct for the same ple, the Children’s Depression Inventory-2 (CDI-
individual across more than one time point (Furr, 2; Kovacs, 2011) is a popular measure of
2018). This ability to make comparisons and childhood depression that has used the Diagnostic
then generalize findings to important areas of and Statistical Manual of Mental Disorders
functioning is central strength of psychological (DSM) as its item source. It was designed to
tests. Without these tests, psychological mea- assess the symptoms of depression in order to
surement would be impractical because of the assist with making the diagnosis of depression. It
time and expense required. Of course, a sample was not designed to assess a stable personality
may be unrepresentative of the population it is trait or temperament.
Types of Tests 19
However, adding further complexity to under- child who meets criteria for attention-deficit
standing how rating scales fit within the array of hyperactivity disorder (ADHD)). Diagnostic
tools available for clinical assessments is the fact schedules have evolved from behavioral assess-
that many widely used rating scales cannot be ment methods, as has the DSM, which now
considered diagnostic schedules. For example, emphasizes the tally of behaviors (symptoms),
although the Behavior Assessment System for as well as impairment/distress due to the symp-
Children-3 (BASC-3; Reynolds & Kamphaus, toms, for making a diagnostic decision.
2015) assesses clinically relevant domains for Personality tests and many rating scales, on the
youth (e.g., hyperactivity, aggression, depres- other hand, are rooted in the psychometric tradi-
sion, anxiety), elevated scores on those domains tion in which they are designed to assess traits
do not necessarily mean that the individual being across a continuum. Although such instruments
assessed meets criteria for a corresponding diag- may not lead directly to a diagnostic decision, as
nosis. Such rating systems routinely have items noted above, they can play other important roles
that do not directly map onto diagnostic criteria. by identifying traits that have implications for
Rather, the content of these scales may be indic- the course or prognosis of a disorder or even for
ative of aspects of the young person’s function- treatment.
ing that may lend themselves to recommendations While diagnostic schedules are practical for
for intervention, as well as help signal a diagno- making diagnostic decisions, such measures have
sis or impairment in a particular domain. limitations for studying the nature of individual
Furthermore, some rating scales may blend the differences or for contributing to other important
elements of personality tests and diagnostic aspects of the assessment process. These limita-
schedules, making sound clinical judgment tions are inherent in diagnostic schedules because
essential in drawing the soundest conclusions they often lack a clear theoretical basis or evi-
from the data collected. The primary purpose of dence of a priori defined trait(s) that can be sup-
the assessment (e.g., diagnostic clarification vs. ported with construct validity evidence.
identifying areas of behavioral, social, or emo- Therefore, the emergence of diagnostic schedules
tional concern) should guide the selection of as the instruments of choice for much of assess-
diagnostic schedules and/or behavior rating ment practice is evidence of the profound impact
scales. Further, we would argue that if given a of behavioral-based diagnostic systems on psy-
choice, clinicians should initially seek tools that chometric test development, as well as the (real
provide a broad screening of a variety of possi- or perceived) need to provide diagnoses as a
ble problems rather than narrowing in too result of all assessments due to managed health
quickly on a specific diagnosis or assessing spe- care.
cific diagnostic criteria. Appropriate conclusions that could be drawn
Mash and Hunsley (2005) have articulated the based on diagnostic schedules include statements
problems with considering a focus on specific like the following:
diagnoses as synonymous with psychological
assessment: • Vanessa suffers from major depression, single
Although formal diagnostic systems provide one episode, and severe.
alternative for framing the range of disorders and • Timothy exhibits nearly enough symptoms to
problems to be considered, there is no need to limit be diagnosed as having conduct disorder.
the range of problems to those detailed in a diag- • Crystal has attention problems that are worse
nostic system. (p. 368)
than those of 99% of the children her age.
Despite these concerns, diagnostic schedules
or checklists may still play a critical role in help- Alternatively, conclusions that could be
ing address a referral question and make treat- offered based on psychometric tests of personal-
ment recommendations that are diagnostically ity or behavioral, emotional, or social function-
relevant (e.g., classroom accommodations for a ing could include:
20 2 Measurement Issues
• Lauren shows evidence of poor adaptability to time efficiency. Furthermore, rating scales allow
new situations and changes in routine, which for the rapid and accurate identification of
puts her at risk for school adjustment domains of behavior that may require diagnosis
problems. or intervention (e.g., Wickerd & Hulac, 2017).
• Kory’s high score on the sensation-seeking In this volume, the term personality test will
scale warrants consideration as part of his occasionally be used generically to apply to per-
vocational counseling and educational sonality trait measures, diagnostic schedules,
planning. syndrome scales, and related measures, assuming
• Shanelle’s somatization tendencies reveal the that the reader is aware of the conceptual distinc-
need for counseling in order to reduce her fre- tions between subtypes of measures.
quency of emergency clinic visits.
• Andersen’s apparent signs of depression indi-
cate a need for further evaluation and Scores, Norms, and Distributions
intervention.
Types of Scores
A central difference between these interpre-
tive statements is that those made based on diag- In this section, some of the basic properties of
nostic schedules are dependent on diagnostic score types are reviewed, with particular empha-
nosologies. A variation of this premise is the third sis on the T-score standard score metric and its
statement exemplifying diagnostic-based conclu- variants. The properties of these scores will be
sions, which may result from a norm-referenced highlighted in order to encourage psychometri-
behavior rating scale that has a scale devoted to cally appropriate score interpretation.
inattention. Such norm-based information can
typically be gleaned from personality tests or
other rating scales as well, underscoring the flex- Raw Scores
ibility of these measures. Moreover, the interpre-
tive statements made based on psychometric tests The first score that the clinician encounters after
can be offered independently of diagnosis. These summing item scores is usually called a raw
conclusions are based on the measurement of score. Raw scores, on most tests, are simply the
traits or tendencies that may or may not represent sum of the item scores. The term raw is probably
diagnostic symptoms or signs, and yet these con- fitting for these scores in that they give little
clusions contribute substantially to the assess- information about a child’s performance as com-
ment process. pared to his or her peers. Raw scores are not par-
Widely used rating scales such as those to be ticularly helpful for norm-referenced
discussed later in this volume have several scales interpretation.
with the same name as a diagnostic category such Raw scores are often used for interpretation of
as depression or anxiety. At the same time, such non-norm-referenced measures such as diagnos-
measures are scaled similarly to traditional per- tic interviews. The Kiddie Schedule for Affective
sonality tests with standard scores based on Disorders and Schizophrenia (K-SADS;
norms. Kaufman et al., 2016) is one such measure. On
Although there is some research on this issue this measure and other interview schedules, one
(e.g., Ferdinand, 2008; Kerr, Lunkenheimer, & merely tallies the number of items (symptoms)
Olson, 2007), generally speaking, we do not endorsed and considers a diagnosis if this num-
know the extent to which these scales demon- ber meets or exceeds a cut score. This tally, in
strate the stability associated with traits or the effect, is nothing more than a raw score. This
diagnostic accuracy of the DSM system. Their score does not yield particular insight into a
popularity for clinical practice, however, contin- child’s behavior in terms of factors such as
ues to increase due to their cost-effectiveness and context or developmental typicality/atypicality.
Scores, Norms, and Distributions 21
It merely identifies the number of symptoms ity tests with good normative samples cannot do.
present, not how deviant this amount of The remaining scores discussed in this section
symptomatology is from the norm nor how
are norm-referenced scores that allow the clini-
impairing it is for the individual. Behavior obser- cian to make determinations of how deviant a rat-
vation schedules also often yield only raw scores. ing is compared to norms.
The Student Observation System (SOS) of the
Behavior Assessment System for Children-3
(BASC-3; Reynolds & Kamphaus, 2015) (see Standard Scores
Chap. 8) is one such scale. As with the K-SADS,
the only result yielded is a frequency count of The standard score is a type of derived score that
various classroom behaviors. Although the raw has traditionally been the most popular for psy-
scores yielded by measures that are not norm- chometric test interpretation. Standard scores
referenced have some utility for diagnosis, these convert raw scores to a distribution that reflects
scores are not useful for answering questions of the degree to which an individual has scored
deviancy (i.e., is the score rare for a child of this below or above the sample mean (Furr, 2018).
age and sex), as discussed next. The typical standard score scale used for person-
ality tests and behavior rating scales is the
T-score, which has a mean of 50 and standard
Norm-Referenced Scores deviation of 10. Another popular standard score
that is coming into more frequent use for person-
Personality test interpretation often focuses on ality test interpretation has the mean set at 100
norm-referenced interpretation, the comparison and the standard deviation at 15, similar to the IQ
of children’s scores to some standard or norm. metric (see Table 2.1). Because they have equal
This assessment framework is critical for the units along the scale, standard scores are useful
field of developmental psychopathology. More for statistical analyses and for making compari-
specifically, for the purposes of assessing psy- sons across tests. The equal units (or intervals)
chological constructs, scores are usually com- that are characteristic of standard scores are
pared to those of children the same age. This shown for various standard scores and percentile
result then provides information on how develop- ranks in Fig. 2.1. This table may also be useful
mentally typical, or age-typical, the child is in for converting a score from one scale to another.
terms of these constructs. Norm-referenced In the T-score metric, the distance between 20
achievement tests, by contrast, may compare and 30 is the same as that between 45 and 55.
children’s scores to those of others in the same Standard scores are particularly useful for test
grade, and college admission counselors may interpretation because they allow for compari-
compare an incoming student’s GPA to that of sons among various subscales, scales, and com-
freshmen who entered the year before. posites yielded by the same test, allowing the
Norm-referencing is of importance in person- clinician to compare traits. In other words, stan-
ality and behavioral assessment because it allows dard scores allow the clinician to answer ques-
the clinician to gauge deviance relative to normal tions such as “Is Janet more anxious than
development, which is often central to the refer- de-pressed relative to others her age?” thus facili-
ral question. Parents who refer a child for a tating profile analysis. Most modern personality
psycho logical evaluation often have norm- tests use T-scores, including the MMPI-A,
referencing in mind. They ask questions such as BASC-3, Achenbach Scales of Empirically
“Is her activity level normal for her age?” or Based Assessments, and Revised Children’s
“Everyone says he is just a boy, but fire setting Manifest Anxiety Scale-2.
isn’t normal, is it?” Norm-referencing allows the Wechsler popularized the scaled score metric
clinician to answer such questions objectively, for intelligence test subtest scores. Scaled scores
which those who do not have access to personal- have enjoyed some use in personality testing
22 2 Measurement Issues
Table 2.1 Standard score, T-score, scaled score, and percentile rank conversion table
Standard score T-score Scaled score Standard score T-score Scaled score
M = 100 M = 50 M = 10 M = 100 M = 50 M = 10
SD = 15 SD = 10 SD = 3 Percentile rank SD = 15 SD = 10 SD = 3 Percentile rank
160 90 99.99 128 69 97
159 89 99.99 127 68 97
158 89 99.99 126 67 96
157 88 99.99 125 67 15 95
156 87 99.99 124 66 95
155 87 99.99 123 65 94
154 86 99.99 122 65 92
153 85 99.98 121 64 92
152 85 99.97 120 63 14 91
151 84 99.96 119 63 89
150 83 99.95 118 62 88
149 83 99.94 117 61 87
148 82 99.93 116 61 86
147 81 99.91 115 60 13 84
146 81 19 99.89 114 59 83
145 80 99.87 113 59 81
144 79 99.84 112 58 79
143 79 99.80 111 57 77
142 78 99.75 110 57 12 75
141 77 99.70 109 56 73
140 77 18 99.64 108 55 71
139 76 99.57 108 55 69
138 75 99 107 55 67
137 75 99 106 54 65
136 74 99 105 53 11 65
135 73 17 99 104 53 62
134 73 99 103 52 57
133 72 99 102 51 55
132 71 98 101 51 52
131 71 98 100 50 10 50
130 70 16 98 99 49 48
129 69 97 98 49 45
97 48 43 68 29 2
96 47 40 67 28 1
95 47 9 38 66 27 1
94 46 35 65 27 3 1
93 45 33 64 26 1
93 45 31 63 25 1
92 45 29 63 25 1
91 44 27 62 25 1
90 43 8 25 61 24 0.49
89 43 23 60 23 2 0.36
88 42 21 59 23 0.30
87 41 19 58 22 0.25
86 41 17 57 21 0.20
85 40 7 16 56 21 0.16
(continued)
Scores, Norms, and Distributions 23
Table 2.1 (continued)
Standard score T-score Scaled score Standard score T-score Scaled score
M = 100 M = 50 M = 10 M = 100 M = 50 M = 10
SD = 15 SD = 10 SD = 3 Percentile rank SD = 15 SD = 10 SD = 3 Percentile rank
84 39 14 55 20 1 0.13
83 39 13 54 19 0.11
82 38 12 53 19 0.09
81 37 11 52 18 0.07
80 37 6 9 51 17 0.06
79 36 8 50 17 0.05
78 35 8 49 16 0.04
78 35 7 48 15 0.03
77 35 6 48 15 0.02
76 34 5 47 15 0.02
75 33 5 5 46 14 0.01
74 33 4 45 13 0.01
73 32 3 44 13 0.01
72 31 3 43 12 0.01
71 31 3 42 11 0.01
70 30 4 2 41 11 0.01
69 29 2 40 10 0.01
Test Score
z score I I I I I I I I I
-4 -3 -2 -1 0 +1 +2 +3 +4
T score I I I I I I I I I
10 20 30 40 50 60 70 80 90
Percentile I I I I I I I I I I I I I
Rank 1 5 10 20 30 40 50 60 70 80 90 95 99
Fig. 2.1 The relation of T-scores, Z-scores, standard deviations, and percentile ranks to the theoretical normal curve
24 2 Measurement Issues
(Kamphaus, 2001). They have a mean of 10 and a shows that percentile ranks have one major dis-
standard deviation of 3 (see Table 2.1). advantage in comparison with standard scores:
In a normal distribution (a frequently unten- Percentile ranks have unequal units along their
able assumption in personality and behavior scale. The distribution in Fig. 2.1 (and in
assessment, as is shown in a later section), a nor- Table 2.1) shows that the difference between
malized standard score divides up the same first and fifth percentile ranks is larger than the
proportions of the normal curve. As can be seen difference between the 40th and 50th percentile
in Table 2.1, a standard score of 85, T-score of ranks. In other words, percentile ranks in the
40, and scaled score of 7 all represent the same middle of the distribution tend to overempha-
amount of deviation from the average. These size percentile ranks at the tails of the distribu-
scores are all at the 16th percentile rank (see tion and tend to underemphasize differences in
Fig. 2.1). performance (Kamphaus, 2001).
However, because many scales on syndrome Here is an example of how confusing this
measures in particular are heavily skewed (the property of having unequal units can be. A clini-
most frequent scenario is that most individuals cian would typically describe a T-score of 55 as
are not experiencing psychopathology and only a average. When placed on the percentile rank
few are, resulting in positive skewness), some distribution, however, a T-score of 55 corre-
test developers opt for the use of linear T-scores. sponds to a percentile rank of 69 in a normal
Linear T’s maintain the skewed shape of the raw distribution of scores (see Table 2.1). The per-
score distribution, which means that the same centile rank of 69 sounds as though it is higher
T-score on different scales may divide up differ- than average. Examples such as this clearly
ent portions of the norming sample. Specifically, show the caveats needed when dealing with an
50% of the norming sample may score below a ordinal scale of measurement (unequal scale
linear T-score of 50 on the anxiety scale, whereas units) such as the percentile rank scale. It is
55% of the norming sample may score below a important to remember that the ordinal proper-
linear T-score of 50 on the aggression scale. ties of the scale are due to the fact that the per-
Essentially, then, the use of linear T-scores makes centile rank merely places a score in the
the relationship of percentile ranks to T-scores distribution. In most distributions, the majority
unique for each scale. of the scores are in the middle of the distribu-
tion, causing small differences between stan-
dard scores in the middle to produce large
Percentile Ranks differences in percentile ranks.
problem of a lack of percentile rank compara- stratification variables include age, gender, race,
bility across scales is described by Tellegen and geographic region, community size, and paren-
Ben-Porath (1992): tal socioeconomic status (SES; Kamphaus,
For example, the raw score distribution of Scale 8, 2001). These variables are used presumably
Schizophrenia (Sc), is more positively (i.e., right-) because they are related to score differences. Of
skewed than that of Scale 9, Hypomania (Ma). these widely used stratification variables, SES is
This means that a linear T-score of, say, 80 repre- known to produce the most substantial score dif-
sents different relative standings on these two
scales in the normative sample. For women in the ferences on intelligence measures (Kamphaus,
MMPI-2 normative sample, the percentile values 2001). The precedent, then, is set for the norm-
of a linear T-score of 80 are 98.6 for Scale 8 and ing of personality and behavioral assessment
99.8 for Scale 9; for men, the corresponding values tools, including along demographic variables.
are similar, 98.6 and 99.7. (p. 145)
However, the user must recognize the potential
For the UT scale score to have the properties role that demographic variables may play in
of percentile rank comparability across scales scores on measures in ways that are not related
and reflection of raw score distribution skewness, to the construct of interest. For example, SES
the UT-score is based on the average skewness may contribute to stigma in terms of how scores
value across all of the clinical scales (Tellegen & for individuals from a low SES background are
Ben-Porath, 1992). This approach meets the interpreted (Phelan, Lucas, Ridgeway, & Taylor,
objectives outlined by the developers, but it relies 2014).
on the assumption that the skewness of the In general, the development of norms for per-
MMPI-2 clinical scales is similar. There is, how- sonality and behavioral assessment measures for
ever, evidence that some MMPI-2 clinical scales youth has improved over the last few decades. We
(e.g., hypochondriasis and schizophrenia) are far will note the characteristics of normative samples
more skewed than others (see Tellegen & Ben- for some such widely used measures in subse-
Porath, 1992). It appears that the UT is a compro- quent chapters of this text. The savvy user of per-
mise metric that meets test development sonality and behavior tests should know the
objectives while, at the same time, not addressing characteristics of a test’s norming sample in order
completely the issue of different skewness across to make test use decisions and gauge the amount
scales. The use of the UT metric remains unique of confidence to place in the obtained scores.
to the MMPI-2 and the adolescent version Such a consideration is a critical first step in
(MMPI-A); thus, it remains unclear whether this designing a clinical assessment battery.
specific metric would have utility for other per- Many tests feature interpretation based on a
sonality tests. normative standard, the national norm sample.
However, a national normative standard may not
be offered for some personality tests, which rely
Norm Development on local norms, a subset of the national normative
sample. Local norms answer different questions
Sampling than do national norms. Hence, their potential
utility has to be evaluated prior to test selection
Intelligence test has routinely collected strati- and interpretation.
fied national samples of children to use as a nor-
mative base, whereas personality and behavioral
rating scales have included such sampling Local Norms
approaches only relatively recently. Stratification
is used to collect these samples in order to Local norms, or norms based on a specific popu-
match, to the extent possible, the characteristics lation in a specific setting or location, may some-
of the population at large on what are thought to times be more useful than national norms,
be important demographic variables. Common particularly in terms of their relevance for the
26 2 Measurement Issues
clinician’s work. For local norms to be meaning- norms are typically the only ones provided.
ful, however, the range of their usefulness must Specifically, intelligence, academic achieve-
be defined clearly. ment, and adaptive behavior scales produce
Regardless of the use of local or national mean score differences between gender groups,
norms, typical norm-referenced questions of but local norms by gender are rarely offered.
interest to psychologists are diagnostic ones. Why then are gender local norms commonly
That is, they help to determine if a child’s offered for personality tests? Tradition could be
“symptoms” are atypical for children of the same the most parsimonious explanation. When com-
age. Common questions might include: paring a child to his or her gender group, the
effects of gender differences in behavior are
• Is Lindsey’s attentional problems more severe removed. Another way of expressing this is to
than what you would expect for other children say that when gender norms are utilized, roughly
his age? the same proportion of boys as girls is identified
• Is Juan’s level of depression high than what is as having problems. Because, for example, boys
typically shown by other adolescents? tend to have more symptoms of inattention and
• Is Natalie more anxious than other children hyperactivity than girls (DSM-5 APA, 2013), the
her age? use of gender local norms would erase this dif-
ference in epidemiology. Gender norm-refer-
One of the goals of diagnostic practice is con- encing would identify approximately the same
sistency. National norms help this consistency percentage of girls and boys as hyperactive,
and allow clinicians to communicate clearly such that a boy would require more severe
with one another. If, for example, Dr. Ob Session symptomatology to be identified as elevated on
in Seattle finds that a patient’s conduct problems hyperactivity relative to other boys. Depression
are above a normative level, then Dr. Hugh Ego is another example of how gender norms may
in New Orleans will know what to expect from affect diagnostic rates. Much evidence suggests
this adolescent when he enters his office for fol- that girls express more depressive symptomatol-
low-up treatment. However, if Dr. Ob Session ogy than boys in adolescence (Hankin, Wetter,
used norms specific to Seattle and Dr. Hugh Ego & Cheely, 2008). The use of gender norms for a
uses norms specific to New Orleans, it is not depression scale would result in the same num-
clear if differences are due to differences in ber of adolescent boys as girls exceeding a par-
the norms used or changes in the patient’s ticular cut score, whereas general national
behavior. norms would retain the known greater preva-
lence among adolescent females.
Are gender local norms a problem? Not as
Gender-Based Norms long as clinicians are clear about the questions
they are asking. A gender norm question would
From this discussion, one may assume that it is be “Is Chloe hyperactive when compared to
always better to use national norms for interpret- other girls her age?” whereas a national norm
ing tests. However, there may be good reasons to question would be “Is Chloe hyperactive in com-
use certain local norms because one wants parison to other children her age?”. General
to limit comparisons to a certain group of national norms are often preferred when a diag-
individuals, not to all individuals. Specifically, nostic question is asked. An example of a diag-
personality and behavior scales often use gen- nostic question is “Does Frank have enough
der-referenced (local) norms. This practice is symptoms of depression to warrant a diagnosis?”
unusual in comparison with other domains of The DSM-5 diagnostic criteria do not have dif-
assessment where, although significant gender fering thresholds for boys and girls, so a gender
differences exist, national combined gender norm is inappropriate.
Norm Development 27
Skewed Distributions
conditions over a brief period of time (typically pline. However, whichever type of discipline the
2–4 weeks). The correlation between the first and parent chose to use, they used it at a fairly consis-
second administrations of the test is then com- tent rate across time.
puted, yielding a test-retest reliability coefficient Because internal consistency coefficients are
that is optimally very close to 1.0. Of course, the imperfect estimates of stability coefficients, both
importance of such reliability, or stability, types of coefficients should be recorded in the
depends on the construct being assessed. If clini- manual for a test (AERA, APA, NCME, 2014). It
cians seek to assess changes in specific, discrete is then up to the professional making use of the
behaviors as a result of an intervention, for test to determine if the reliability is suitable for
example, then test-retest reliability becomes less the purpose for which the tool is to be used.
of a concern. On the other hand, if a clinician
seeks to evaluate what are presumably relatively
stable indicators of behavioral functioning or Variables That Affect Reliability
personality, the test-retest reliability of the mea-
sure becomes paramount. Clinicians who use personality or behavior tests
should be especially cognizant of factors that can
affect reliability. Some factors that the clinician
Internal Consistency Coefficients should keep in mind when estimating the reliabil-
ity of a test for a particular child include the
Another type of reliability coefficient typically following:
reported in test manuals is an internal consis-
tency coefficient. This estimate differs from test- 1. Reliability is affected by homogeneous sam-
retest or stability coefficients in that it does not ples. Simply, the more homogenous the sam-
directly assess the stability of the measure of per- ple, the higher the reliability tends to be.
sonality or behavior over time. Internal consis- 2. Reliability can differ for different score levels.
tency coefficients assess what the name A test that is very reliable for emotionally dis-
implies—the average correlation among the turbed students is not necessarily as reliable
items in a test or scale. In other words, this index for nondisabled students without research evi-
of reliability assesses the homogeneity of the test dence to support its use (AERA, APA, NCME,
item pool. Internal consistency coefficients are 2014).
inexpensively produced, since they only require 3. Reliability can suffer when there is a long
one administration of the test. Typical formulae interval between assessments (Furr, 2018).
used for the computation of internal consistency 4. Reliability can be affected by rater or child
coefficients include split-half coefficients, characteristics such as age, reading level, and
Kuder-Richardson Formula 20, and coefficient fatigue. Reliability of personality measure-
(or Cronbach’s) alpha. ment, for example, may drop if the child does
On occasion, there are differences between not understand the test items.
internal consistency and test-retest coefficients 5. Analogously, error may be introduced if a
that can affect test interpretation. A test may, for poor translation of a test is used.
example, have a relatively poor internal consis-
tency coefficient and yet a strong test-retest coef-
ficient (Kamphaus, 2001). For example, Shelton, Reliable Specific Variance
Frick, and Wootton (1996) reported that a mea-
sure of corporal punishment had very low levels Subtest specificity is the amount of reliable spe-
of internal consistency but moderate levels of cific variance that can be attributed to a single
temporal stability. This appeared to be because subtest or scale. Kaufman (1994) popularized the
parents tended to have a preferred method of dis- use of subtest specificity in clinical assessment as
cipline. Thus, use of one type of discipline did a way of gauging the amount of confidence a cli-
not correlate highly with another type of disci- nician should have in conclusions that are based
Reliability 31
on a single subtest or scale. In effect, knowledge computed for some tests and may appear in
of subtest specificity makes clinicians more cau- the test’s manual or in texts that extensively
tious about drawing conclusions based on a sin- cover the test in question.
gle scale.
A reliability coefficient represents the amount
of reliable variance associated with a scale. An Standard Error of Measurement
example would be an anxiety scale taken from a
larger battery of 13 tests, all of which are part of The standard error of measurement (SEM) gives
a major personality test battery. The anxiety an indication of the amount of error associated
scale has a test-retest reliability coefficient of with test scores. In more technical terms, the
0.82. On the surface, this test appears reliable. If SEM is the standard deviation of the error distri-
this scale produces the child’s highest score, the bution of scores. The reliability coefficient of a
examiner may wish to say that the child has a test is one way of expressing the amount of error
problem with anxiety. The examiner can then associated with a test score in order to allow the
make this statement with confidence because the user to gauge the level of confidence that should
test is relatively reliable, right? Not necessarily. be placed in the obtained scores. An examiner
As Kaufman (1979) points out, the conclusion may report a personality test score for a child as
being drawn by the clinician is about some skill, being T = 63 with a test-retest reliability coeffi-
trait, or ability (in this case, anxiety) that is spe- cient of 0.95. This practice, however, is unortho-
cific or measured only by this one scale. The dox and clumsy. The typical practice is to report
reliability coefficient, on the other hand, reflects a test score along with the test’s standard error of
not just reliable specific variance but also reli- measurement, as is frequently done for opinion
able shared variances. Subtest specificity is typi- polls conducted by the popular media (e.g., the
cally computed in the following way (Kamphaus, error rate or margin of error of this poll is…).
2001): The standard error of measurement is simply
another way of reflecting the amount of error
1. Compute the multiple correlation (R) between associated with a test score.
the scale in question and all other scales in the In classical test theory, if a child were admin-
battery, and square it (R2). This computation istered a personality test 100 times under identi-
yields the amount of reliable shared variance cal conditions, he or she would not obtain the
between the scale in question, in this case same score on all 100 administrations. Rather,
anxiety, and the other scales in the battery. the child would obtain a distribution of scores
2. Subtract the squared multiple correlation
that approximates a normal curve. This error
coefficient from the reliability coefficient or distribution would have a mean. The mean of
rtt. If R2 = 0.30, 0.82–0.30 = 0.52. This for- this theoretical distribution of scores is the
mula yields the reliable specific variance. child’s true score. A true score is a theoretical
3. Compare the amount of reliable specific vari- construct that can only be estimated. This error
ance (0.52) to the amount of error variance distribution, like other distributions, not only
(1 − 0.82 = 0.18). If the reliable specific vari- has a mean, but it can also be divided into stan-
ance exceeds the error variance by 0.20 or dard deviations. In an error distribution, how-
more, then the scale is considered to have ever, instead of being called a standard deviation,
adequate specificity for interpretive purposes. it is called the SEM. As one would predict, then
By convention, if the reliable specific vari- in this error distribution of scores ±1 SEM
ance exceeds the error variance by 0.19 or divides up the same portion of the normal curve
less, then the test lacks specificity, and it (68%) as does a standard deviation, and ±2
should be cautiously interpreted. If the reli- SEMs divides up the same proportion of the
able specific variance does not exceed the error distribution (95%) as ±2 standard devia-
error variance, then interpretation of the scale tions does for a normal distribution of obtained
is ill-advised. Subtest specificity values are scores.
32 2 Measurement Issues
the Children’s Depression Inventory-2 (CDI-2), they utilize. By doing so, we think that clinicians
where items were based on accepted diagnostic will be better able to judge the implications of
nosologies such as the DSM. Even in such cases, test results for interpretation. Construct irrele-
however, personality test developers usually do vance, in particular, is of great concern as we
not go to the lengths of other test developers to note in later sections.
document adequate sampling of the content (or
psychopathology) domain. Few personality tests
or behavioral rating scales, for example, use Criterion-Related Validity
panels of experts to develop item content.
Problems with regard to content validity may Criterion-related validity assesses the degree to
be identified as cases of construct underrepre- which tests relate to other tests in a theoretically
sentation or construct irrelevance. A depression appropriate manner. There are two kinds of
scale may suffer construct underrepresentation, criterion-
related validity: concurrent and
for example, if it lacks both cognitive (e.g., predictive.
excessive self-deprecation) and vegetative symp-
toms of depression (e.g., hypersomnia). In this Concurrent Validity This type of validity stipu-
scenario, it may be said that there are not enough lates that a test should show substantial correla-
items on the scale that are known to be “indica- tions with other measures to which it is
tors,” or symptoms of depression, resulting in theoretically related. One of the important crite-
questionable content, ergo inadequate construct, ria for the evaluation of personality or behavior
validity. measures since their inception has been that they
The reader will note in later chapters that con- show a substantial correlation with other indica-
struct irrelevant items are a more serious prob- tors of psychopathology, such as well-validated
lem in behavioral assessment. This problem is tests or clinicians’ratings or diagnoses. The typi-
likely to occur when only empirical methods cal concurrent validity investigation involves
(i.e., factor analysis) are used to construct scales administering a new behavior rating scale and an
and select items for scales. Examples of con- existing well-validated measure of psychopathol-
struct irrelevance are listed in Chap. 17 as they ogy to a group of children. If a correlation of 0.20
relate to the assessment of ADHD. In compari- is obtained, then the concurrent validity of the
son with some other syndromes, ADHD is a new test would be in question. A 0.75 correlation,
well-studied condition with a widely agreed- on the other hand, would be supportive of the
upon set of symptoms (e.g., motor hyperactivity validity of the new test.
and inattention). What if, however, an item such
as “My child is adopted” was placed on an inat- Predictive Validity Predictive validity refers to
tention subscale of a parent rating scale? Such an the ability of a test to predict (as shown by its
item would likely be identified as a source of correlation) some later criterion. Research inves-
construct irrelevant variance for this scale, a tigating predictive validity is conducted very
source that would lead to less valid assessment similarly to a concurrent validity study, with one
of attention problems for the child undergoing important exception. The critical difference is
evaluation. that in a predictive validity study, the new per-
In our view, construct irrelevance and con- sonality test is first administered to a group of
struct underrepresentation are likely to become children and then is correlated with other mea-
problems at the item/symptom/behavior selec- sures sometime in the future, with the time frame
tion stage of test development. We, therefore, of relevance determined by the construct itself
caution test users to carefully review the process (e.g., later educational outcomes could be
of item selection and scale construction as well defined as the end of a unit, term, year, or
as the resulting scale content for each test that post-graduation).
34 2 Measurement Issues
reduce the correlations between all scales (or between a scale and a factor means the same
items) in a test to a smaller set of common fac- thing as a high positive correlation between two
tors. This smaller set of common factors will scales in that they tend to covary to a great
presumably be more interpretable than all of the extent. One can see from Table 2.2 that the
scales (or items) on a personality test considered Hysteria scale is highly correlated with Factor 1,
as individual entities. for example, and that Mania is not highly corre-
Factor analysis begins with the computation lated with Factor 1, but it is highly correlated
of an intercorrelation matrix showing the correla- with Factor 2.
tions among all of the items or scales in a test Once the factor matrix, such as that shown in
battery (most studies of behavior or personality Table 2.2, is obtained, the researcher must label
tests use item intercorrelations as input). These the obtained factors. This labeling is not based
intercorrelations then serve as the input to a on statistical procedures but on the theoretical
factor-analytic program that is part of a popular knowledge and perspective of the individual
statistical analysis package. researcher. For the MMPI-A, there is general
The output from a factor analysis that is fre- agreement as to the names of the factors. The
quently reported first in test validation research is first factor is typically referred to as general mal-
a factor matrix showing the factor loading of each adjustment and the second as overcontrol. The
subtest on each factor. A factor loading is, in third and fourth factors are named after the
most cases of exploratory factor analysis, the cor- scales with the highest loadings on each social
relation between a scale and a larger factor.1 introversion and masculinity-femininity
Factor loadings range from −1 to +1 just as do (Butcher et al., 1992).
correlation coefficients. Test developers often eliminate scales or items
For illustration purposes, selected factor based on factor analyses. They also commonly
loadings for the MMPI-A clinical scales in the design their composite scores based on factor-
standardization sample (Butcher et al., 1992) are analytic results. This process was not followed in
shown in Table 2.2. A high positive correlation the development of the MMPI-A, as this test was
developed long before the ready availability of
When orthogonal (independent or uncorrelated) rotation
1
factor-analytic procedures. Although the
techniques are used (and these techniques are very fre- MMPI-A appears to be a four-factor test, it pro-
quently used in test validation research), the factor load-
ing represents the correlation between the subtest and a
duces ten clinical scale T-scores, and no compos-
factor. This is not the case when oblique or correlated ite scores corresponding to the four obtained
methods of factor analysis are used (Furr, 2018). factors are offered. More recently developed
36 2 Measurement Issues
tests, such as the BASC-3, made heavy use of test, the confirmatory factor analysis will, in fact,
factor-analytic methods in the development of yield four factors. Statistics assessing the fit of
scale and composite scores (see Chap. 7). these four factors to the data may, however, indi-
Generally, then, consumers of factor-analytic cate a lack of congruence (i.e., correlation)
research seek comparability between the factors between the hypothesized four factors and the
and composite scores offered for interpretation. obtained four factors.
If there is, for example, a one-to-one relation- Thorough confirmatory factor-analytic studies
ship between the number of factors found and use a variety of statistics to assess the fit of the
the number of composite scores produced, then hypothesized factor structure to the data. These
the validity of the composite scores is likely statistics may include a chi-square statistic,
enhanced. goodness- of-fit index, adjusted goodness-of-fit
index (GFI), comparative fit index (CFI), root
mean square error of approximation (RMSEA),
Confirmatory Factor Analysis the Tucker-Lewis Index, Akaike Information
Criterion (AIC), and standardized root mean
The procedures discussed thus far are generally square residual (SRMR). Several statistics are
referred to as exploratory factor-analytic proce- desirable for checking the fit of a confirmatory
dures. Another factor-analytic technique is called factor analysis because all of these statistics have
confirmatory factor analysis (Kamphaus, 2001). strengths and weaknesses. The chi-square statis-
These two factor-analytic procedures differ in tic, for example, is highly influenced by sample
some very important ways. In exploratory factor size (Furr, 2018).
analysis, the number of factors to be yielded is Of course, factor-analytic results are heavily
typically dictated by the characteristics of the influenced by the quality of the data being input.
intercorrelation matrix. That is, the number of If, for example, a factor analysis is being per-
factors selected is based on the amount of vari- formed on scales with reliability indexes of 0.80
ance that each factor explains in the correlation and above, one is more likely to obtain factor-
matrix. If a factor, for example, explains 70% of analytic results that will be both meaningful and
the variance in the correlation matrix, then it is replicable by others.
typically included as a viable factor in further As alluded to earlier, however, factor analysis
aspects of the factor analysis. If, on the other is occasionally used as a method of item selec-
hand, the factor only accounts for 2% of the vari- tion and scale refinement with items as input.
ance in a factor matrix, then it may not be The reliability of these individual items is more
included as a viable factor. likely to be either poor or unknown. Thus, if an
In direct contrast, in confirmatory factor anal- unreliable indicator of a construct is used (i.e., an
ysis, the number of factors is not dictated by data item), some aspect of the validity of the resulting
but rather by the theory underlying the test under scale may be questionable. It is desirable then to
investigation. In confirmatory factor analysis, know the reliability and validity of indicators
the number of factors, as well as the scales/items included in a factor analysis and to have the
that load on each factor, is selected a priori results of such analyses replicated by individual
according to a hypothesized model (Furr, 2018). researchers.
In addition, the general fit between the factor
structure dictated a priori, and the obtained data
are assessed. If there is a great deal of correspon- Cluster Analysis
dence between the hypothesized structure and
the obtained factor structure, then the validity of Similarly to factor analysis, cluster analysis
the personality test is supported, and the theory attempts to reduce the complexity of a data set. In
is confirmed (hence the term confirmatory). If, factor analysis, it is typical to try to reduce a large
for example, a researcher hypothesized the exis- number of variables (e.g., items) to a smaller set.
tence of four factors in a particular personality In cluster analysis, researchers are most often
Validity 37
interested in grouping individuals (as opposed to depression; anxiety alone; depression alone;
variables) into groups of people who share com- absence of anxiety and depression).
mon characteristics. Cluster analysis and other person-centered
Several steps are common to cluster-analytic techniques, such as LCA, may also be a useful
techniques, including the following: method for classifying children without problems
and those with subsyndromal impairment that
1. Collect a sample of individuals who have been requires prevention or treatment (e.g., Kamphaus
administered one test yielding multiple scores et al., 1999). Despite this potential usefulness,
or a battery of tests. one should be aware that the process of determin-
2. For each variable (e.g., depression scores),
ing the clinical relevance of cluster analysis
compute the distance between each pair of results involves some judgment and possible
children. arbitrariness. The consumer of such work is
3. These distances between each individual on encouraged to bear in mind the limitations inher-
each variable are then used to produce a prox- ent in considering youth as falling into distinct
imity matrix. This matrix serves as the input subtypes based on indicators of psychological
for the cluster analysis in the same way that functioning.
correlation or covariance matrices are used as
input in factor analysis.
4. Apply a cluster-analytic method that sorts
Sensitivity and Specificity
individuals based on the distances between
individuals that were plotted in the proximity Making a diagnosis is one of the primary rea-
matrix. In simple terms, clustering methods in sons for conducting an evaluation. A test that is
this step match individuals with the smallest to be used for such a purpose should possess evi-
distance between individuals on a particular dence of sensitivity or the ability to identify true
variable. positives (i.e., the percentage of children who
5. This sorting process continues until groups of actually have the disorder). A prototypical study
individuals are formed that are as homoge- might involve administering an electronic mea-
neous as possible (i.e., have profiles of scores sure of inattention to a group of children with
of similar level and shape). ADHD and a group without any psychiatric
6. Just as in factor analysis, the researcher has to diagnoses (“normals”). In this type of investiga-
decide next on the number of clusters that is tion, performance-based measures of inattention
the most clinically meaningful. Statistical often demonstrate good sensitivity by correctly
indexes are provided as an aid to the researcher identifying the vast majority of cases of ADHD,
in this step. a finding that then triggers investigation of spec-
ificity. Specificity refers to the relative percent-
An increasingly popular approach to accom- age of true negatives or the correct identification
plishing the goals of cluster analysis is latent of individuals who do not have the disorder as
class analysis (LCA) or latent profile analysis. not having the disorder. This same measure of
This approach allows for classification of indi- inattention may also identify only 50% of the
viduals based on their scores on a number of rel- nondiagnosed sample as “normal.” Therefore, it
evant constructs but differs from cluster analysis may have demonstrated good sensitivity but
in that it also provides a probability that a given inadequate specificity (i.e., a high rate of false
case belongs in a given class (Porcu & Giambona, positives). Performance-based measures of inat-
2017). Such an approach can be useful in tention often produce results of this nature. In an
research on developmental psychopathology, exhaustive review of the literature on such mea-
particularly insofar as it can provide information sures, Riccio and Reynolds (2003) concluded
about risk and protective factors associated with that while sensitivity is typically good, evidence
different classes (e.g., comorbid anxiety + of specificity is often poor.
38 2 Measurement Issues
To be useful, tests should demonstrate the child to understand the test items. Although con-
ability to differentiate among diagnostic catego- cern is often expressed about the ability of chil-
ries or have problem specificity (both clinically dren to read test items, parents also have
and subclinically), not just allow distinctions difficulty due to limited educational attainment
between a diagnostic group and normality (Mash or cultural or linguistic differences. The manuals
& Hunsley, 2005). Unfortunately, few test man- of many commonly used assessment instruments
uals provide evidence of this nature, and many now report the reading level of the test content,
journal articles test only the relatively easy dis- often using indices such as the Flesch-Kincaid
tinction between persons with some diagnosis grade level formula. In circumstances in which
and others with no diagnosis at all. The clini- this information is not available and a clinician
cian, however, routinely has the more difficult questions the readability of test content for a
task of differentiating among diagnostic catego- specific client, he/she may opt not to use the
ries, many of which share common features measure, may have the client read portions of the
(e.g., problems paying attention). Again, elec- instructions and items aloud to gauge readabil-
tronic measures of inattention have not shown ity, or read the entirety of the measure aloud to
good evidence of diagnostic group differentia- the client.
tion. In fact, Riccio and Reynolds (2003) con- Related to this point is the problem of transla-
cluded that when children with a number of tional equivalence or the degree to which a trans-
problems are included, the proportion of chil- lation of a test is equivalent to its original
dren correctly classified drops significantly. language form (AERA, APA, NCME, 2014).
Additional work has focused on improving Evidence of translational equivalence should be
problem specificity of assessment tools or the offered to reassure the test user that a threat to
ability to distinguish particular problems from validity is not present.
each other (see Mash & Hunsley, 2005).
Clinicians routinely are faced with this task
which is also often considered “differential Response Sets
diagnosis.” The call for a larger more sound evi-
dence base also raises awareness of positive pre- A response set is a tendency to answer questions
dictive power (i.e., the ability of an item/test to in a biased fashion, thus masking the true feel-
correctly identify a child with a particular prob- ings of the informant. These response sets are
lem) and negative predictive power (i.e., the often mentioned and, in some personality tests,
ability of an item/test to correctly identify a addressed in construction and interpretation.
child without a problem; Pelham, Fabiano, & The social desirability response set is “the
Massetti, 2005). More research on these issues tendency for a person to respond in a way that
can only serve to assist in clinical decision-mak- seems socially appealing” (Furr, 2018, p. 321).
ing, but increased evidence will still not replace Some personality tests include items and scales
clinical judgment in integrating information to assess the potential effects of such a response
from a variety of sources that have varying set. “I like everyone that I meet” might be an
degrees of validity. item on such scales. The acquiescence response
set is the tendency to answer true or yes to a
majority of the items (Furr, 2018). A third
Threats to Validity response set to be aware of is called deviation,
and it comes into play when an informant tends
Readability to give unusual or uncommon responses to items
(Anastasi & Urbina, 1998). Readers are referred
An obvious but easily overlooked, threat to to Furr (2018) for further discussion of these and
validity is lack of ability of the parent, teacher, or other response sets or biases (e.g., malingering,
Conclusions 39
guessing, random/careless responding) that may ing personality tests. By the time that an
impact results and data interpretation. assessment instrument is well-known and widely
used in clinical settings, it usually has demon-
strated adequate reliability and construct valid-
Guarding Against Validity Threats ity. However, as Mash and Hunsley (2005)
describe, the question of utility remains as to
Many personality and behavior scales, such as whether the instrument provides:
those described in later chapters, typically have psychologists with the kinds of information that
validity scales that are geared toward flagging can be used in ways that will make a meaningful
potential response sets, such as inconsistent difference in relation to diagnostic accuracy, case
responding, acquiescence, or malingering (“fak- formulation considerations, and treatment out-
comes. (p. 365)
ing bad”). For example, some tests include fake
bad scales, which assess the tendency to exag- This concept can also be applied to the inclusion
gerate problems. Some tests also often have con- of a particular informant in the assessment pro-
sistency indexes. One such index allows the cess as well as to the relative cost-effectiveness
examiner to determine if the informant is of one tool versus another comparable tool.
answering questions in a predictable pattern. For In short, a rating scale, for example, may be a
example, a consistency index might be formed valid indicator of depression, but its utility indi-
by identifying pairs of test items that correlate cates how valuable that particular rating scale is
highly. If an informant responds inconsistently for an assessment of depression relative to other
to such highly correlated items, then his or her measures and the cost (monetary and time)
veracity may be suspect. involved in administering it. Utility analysis of a
Examiners often also conduct informal valid- measurement tool involves estimating the mon-
ity checks. One quick check is to determine etary value of the measure itself along with the
whether the informant responded to enough items outcomes tied to the outcomes of decisions made
to make the test result valid. Another elementary using the measure (Furr, 2018).
validity check involves scanning the form for pat- Validity, including incremental validity (i.e.,
terned responding. A response form that rou- the improved assessment decision as a result of
tinely alternates between true and false responses adding a measure), is a necessary condition for
may reflect a patterning of responses on the part utility, and establishing such validity evidence
of the rater. for an assessment tool, and especially an entire
One way to limit the influence of response sets assessment battery, is arduous. Nevertheless,
is to ensure that informants are clear about the various forms of validity evidence are likewise
clinician’s expectations. Some clients may also not sufficient for demonstrating utility.
need to take the personality test under more con- Ultimately, in addition to cost-effectiveness, the
trolled circumstances. If an examiner has reason clinician must take into account the assess-
to believe, for example, that a child is opposi- ment’s role in translating to effective interven-
tional, then the self-report personality measure tion and subsequent positive change for a child
may best be completed in the presence of the (Mash & Hunsley, 2005).
examiner.
Conclusions
Utility
Knowledge of psychometric principles is crucial
Clinical utility is a concept that is not discussed for the proper interpretation of personality tests.
as much as reliability and validity in the discus- As psychometrics become more complex, clini-
sion of psychological assessment tools, includ- cians have to become increasingly sophisticated
40 2 Measurement Issues
24. Sensitivity refers to the ability of scores from 25. Clinical utility is an additional consideration
a test to correctly identify true positives (e.g., for the selection of assessment tools and
those with a diagnosis) and specificity to the informants. Utility concerns how well a par-
ability of scores from a test to correctly iden- ticular tool provides information that is
tify true negatives (e.g., those without a unique and cost-effective relative to other
diagnosis). tools (or informants).
Classification and Developmental
Psychopathology 3
but we have selected two that we feel are most sification is a method of determining when psy-
critical to the clinical assessment of children and chological functioning is abnormal, deviant, and/
adolescents. First, understanding the theories that or in need of treatment. On a second level, clas-
guide different models of classification is neces- sification is a method of distinguishing among
sary because the framework that one uses to define different dimensions or types of psychological
and classify psychological functioning determines functioning. When classification is viewed this
how one designs and interprets an assessment bat- way, it becomes clear that clinical assessment is
tery. Second, the clinical assessment of children at least partly a process of classification. It
must be conducted within the context of a broad involves (1) determining whether some areas of
knowledge of developmental psychopathology. psychological functioning in a child and adoles-
Developmental psychopathology refers to an inte- cent are pathological and in need of treatment
gration of two scientific disciplines: child develop- and (2) determining what types of pathology may
ment and child psychopathology. The integration be present. Stated another way by Achenbach
rests on the basic assumption that the most appro- (1982), “assessment and classification are two
priate way to view the emotional and behavioral facets of what should be a single process: assess-
functioning of children, both normal and problem- ment aims to identify the distinguishing features
atic, is within a comprehensive framework that of individual cases; taxonomy (classification) is
includes the influence of developmental processes the grouping of cases according to their distin-
(Rutter & Garmezy, 1983). A noted developmental guishing features” (p. 1). Therefore, understand-
psychopathologist, Judy Garber, summarized her ing the issues involved in classification is
view of this field as being “concerned with both essential to clinical assessment.
the normal processes of change and adaptation and The first issue is acknowledging that any clas-
the abnormal reactions to stress or adversity, as sification system of psychological functioning
well as the relationship between the two” (Garber, will be imperfect. Psychological phenomena do
1984, p. 30). Developmental psychopathology, not fall neatly into categories of normal and
then, is a framework for understanding children’s abnormal or into clear non-overlapping types of
emotions and behaviors that has many implica- dysfunction. This seems to be especially true
tions for the assessment process. with children; there is often no clear demarcation
It is beyond the scope of this book to provide of when a dimension of behavior should be con-
an intensive and exhaustive discussion of theories sidered normal and when it should be considered
of classification or the many important findings pathological. Further, there is often a high degree
in the field of developmental psychopathology. of overlap between the various forms of psycho-
Instead, this chapter will illustrate how critical pathology in children. Finally, any classification
these two areas of knowledge are to the assess- system is only as good as the research that is used
ment process and will provide at least a basic to create it. As research advances, so too should
framework for applying knowledge in these areas the classification system. Therefore, any system
to the assessment of children and adolescents. of classification is bound to be imperfect. Due to
The discussion that follows highlights some of this imperfection, many experts have argued
the issues in both areas that we feel have the most against the need for any formal classification sys-
relevance to the assessment process. tem. Instead, they argue that psychological func-
tioning should be assessed and described
idiosyncratically for each individual person. That
Classification is, each person is a unique individual whose psy-
chological functioning should simply be
Classification refers to the process of placing described in ways that maintain this uniqueness
psychological phenomena into distinct categories without being compared with other individuals or
according to some specified set of rules. There fitted into artificial categories. This argument has
are two levels of classification. One level of clas- quite some intuitive appeal given the complexity
Classification 45
of human nature. However, there are several using the term Major Depression would be mis-
compelling arguments for why we need good leading and actually hurt communication if the
classification systems, in spite of the fact that term was used without systematically ensuring
even the best system will be imperfect. that the DSM-5 criteria for Major Depression
were actually being displayed by the person, such
as making the diagnosis based on responses from
Communication a projective technique.
classification, the clinical assessor must be espe- difference between children with T-scores of 69
cially cognizant of these issues. Many of these (not classified) and children with T-scores of 70
dangers can be limited if classification systems (classified). Stating that this is an illusion does
are used appropriately. Therefore, in the discus- not imply that all psychological traits are on a
sion that follows, we have tried to not only out- continuum with normality, because some are not.
line the dangers of classification but also to For example, in Jerome Kagan’s work with
present practices that minimize or eliminate behaviorally inhibited children, there seem to be
potentially harmful effects. a number of qualitative differences between chil-
Because psychological phenomena, and the dren with the behaviorally inhibited temperament
persons they represent, do not fall neatly into and those without this temperamental style
categories, one loses information by attempting (Kagan & Snidman, 1991). However, if one was
to fit people into arbitrary categories. People using a measure of behavioral inhibition with
within the same category (e.g., Major some cutoff for classifying inhibition (e.g., a
Depression) share certain characteristics (e.g., T-score of 65), there may be some children close
depressed mood, loss of interest in activities, to this threshold (e.g., T-scores of 60–64) who
disturbances in sleep, impaired concentration), were not classified due to imperfections in the
but there will also be many differences among measurement technique (Ghiselli, Campbell, &
persons within a category (e.g., the number of Zedeck, 1981). Therefore, whether the illusion of
depressive episodes, whether the depression a clear break is due to a normally distributed trait
started after the death of a relative). The shared or due to measurement error, it is still an
characteristics should provide some important illusion.
information about the persons in the category A third danger of classification is that of stig-
(e.g., prognosis, response to treatment), or the matization associated with psychological labels,
classification is useless. However, given the loss often the end result of classification. How strong
of information inherent in any classification the effect of labeling is on psychological func-
grouping, classification should not be consid- tioning is not clear from research. There is no
ered the only information necessary for deter- evidence to suggest that the act of labeling cre-
mining the child’s treatment needs. Instead, any ates significant pathology through a self-fulfill-
classification, whether it is a diagnosis or an ing prophesy. However, it is also clear that labels
elevation on a behavior rating scale, should be can affect how others, either lay persons (e.g.,
one part of a larger description of the case (i.e., Snyder, Tanke, & Berscheid, 1977) or clinicians
the case conceptualization). This approach (Rockett, Murrie, & Boccacini, 2007), interact
allows one to take advantage of the positive with children who have been diagnosed with
aspects of formal classification (e.g., enhanced certain mental health problems. Given this
communication), yet it acknowledges the limits potential danger, classificatory terms (labels)
of such systems and integrates classification should be used cautiously and only when there
into a broader understanding of the case. is a clear purpose for doing so (e.g., when it
Throughout this book, we provide case studies influences treatment considerations). Also,
that illustrate this approach. In each case, diag- when such terms are used, great efforts should
noses or other methods of classification are inte- be made to clearly define the meaning of the
grated into a more complete clinical description term to limit the potential for misinterpretations.
of the child being assessed. And, finally, terms should be worded to show
A second danger of classification systems is that the person is not the disorder but that the
that they foster the illusion of a clear break disorder is something a person may have at
between normal and pathological psychological some given point in time. For example, one
functioning. For example, if you are classifying should use the phrase “a child with conduct dis-
based on elevations of a rating scale (over a order” rather than stating “a conduct disordered
T-score of 70), it gives the illusion of a dramatic child.”
Classification 47
behavior espoused by the system, such as psy- Interpretations are typically made by comparing
chodynamic theories or biological theories. In the an individual case to a representative normative
first case, the diagnosis is based on the internal sample and choosing some level of functioning
conflict surrounding key personality dynamics, as being so rare in the average population (e.g.,
whereas in the latter case, the diagnosis is based above the 98th percentile) that it should be con-
on abnormalities in physical processes. In both sidered deviant. Classifications are thus based on
cases, the disorder is within the individual and how a child falls on a certain dimension of func-
can largely only be inferred from the outward tioning (e.g., anxiety/withdrawal) relative to
symptoms displayed by the person. Second, some comparison group (e.g., compared to other
because of the emphasis on a pathological core children of the same age group).
(e.g., the disease entity), medical model systems In Table 3.1, we provide an example of a mul-
typically make sharp distinctions between disor- tivariate approach to classification of childhood
dered and non-disordered individuals. There is emotional and behavioral functioning reported
typically an underlying assumption that there are by Lahey and colleagues (2004). This system was
qualitative differences between individuals with based on a series of exploratory and confirmatory
and without a disorder. factor analyses of caretaker ratings of 1358 chil-
dren and adolescents ages 4–17. It shows six
Multivariate Approaches The second major dimensions of behavior that can be subsumed
approach to classification that has been extremely under two broad dimensions of Externalizing and
influential in the clinical assessment of children Internalizing problems.
has been labeled the multivariate statistical
(Quay, 1986) or the empirically based approach
(Achenbach, Rescorla, & Ivanova, 2012). In this he Current Status of Different
T
approach, multivariate statistical techniques are Approaches to Classification
used to isolate interrelated patterns or “syn-
dromes” of behavior. Therefore, unlike the medi- These two basic approaches to classification are
cal model disorders that are defined by theory important for clinical assessment because the
and clinical observations, behavioral syndromes design of assessment instruments often is consis-
are defined by the statistical relationship between tent with one of these basic approaches. More
behaviors or their patterns of covariation. In this importantly, the interpretation of assessment
approach, behaviors form a syndrome if they are instruments is basically a process of classifica-
highly correlated with each other, and there is no tion. Therefore, it is often guided by these models
necessary assumption of a pathological core to or some variation of them. The clinical assessor
underlie the symptoms, as is the case in medical should be aware of the issues involved in classifi-
models of classification. Instead, the disorder is cation generally, and the advantages and disad-
the pattern of behaviors and not what may be vantages of these two models of classification
underlying this pattern. specifically, to aid in the interpretation of assess-
ment measures.
In addition to being based on statistical covari- Research has indicated that both the medical
ation, the multivariate approach is also different and the multivariate models have flaws that make
from medical models of classification because it the exclusive use of either approach problematic.
emphasizes quantitative distinctions rather than For example, the reliance of medical model
qualitative distinctions. Once behavioral syn- approaches on theory has led to many “disorders”
dromes are isolated through statistical analyses, being created with little support from research
then a child’s level of functioning along the vari- (Achenbach et al., 2012). Also, the medical
ous dimensions of behavior is determined. model approach, with its emphasis on qualitative
Behavioral syndromes are conceptualized along distinctions, masks a continuum with normality
a continuum from normal to deviant. that seems most appropriate for understanding
Classification 49
many dimensions of functioning. Finally, despite by the American Psychiatric Association (APA,
the promise that the focus on pathogenic pro- 2013). The DSM-5 is the most recent edition of
cesses that underlie behavioral symptoms could the manual that has been highly influential in
lead to more targeted and effective treatments, making mental health diagnoses since the 1950s,
most successful treatments focus largely on the especially in the United States. Its history and its
behavioral and outward manifestations of the dis- current structure will be described in more detail
order and not on the underlying causes. In fact, in below. However, whereas the most recent edi-
most individual cases, it is hard to be confident in tions of the manual suggest that disorders should
what is the actual or most important cause of the involve a dysfunction within the individual (i.e., a
child’s emotional and behavioral problems. medical model), disorders are largely defined by
In contrast, the multivariate approach with its the number of symptoms present. Further, the
reliance on statistical covariation in the absence types of symptoms used to define the disorder are
of clear theory has resulted in syndromes that are often based on behavioral covariation, and the
often hard to generalize across samples and cutoff for making the diagnosis is typically based
across different methods used to assess children’s on empirical findings as to when the level of
emotional and behavioral problems. Also, symptoms leads to impairment in the individual
whereas some psychological phenomena in chil- (Lahey et al., 2004). Thus, in practice, the criteria
dren and adolescents are best conceptualized on a used to make diagnoses from the DSM-5 are
continuum with normality, there are others that often more closely aligned to the multivariate
may fit with more qualitative distinctions (e.g., model.
Kagan & Snidman, 1991; Lahey et al., 1990) and
are not captured well by the multivariate The National Institute of Mental Health
approach. Most importantly, however, the focus Research Domain Criteria (RDOC) The
on the behavioral syndromes makes it hard to RDoC project is a recent attempt to develop a
determine whether the same symptoms may reliable and valid method for classifying the psy-
result from different causes or whether the same chological and biological mechanisms that might
underlying cause could lead to a multitude of dif- lead to the symptoms that define disorders in the
ferent behavioral expressions. DSM-5. Thus, from its outset, it was not designed
This tension between the benefits of medical to “compete” with the DSM-5 (Cuthbert & Insel,
and multivariate approaches to classification can 2010). Instead, it was designed to provide a med-
be seen in the recent initiation of the Research ical model complement to the DSM-5 that moved
Domain Criteria project initiated by the National away from clinical descriptions of outward
Institute of Mental Health (RDoC; Sanislow behaviors and symptoms to focus more on the
et al., 2010) and the publication of the Fifth aberrant mechanisms that lead to them. Pathology
Edition of the Diagnostic and Statistical Manual in the RDoC was to be defined as abnormality in
of Mental Disorders (DSM-5) that was published any of five domains: negative affect, positive
50 3 Classification and Developmental Psychopathology
affect, cognition, social processes, and regulatory its many revisions since its first publication in
processes. Abnormality in these processes could 1952. The biggest changes came with the publi-
be defined using a matrix that included genes, cation of its third revision in 1980. In the first two
molecules, cells, physiology, and behavior. By editions, the definitions of disorders were clearly
focusing on these mechanisms, the RDoC hopes based on a medical model approach to classifica-
to overcome the problem that behavioral symp- tion. The definitions assumed an underlying path-
toms are frequently multi-determined and, as a ological core, and the conceptualization of the
consequence, diagnoses based on them are fre- core was based on psychodynamic theory. In the
quently heterogeneous in terms of their patho- third edition, there was an explicit switch from a
physiology. It also hoped to explain that the high medical model view of disorders and the reliance
degree of comorbidity among different behav- on psychodynamic theory. In the third edition
ioral disorders is often not due to the co- and continuing into later editions, a functional
occurrence of separate causal mechanisms but approach to defining disorders was used in which
instead reflect different behavioral manifesta- criteria for each diagnosis was based on behav-
tions of the same underlying pathophysiological ioral or emotional symptoms that caused either
process. significant distress or disability (i.e., impairment
The goal of this brief description of the RDoC in one or more important areas of functioning;
is to illustrate the ongoing debate over the rela- American Psychiatric Association, 1980).
tive merits of the two different approaches to
classifying mental disorders. Hereafter, the rest Another major change with the third edition
of this chapter, and in fact, the rest of the book, that has been maintained in subsequent revisions is
largely does not focus on the RDoC but instead an increase in the level of specificity with which
focuses on the DSM. This decision is not due to disorders are defined. In the first two editions of
any inherent disagreement with the underlying the manual, disorders were often poorly defined,
assumptions of this approach, and, in fact, we which led to problems obtaining high levels of
feel that this approach has great promise to reliability in the diagnostic classifications (Spitzer
addressing some of the issues in current classifi- & Cantwell, 1980). In contrast, later revisions
cation systems raised by the developmental psy- include more detailed diagnostic definitions with
chopathology framework discussed in the next explicit symptom lists, which has led to an increase
section. Instead, it was based on several practical in the reliability of the system (e.g., Spitzer,
considerations. First, the RDoC is in the very Davies, & Barkley, 1990). To illustrate this change,
early stages of development, and, as a result, little the DSM-II (APA, 1968) definition of Hyperkinetic
data are available to support the reliability and Reaction of Childhood is contrasted with the anal-
validity of this approach. Second, and also due to ogous DSM-III-R definition of Attention-Deficit
it very short history, most available instruments Hyperactivity Disorder (ADHD) in Box 3.1.
used in the clinical assessment of children’s emo-
tions, behavior, and personality do not provide
data directly relevant to this approach. Third, and Box 3.1 A Comparison of DSM-II and DSM-III-R
perhaps most importantly, while the ability of Diagnostic Criteria
RDoC to guide research on the etiology of many
forms of psychopathology in children and ado- DSM-II: Hyperkinetic Reaction of
lescents has some immediate utility, its ability to Childhood (Adolescence)
guide treatment is still an untested assumption. This disorder is characterized by overactiv-
ity, restlessness, distractibility, and short atten-
Diagnostic and Statistical Manual of Mental tion span, especially in young children; the
Disorders, fifth Edition (DSM-5; APA, behavior usually diminishes in adolescence. If
2013) The DSM approach to defining psychiat- this behavior is caused by brain damage, it
ric disorders has undergone dramatic changes in
Classification 51
should be diagnosed under the appropriate ing possible consequences (not for
nonpsychotic organic brain syndrome. the purpose of thrill seeking), e.g.,
runs into street without looking
DSM-III-R: Attention-Deficit Hyperactivity B. Onset before the age of 7
Disorder C. Does not meet criteria for a Pervasive
Note: Consider a criterion met only if Develop-mental Disorder
the behavior is considerably more frequent
than that of most people of the same mental Source: Diagnosis and Statistical Manual
age. of Mental Disorders, Second Edition,
American Psychiatric Association, 1968
A. A disturbance of at least 6 months dur- and Diagnosis and Statistical Manual of
ing which at least eight of the following Mental Disorders, Third Edition, Revised,
are present: American Psychiatric Association, 1987.
1. Often fidgets with hands or feet or Reproduced with permission of the
squirms in seat publisher
2. Has difficulty remaining seated
when required to do so
3. Is easily distracted by extraneous One source of confusion in the most recent
stimuli editions of the DSM has been a disconnect
4. Has difficulty awaiting turn in between its overall definition of a mental disorder
games or group situations and the specific criteria used to define each indi-
5. Often blurts out answers to ques- vidual disorder. That is, the DSM-5 defines a
tions before they have been mental disorder as “a syndrome characterized by
completed clinically significant disturbance in an individu-
6. Has difficulty following through on al’s cognition, emotional regulation, or behavior
instructions from others (not due to that reflects a dysfunction in the psychological,
oppositional behavior or failure of biological, or developmental processes underly-
comprehension), e.g., fails to finish ing mental functioning” (p. 20; APA, 2013).
chores Thus, this overall definition fits with a medical
7. Has difficulty sustaining attention model approach to classifying psychopathology.
in tasks or play activities That is, the disorder is the “dysfunction” that
8. Often shifts from one uncompleted underlies mental functioning. However, the crite-
activity to another ria for specific disorders largely focus on the
9. Has difficulty playing quietly number of emotional or behavioral symptoms
10. Often talks excessively that need to be present to make the diagnosis, and
11. Often interrupts or intrudes on oth- they do not specify how to determine if they are
ers, e.g., butts into other children’s due to a dysfunction or not. The symptoms
games included in the diagnostic criteria were often
12. Often does not seem to listen to selected based on their pattern of covariation and
what is being said to him or her the point at which the level of symptoms seems to
13. Often loses things necessary for
result in significant impairment in a person (Frick
tasks or activities at school or at et al., 1994). Thus, the actual criteria for the dis-
home (e.g., toys, pencils, hooks, orders in the more recent editions of the DSM are
assignments) based largely on a multivariate or empirical
14. Often engages in physically dan- model of classification.
gerous activities without consider- In summary, in the history of the DSM, the
third edition reflected a major changing point in
52 3 Classification and Developmental Psychopathology
how mental disorders were defined. The later edi- medical models for classification is the fostering
tions reflect much more modest changes. The of a perception that there are clear and dramatic
fifth edition of the DSM makes several changes. differences between healthy and disordered indi-
First, one goal of the DSM-5 was to work in con- viduals and ignoring the fact that there can be
junction with the World Health Organization significant variations in severity even within indi-
(WHO) to “harmonize” as much as possible the viduals with the same disorder. Further, placing
criteria in the DSM-5 with the criteria that was individuals into discrete disorders ignores the
included in the 11th edition of the International fact that multiple disorders might share common
Classification of Diseases (ICD-11; WHO, 2018). etiological factors (Lahey, Krueger, Rathouz,
Second, the DSM-5 was configured to include Waldman, & Zald, 2017). To begin to address the
information on the gender differences in the first issue and recognize variations in severity
causes and expression of disorders at multiple within a diagnostic category, almost all disorders
levels. That is, if warranted by research, gender- in the DSM-5 include specifiers for mild, moder-
specific manifestations of a disorder are reflected ate, and severe manifestations. While most of
in its criteria, and, for all disorders, issues related these specifiers focus on variations in the number
to gender that are pertinent to the diagnosis are of symptoms displayed above those needed to
included in the introductory text in a section meet the diagnostic criteria, some include other
labelled “Gender-Related Diagnostic Issues.” For methods for specifying severity. For example, the
example, in this section for the diagnosis of Oppositional Defiant Disorder (ODD) severity
Conduct Disorder (CD), the DSM-5 includes the specifier focuses on the number of situations in
statement that “whereas males tend to exhibit which the child shows the symptoms of the disor-
both physical and relational aggression (behavior der, based on research showing the importance of
that harms the relationships of others), females the pervasiveness of symptoms for the level of
tend to exhibit relatively more relational aggres- impairment associated with this disorder (Frick
sion ” (p. 474; APA, 2013). Third, another goal of & Nigg, 2012). To address the second issue
the DSM-5 was to reflect research on cultural related to common etiological dimensions,
issues related to the disorders in the manual as grouping of disorders into chapters was guided
well. All disorders include a section in their intro- more by etiological considerations than shared
ductory text labelled “Culture-Related Diagnostic behavioral manifestations. To illustrate this
Issues,” similar to what was described above for change, past versions of the manual grouped the
gender. However, the DSM-5 also includes a sep- diagnoses of ADHD, ODD, and CD together into
arate section on “Cultural Formulation” that con- a category labelled “Attention-Deficit and
tains a detailed discussion of cultural issues that Disruptive Behavior Disorders” largely because
should be considered when making diagnoses, of the very high degree of behavioral covaration
such as cultural identity of the individual being in the symptoms of these disorders. However,
diagnosed, cultural variations in how distress can due to findings that ADHD is associated with
be expressed, and cultural features that can influ- early appearing alterations or immaturities in
ence the relationship between the individual and neural development (Frick & Nigg, 2012), the
the clinician making a diagnosis. DSM-5 chose to place ADHD in the chapter enti-
The final two overarching goals for the DSM-5 tled, “Neurodevelopmental Disorders,” which
are particularly relevant to this chapter of the have as their defining features that they manifest
book because they relate directly to the discus- early in development and produce impairments in
sion on the different models of classification (i.e., personal, social, academic, and occupational
the previous section) and research in develop- functioning (American Psychiatric Association,
mental psychopathology (i.e., the following sec- 2013). In contrast, ODD and CD were placed into
tion). Specifically, a fourth goal of the DSM-5 a separate category of the Disruptive, Impulse
was to foster a more dimensional approach to Control, and Conduct Disorders which are
diagnosis. As noted previously, one criticism of defined by problems in the self-control of emo-
Developmental Psychopathology 53
tions and/or behaviors (American Psychiatric these principles. That is, we focus on the impor-
Association, 2013). tance of using research on normal developmen-
Finally, a fifth important goal of the DSM-5 tal processes to enhance the clinical assessment
was to promote a lifespan view of mental disor- of children and adolescents.
ders, whereby continuities and changes in mani-
festations of disorder across development are
recognized. This lifespan perspective is a critical Developmental Psychopathology
part of the developmental psychopathology
approach to assessment that will be the focus of As mentioned previously, the overriding princi-
the next section of the text. Based on this goal, ple of developmental psychopathology is that
the DSM-5 eliminated a section that was included children’s emotional and behavioral functioning
in previous editions that group disorders “first must be understood within a developmental con-
manifested in childhood and adolescence.” Such text (Rutter & Garmezy, 1983). Therefore, it fol-
a chapter grouping disorders based on timing of lows that the assessment of children’s emotional
onset was not consistent with the goal of promot- and behavioral functioning must also be con-
ing a lifespan view of all disorders. For example, ducted within a developmental framework. Such
it minimized the importance of certain disorders themes as understanding behavior in a develop-
included in this section when making diagnoses mental context and conducting assessment within
in adults, such as ADHD (Barkley, Murphy, & a developmental framework are broad principles
Fischer, 2008). The DSM-5 also includes in the that have several important specific implications
introductory text for all disorders a section for the assessment process.
labelled “Development and Course” to include
age-related issues that might influence the mani-
festation of a disorder or its diagnosis. To illus- Developmental Norms
trate, this section for Generalized Anxiety
Disorder includes the statement that “children First, a developmental approach recognizes that a
and adolescents tend to worry more about school child’s emotional and behavioral functioning
and sporting performance, whereas older adults must be understood within the context of devel-
report greater concern about the well-being of opmental norms. To be specific, there are numer-
family or their own physical health” (p. 223; ous behaviors of children that are common at one
APA, 2013). Similarly, in the introductory mate- age but relatively uncommon at others. For exam-
rial for Specific Phobia, the DSM-5 notes that ple, bedwetting is quite common prior to age 5
excessive fears are quite common in young chil- and even at age 5 is present from 15% to 20% of
dren and are usually transitory, and, as a result, children (Doleys, 1977; Walker, Milling, &
when a diagnosis is being considered in a child, Bonner, 1988). Similarly, childhood fears tend to
“it is important to assess the degree of impair- be quite common, and the types of fears that are
ment and duration of the fear, anxiety, and avoid- most common show a regular progression with
ance” (p. 200; APA, 2013). the development of the child (Campbell, 1986).
Thus, the DSM-5 attempted to include infor- For example, separation anxiety is not uncom-
mation that would help a clinician consider a mon in infants toward the end of the first year of
number of important developmental issues life (Bowlby, 1969), whereas fears of the dark
when using the manual to make diagnoses of and imaginary creatures are quite common in
mental disorders. These changes were a very preschool and school-age children but decrease
important advance in the classification of psy- in prevalence with age (Bauer, 1976).
chopathology, especially for children and ado- These are just a few of the many developmen-
lescents. In the following section, we expand on tally related changes in the prevalence of specific
the underlying principles that guided these child behaviors. Knowledge of these develop-
changes and describe practical implications of mental changes in the behavior of children is
54 3 Classification and Developmental Psychopathology
crucial to clinical assessment because the same cess will provide only a limited, and sometimes
behavior may be developmentally appropriate at misleading, understanding of a child’s psychoso-
one age but indicative of pathology at another. cial functioning. For example, understanding the
Therefore, assessment of children and adoles- family environment of a child with behavioral
cents must allow for developmentally based problems will only provide a limited perspective
interpretations. The critical nature of these inter- on how these behavior problems developed with-
pretations means that selection of assessment out also considering the child’s temperament,
techniques must be based, at least in part, on the which may make the child more difficult to raise.
availability of age-specific norms. Further, given Such transactional relations among developmen-
the rapid developmental changes experienced by tal processes necessitate that a professional
children and adolescents, comparisons must be design batteries that assess for many different
made within fairly limited age groups. Whereas types of processes and consider potential transac-
for adults using a comparison group that spans tional relations when interpreting the testing
the ages from 25 to 35 may be justifiable, a com- results.
parison group of children that spans the ages Second, the process-oriented developmental
5–15 would be meaningless, given the many approach recognizes that the same developmen-
changes in development that are subsumed within tal process (e.g., a permissive rearing environ-
this period. Because the normative information ment) may result in many different outcomes
provided by an assessment instrument and the (e.g., some children who are creative, others
appropriate use of norm-referenced information who are dependent, and others who are antiso-
by the assessor are critical components to the cial), a concept called “multifinality.” The com-
clinical assessment of children, these issues are plementary concept is “equifinality,” which
discussed in great detail throughout this book. refers to the concept that the same outcome
(e.g., antisocial behavior) can result from very
different developmental processes across indi-
Developmental Processes viduals (e.g., very strong emotional reactivity
leading to strong angry and aggressive reactions
Unfortunately, many assessors believe that sim- or weak emotional reactivity leading to prob-
ply comparing assessment information to age lems in the experience of empathy and guilt).
norms is all that is needed to take a developmen- The implications for the assessment process are
tal approach to child and adolescent assessment. that assessors need to expect that the same per-
This is a much too limited view of development sonality pattern or psychopathological condi-
and how it can be applied to understanding both tion may result from very different processes
normal and pathological outcomes in children. A across individuals and batteries need to be
developmental approach is a “process-oriented” designed to assess for these various “causal
approach. Put simply, this means that any devel- pathways.” This concept, in fact, may be one of
opmental outcome, be it a normal personality the most critical in a developmental psycho-
dimension or a problematic pattern of behavior, pathological approach to assessment because it
is the end result of an interaction of numerous places the assessment process in the important
interrelated maturation processes (e.g., socio- role of uncovering the unique causal pathway
emotional, cognitive, linguistic, biological). This that leads to a child’s current functioning so that
process-oriented approach has several important interventions can be better tailored to unique
implications for how assessors conceptualize needs of the child (see, e.g., Frick, 2012).
what they are trying to assess and how they go Third, how the various processes unfold over
about doing it. development leads to specific “tasks” that may
First, given the interrelated nature of the matu- make certain behaviors more likely to occur at
rational processes, this approach recognizes that certain points in development. These unique
a focus on any single type of developmental pro- demands, or “developmental tasks,” lead to many
Developmental Psychopathology 55
of the age-related changes in children’s emotions lescence, the conduct problems may be best
and behaviors discussed previously. Comparing a conceptualized as an exaggeration of a normal
child’s behavioral or emotional functioning to developmental process (e.g., identity develop-
developmental norms can help to determine ment). In contrast, a preadolescent child who
whether the child’s functioning is deviant com- exhibits deviant levels of conduct problems is
pared to other children who are experiencing showing a behavior that seems more qualitatively
similar developmental demands. However, sim- different from what is expected from normal
ply comparing a child’s behavior to developmen- developmental processes. It may, therefore, be an
tal norms and determining whether the child’s indication of more severe pathology. This conclu-
functioning is deviant compared to other children sion is consistent with research indicating that
of the same age does not allow the assessor to conduct problems that have onset in adolescence
determine whether (1) the child’s problems are more likely to be transient, whereas conduct
should be considered an exaggeration of the nor- problems with a prepubertal onset tend to be
mal maturational processes operating at that more severe and chronic (Moffett, 2018).
stage of development or (2) the child’s problems
should be considered as a qualitative deviation
from normal development (i.e., not consistent Stability and Continuity
with the specific demands of that developmental
stage), with the latter often being indicative of a Issues regarding the stability of childhood behav-
more severe pathological process. ioral and emotional functioning are important to
These two different interpretations of devia- a developmental perspective to psychopathology.
tions from developmental norms can be illus- These complex issues have important implica-
trated in the assessment of conduct problems in tions for the assessment process. The basic issue
children and adolescents. Research has docu- of stability is not unique to the assessment of
mented an increase in acting-out behavior for children and adolescents but has been a long-
both boys and girls that coincides with the onset standing controversy in psychological assess-
of adolescence (e.g., Offord et al., 1989). The ment throughout the life span. For example,
first implication of this finding is that the assess- many have questioned the whole concept of per-
ment of adolescent conduct problems should be sonality, because it implies a consistency of
based on a comparison to adolescent norms so behavior over place and time that is often not
that age-specific deviations can be determined. apparent in human behavior (Mischel, 1968).
However, to interpret these age-specific devia- This issue is even more relevant to children
tions, it is also helpful to realize that an increase because childhood behavior seems even less sta-
in acting-out behavior in adolescence is consis- ble over time and situations than found in adults,
tent with the identity formation process outlined making the concept of personality in children
in Erik Erikson’s psychosocial theory of person- even more controversial. Our view of the debate,
ality development (Erikson, 1968). Specifically, which is similar to the view of many other theo-
Erikson characterizes adolescence as a time when rists (e.g., Buss, 1995; Martin, 1988), is that the
youths are struggling with the development of an concept of personality can be useful if conceptu-
individual identity, one that is separate from their alized appropriately but dangerous if viewed
parents. Rebellion and questioning of authority wrongly.
are manifestations of adolescents’ rejection of For example, many measures of children’s
parental and societal values, as they struggle to behavior or other aspects of personality show
develop their unique identity. Understanding this much less stability in children than do analogous
process allows for an additional interpretation constructs in adults. Specifically, Roberts and
when one documents deviations from age norms. DelVecchio (2000) reported a meta-analysis of
If a youth exhibits developmentally deviant lev- 152 longitudinal studies assessing the average
els of conduct problems for the first time in ado- stability of personality traits in different age
56 3 Classification and Developmental Psychopathology
groups over a 6- to 7-year period. They reported increase in stability through aggregation of
that the average stability coefficient for children behaviors can also be conceptualized from basic
and adolescents was 0.31, compared to 0.54 for measurement theory. It has consistently been
young adults, 0.64 for middle adulthood, and shown that increasing the number of items on a
0.75 for adults over the age of 50. These findings measure of a trait also increases its reliability
are not surprising given the rapid developmental (Ghiselli, Campbell, & Zedeck, 1981). Hence,
changes that are occurring in childhood. However, the increased stability may be a function of a
the findings have important implications for the more reliable method of measurement.
interpretation of personality measures in chil- Finally, stability can be affected by whether
dren. Specifically, interpretations of dispositional one is viewing the developmental outcome or the
characteristics must be made cautiously in chil- processes that may have led to this outcome. For
dren so that there is no implication of stability example, the type of adjustment problem may be
over time, unless data are available to support episodic, as in the case of depression, but the fac-
such an interpretation. Given that the data are tors that led to this problem, such as problems in
lacking in most cases, the term personality may emotional regulation, may continue even after the
be misleading for many domains of child depression has remitted and could place the child
behavior. at risk for other adjustment problems in the
Although we feel that this caution is war- future. As such, depression may not appear sta-
ranted, there are also several ways in which this ble, but the problems in emotional regulation are
general statement must be qualified. First, there stable (Keenan, 2000). Again, this example illus-
is clearly some continuity in children’s behavior, trates the relevance of the process-oriented
and the degree of stability (or instability) seems approach to assessment that is consistent with the
dependent on which domain of behavior is being developmental psychopathology framework.
assessed. For example, research generally indi- This framework illustrates the importance of not
cates that externalizing behaviors (e.g., hyperac- simply assessing developmental outcomes (e.g.,
tivity, aggression, antisocial behavior) tend to be psychopathological conditions) but also assess-
more stable over time than internalizing behav- ing the various interacting processes that can lead
iors (e.g., fears, depression; Frick & Loney, 1999; to these outcomes.
Ollendick & King, 1994). Therefore, interpreta-
tions of the stability of behavior must be depen-
dent on the dimension of behavior that is of Situational Stability
interest.
In addition, aggregates of behaviors (behav- Explicit in the developmental psychopathologi-
ioral domains) tend to be more stable than dis- cal perspective is a transactional view of behav-
crete behaviors. For example, Silverman and ior. That is, a child’s behavior influences his or
Nelles (1989) reported on the 1-year stability of her context and is also shaped by the context. As
mothers’ reports of fears in their children between a result, one expects a high degree of situational
the ages of 8 and 11. Over the 1-year study variability in children’s behavior based on the
period, there was only a 10% overlap between differing demands present across situations.
Time 1 and Time 2 in the 10 specific objects or Providing some of the best data on this issue,
situations that mothers reported eliciting the most Achenbach, McConaughy, and Howell (1987)
fear in their child. However, the correlation conducted a meta-analysis of 119 studies that
between the absolute number of fears was quite reported on the correlations between the reports
high. Therefore, although the specific types of of different informants on children’s and adoles-
fears were not stable over the study period, the cents’ (ages 1½–19 years) emotional and behav-
level of fears was stable. Some have argued that ioral functioning. The correlations between
aggregation allows one to pick up generalized different types of informants (e.g., parent-
response tendencies that are not captured by dis- teacher) were fairly low, averaging about 0.28.
crete behaviors (Martin, 1988). However, this This low correlation is not a good indicator of
Developmental Psychopathology 57
and adolescents. In a special issue of the Journal 2. Children’s behaviors and emotions must be
of Consulting and Clinical Psychology (Kendall understood within a developmental context.
& Clarkin, 1992), several articles highlighted the Therefore, an important characteristic of
unique treatment needs of children with various assessment instruments for children is their
types of comorbid psychopathological condi- ability to provide developmentally sensitive
tions. The high degree of comorbidity and its normative comparisons. On a more general
importance to the design of effective treatment level, appropriate interpretations of test
programs makes the assessment of comorbid scores, even if they are based on age-specific
conditions crucial in the clinical assessment of norms, should be guided by a knowledge of
children and adolescents. developmental processes and their effect on a
Because of the importance of comorbidity, child’s behavioral and emotional functioning.
most clinical assessments of children and adoles- 3. Children’s behavior is heavily dependent on
cents should be comprehensive. Specifically, the contexts in which the child is participat-
assessments must cover multiple areas of func- ing. Therefore, assessments of children must
tioning so that not only are the primary referral be based on multiple sources of information
problems assessed adequately, but potential that assess a child’s functioning in multiple
comorbid problems in adjustment are also contexts. In addition, an assessment of the rel-
assessed. The clinical assessment must be evant aspects of the many important contexts
designed with a thorough understanding of the in which a child functions (e.g., at school, at
high degree of comorbidity present in child psy- home, with peers) is crucial to understanding
chopathology generally and the most common the variations in a child’s behavior across
patterns of comorbidity that are specific to the settings.
referral question. 4. Most assessments of children must be com-
prehensive. This is necessary not only because
of the need to adequately assess the many
Practical Implications for Assessment important situational contexts that influence a
child’s adjustment. Children often exhibit
Although we have tried to summarize some of problems in multiple areas of functioning that
the major findings in the field of developmental span emotional, behavioral, learning, and
psychopathology that have particular relevance social domains. Effective treatments must be
to clinical assessment, sometimes it is difficult based on the unique strengths and weaknesses
to translate research into guidelines for prac- of the child across these multiple psychologi-
tice. The following is a summary of some of cal arenas.
the major implications of the findings dis-
cussed in this section applied to the clinical
assessment of children and adolescents. These Conclusions
implications for assessment are expanded on
and applied to specific situations throughout The main theme of this chapter and, in fact, of
this book: this entire text, is that appropriate assessment
practices are based on a knowledge of the
1. A competent assessor needs to be knowl- basic characteristics of the phenomena one is
edgeable in several areas of basic psycho- trying to assess. As a result, the competent
logical research to competently assess assessor is knowledgeable not only in test
children and adolescents. In addition to com- administration but is well-versed in psycho-
petence in measurement theory, knowledge metric theory, child development, and child-
of developmental processes and basic char- hood psychopathology.
acteristics of childhood psychopathology is Clinical assessment can be conceptualized as a
also essential. process of classification. Therefore, understanding
60 3 Classification and Developmental Psychopathology
anyone involved in the clinical assessment pro- Table 4.1 Content in the Standards for Educational and
Psychological Testing (AERA, APA, NCME, 2014)
cess should be familiar with the both the Ethical
Principles and the Standards and have a copy of Part 1. Foundations
them available for periodic reference. Further, 1. Validity
(a) Cluster 1. Establishing Intended Uses and
any class devoted to training clinical assessors Interpretations
should have both of these guides as required (b) Cluster 2. Issues Regarding Samples and
reading. Settings Used in Validation
In this chapter, we will not provide an exhaus- (c) Cluster 3. Specific Forms of Validity Evidence
tive restatement of all of the important points 2. Reliability/Precision and Errors of Measurement
made by either the APA Ethical Principles or the (a) Cluster 1. Specifications for Replications of
the Testing Procedure
Standards. In Table 4.1 we provide a summary of
(b) Cluster 2. Evaluating Reliability/Precision
the content of the Standards. Many of the
(c) Cluster 3. Reliability/Generalizability
Standards relate to material covered in other Coefficients
parts of this text. For example, issues related to (d) Cluster 4. Factors Affecting Reliability/
validity (Standards, Section 1), reliability Precision
(Standards, Section 2), and norm- referenced (e) Cluster 5. Standard Errors of Measurement
scores (Standards, Section 5) were covered in (f) Cluster 6. Decision Consistency
the chapter on Measurement Issues (Chap. 2). (g) Cluster 7. Reliability/Precision of Group
Means
Issues related to scoring, reporting, and interpre- (h) Cluster 8. Documenting Reliability/Precision
tation are covered in more detail in our chapter 3. Fairness in Testing
on Integrating and Interpreting Assessment (a) Cluster 1. Test Design, Development,
Information (Chap. 15) and Report Writing Administration, and Scoring Procedures That
(Chap. 16). Also, some parts of the standards Minimize Barriers to Valid Score Interpretations for
the Widest Possible Range of Individuals and
cover areas that are not directly relevant to the Relevant Subgroups
clinical assessment of child and adolescent per- (b) Cluster 2. Validity of Test Score
sonality and behavior, such as the sections on Interpretations for Intended Uses for the Intended
test design and development (Standards, Section Examinee Population
4), workplace testing and credentialing (c) Cluster 3. Accommodations to Remove
Construct-Irrelevant Barriers and Support Valid
(Standards, Section 11), and use of tests for pro- Interpretations of Scores for Their Intended Uses
gram evaluation, policy studies, and account- (d) Cluster 4. Safeguards against Inappropriate
ability (Standards, Section 13). Instead, in this Score Interpretations for Intended Uses
chapter, we attempt to highlight a few issues that Part 2. Operations
we have found to be especially important to the 4. Test Design and Development
clinical assessment of children and adolescents (a) Cluster 1. Standards for Test
and constitute the basics of fair and ethical prac- (b) Cluster 2. Standards for Item Development
and Review
tice in testing.
(c) Cluster 3. Standards for Developing Test
Administration and Scoring Procedures and
Materials
ights and Responsibilities of Test
R (d) Cluster 4. Standards for Test Revision
Users 5. Scores, Scales, Norms, Score Linking, and Cut
Scores
(a) Cluster 1. Interpretations of Scores.
Informed Consent
(b) Cluster 2. Norms
(c) Cluster 3. Score Linking
Before providing any clinical service to a child (d) Cluster 4. Cut Scores
are adolescent, it is both a legal and ethical 6. Test Administration, Scoring, Reporting, and
requirement to obtain informed consent for these Interpretation
services. A sample consent form is provided in (a) Cluster 1. Test Administration
Box 4.1. As illustrated in this form, the basic ele- (b) Cluster 2. Test Scoring
ments of informed consent are: (continued)
Rights and Responsibilities of Test Users 63
Box 4.1 Sample Parental Consent Form for Clinical Assessments of Children and Adolescents
Purpose of the Service: The purpose of this clinic is to provide a comprehensive, state-of-the-art,
affordable psychological assessment of children and adolescents who are experiencing behavioral,
emotional or learning problems. The results of the evaluation will be summarized in a report that will be
reviewed with my child and me. This report will include recommendations on interventions that may
enhance my child’s adjustment.
Procedures:
a. I will be asked questions about my child’s interests, friends, attitudes, emotions, and
behaviors. I will also be asked questions about my neighborhood, my family, and my
emotions and behaviors.
b. My child will be asked questions about his/her interests, attitudes, emotions, and
behaviors. He/she will also be asked questions about our family, his friends, his
neighborhood, and his school. He/she will participate in tasks that assess his/her thoughts,
emotions, and behaviors.
c. My child will be given standardized tests assessing his/her intelligence and learning.
d. My child’s teacher will be asked to answer questions and complete rating scales assessing
my child’s emotions, behavior, and learning at school.
Benefits: The benefit will be a thorough psychological assessment of my child or adolescent and a report
making recommendations for any interventions that may help him or her.
Risks/Discomforts: Some persons may not feel comfortable answering personal questions about their
emotions and behaviors. The purpose of asking particular questions will be explained to me and my child
and I and my child can refuse to answer specific questions, although this may limit the conclusions that
can be made in the final report.
(continued)
Rights and Responsibilities of Test Users 65
Box 4.1 (continued)
Right to Refuse and Alternatives: Since this is a service that I have voluntarily requested for my child, I
can discontinue the testing at any time. I also recognize that similar services may be available at other
mental health clinics.
Privacy: Records with identifying information will be kept in a locked file cabinet that can only be
accessed by approved C-HAS staff. The information from the evaluation will not be shared with anyone
without a signed release by me or another legal guardian of my child. The only exceptions are if someone
is in imminent danger of harming themselves or others, if information is provided on current or past child
or elder abuse, or if I attempt to receive third-party payment for services.
Financial Information: The cost for participation in this study is ____________. Payment in full is
required prior to testing. Participants who withdraw will pay a pro-rated fee based on the assessments
given and the time involved.
Signatures:
I consent for my child to receive the services described above.
____________________________________ ________________________
Parent/guardian Signature Date
*Reader of the consent form, please sign the statement below if the consent form was read to the parent
because he/she is unable to read: The parent/guardian has indicated to me that he/she is unable to read. I
certify that I have read this consent form to the parent/guardian and explained that by completing the
signature line above, he/she has provided consent for their child to receive the services listed above.
____________________________________ ________________________
Signature of Reader Date
Importantly, children under the age of consent different ages is discussed in more detail in the
cannot legally provide informed consent for next section covering rapport building strate-
services. However, children still have the right to gies. The one notable difference in the informed
have the policies and procedures explained to consent given to parents and legal guardians and
them in a language that is appropriate to their these “assent procedures” for the child is that the
development level. As noted in the APA Ethical child is not provided the right to refuse partici-
Principles, for persons who are legally incapable pation. Instead, the assessor should consider the
of giving informed consent, psychologists child’s preferences and best interests. This dif-
nevertheless: ference is based on the notion that minors are
not yet capable of weighing the costs and bene-
• “provide an appropriate explanation, fits of receiving services in order to make
• seek the individual’s assent, and informed decisions on the receipt of services. A
• consider such persons’ preferences and best form that can be used to document this “assent”
interests” (3.10; APA, 2016). procedure is provided in Box 4.2.
Another way that jurisdictions can vary is in
To highlight the key parts of this principle, the the reporting child abuse. Psychologists are
child being tested should receive a description typically mandatory reporters of abuse, which
of the testing procedures, and this explanation means that they are legally mandated to report
should be tailored to the child’s development any suspected cases of child abuse (both physical
level. How to describe testing to children at and sexual) and neglect. However, jurisdictions
66 4 Ethics in Testing: Rights, Responsibilities, and Fairness
Box 4.2 Sample Child Assent Form for Clinical Assessments of Children and Adolescents
My parents have request that I undergo a comprehensive psychological evaluation through the Caring
Huskies Assessment Service (C-HAS).
a. My parent will be asked questions about my interests, attitudes, emotions, and behaviors.
I will also be asked questions about my neighborhood, my family, and my emotions and
behaviors.
b. I will be asked questions about my interests, attitudes, emotions, and behaviors. I will
also be asked questions about my family, friends, neighborhood, and school. I will also
participate in tasks that assess my thoughts, emotions, and behaviors.
d. My teacher will be asked to answer questions and complete rating scales assessing my
emotions, behavior, and learning at school.
Since this is a service that my parent has voluntarily requested for me, my parent can stop the testing at
any time.
All of the information obtained during the evaluation will be summarized in a report that will be reviewed
with my parent and me. The information from the evaluation will not be shared with anyone else without
the permission of my parent or legal guardian. The only exceptions are if someone is going to hurt
themselves or others or if someone is hurting a child, including me.
____________________________________ ________________________
Child Signature Date
____________________________________ ________________________
Witness Date
*Witness must be present for the assent process, not just the signature by the minor.
can vary in terms of the definitions of abuse and One final note about informed consent is that
neglect and how and to whom such reports of it should be the first thing done in a clinical
abuse should be made. Again, it is thus assessment, without exception. That is, clini-
imperative that assessors are knowledgeable cians can be tempted to start their therapeutic
about all applicable laws and policies that relationship off with something less formal than
govern the reporting of abuse, so that these can reviewing the basic elements of informed con-
be strictly adhered to and accurately conveyed sent in order to help the parent and child become
during informed consent. comfortable with the clinic. In addition, a parent
Rights and Responsibilities of Test Users 67
may be anxious to tell the clinician about their • Ethical principles related to testing children
concerns and why they are having their child and adolescents (Chap. 4)
tested without waiting to complete the consent • Measurement/psychometric theory (Chap. 2)
procedures. However, it is imperative that par- • Developmental psychopathology research
ents be fully informed about the testing process (Chap. 3)
and their rights and responsibilities prior to • Research on common problems in develop-
starting the assessment. As noted in the next ment experienced by children and adolescents
chapter, this actually can help establishing rap- (Chaps. 17, 18 and 19)
port by making it clear that both the parent and
child are informed participants in the assessment Thus, a critical component of the principle of
process by clearly explaining everything that “competent use” in psychological assessment is
will be done during the evaluation. Also, a par- that clinical assessors should be competent in the
ent should be fully informed on the limits to methods used (i.e., be a competent user).
confidentiality before they potentially provide However, just as importantly, the assessor should
information that the assessor may be legally engage in competent test administration (i.e.,
required to disclose (e.g., suspected cases of competent use). These are related but not identi-
child of abuse). cal issues. That is, a person can have adequate
training and competency in the tests administered
but fail to engage in appropriate administration
Competent Use procedures. There are two main parts to compe-
tent use. First, Standard 6.1 of the Standards
A critical ethical principle guiding test adminis- states that, “test administrators should follow
tration is that that an assessor should only admin- carefully the standardized procedures for admin-
ister tests for which he or she is competent. As istration and scoring specified by the test devel-
stated by the Standards (i.e., Standards 6.0): oper” (p. 114, AERA, APA, NCME, 2014).
Those responsible for administering, scoring, Second, testers need to ensure that administration
reporting, and interpreting should have sufficient procedures avoid introducing “construct-irrele-
training and supports to help them follow the vant variance” into the testing process. Construct-
established procedures. (p. 114; AERA, APA, irrelevant variance refers to factors that affect a
NCME, 2014)
person’s performance on a test that are unrelated
Tests can vary greatly in how much training is to the construct that is being measured. For
needed to be proficient at their administration, example, noises and other disruptions in the test-
scoring, and interpretation. However, all tests ing area, extremes of temperature, poor lighting,
specify a minimal level of training, and this typi- illegible materials, and malfunctioning comput-
cally includes both didactic and practical (i.e., ers can all influence a person’s performance on a
supervised practice) training. It is incumbent on test. Unless the goal of the test is to assess the
the assessor to assure that this specified level of child’s performance under certain conditions
competence is achieved before using a test in a (e.g., ability to ignore distractions), these influ-
clinical assessment. ences then could add variability to the child’s
When developing competencies, most clinical performance that is not related to the skill or trait
assessors focus on learning and mastering the being assessed.
administration and scoring of individual tests. These factors are all related to the administra-
However, we feel that there are also some more tion of tests. Similar issues are related to scoring
general competencies that are important for ade- the test. Like administration, scoring procedures
quately conducting clinical assessments of chil- can vary greatly in terms of their difficulty, with
dren and adolescents. This view led to the content some requiring quite detailed coding of responses
of this text. Specifically, assessors should have and others requiring simply entering responses
both didactic and supervised experience in: into a computer. Problems with the former can
68 4 Ethics in Testing: Rights, Responsibilities, and Fairness
result from inadequate training, and problems in support certain interpretations from a test (Frick,
the latter can result from carelessness. In both 2000). As stated by the APA Ethical Principles:
cases, however, such errors can lead to scores that Psychologists base the opinions contained in their
lead to invalid interpretations about the child recommendations, reports, and diagnostic or eval-
being tested. uative statements, including forensic testimony,
on information and techniques sufficient to sub-
stantiate their findings. (9.01; APA, 2016)
Psychologists take reasonable steps to ensure that psychologists have a primary obligation and take
explanations of results are given to the individual reasonable precautions to protect confidential
or designated representative, unless the nature of information obtained through or stored in any
the relationship precludes provision of an expla- medium. (4.01; APA, 2016)
nation of results (such as in some organizational
consulting, pre-employment or security screen- A few critical questions should be considered for
ings, and forensic evaluations), and this fact has
been clearly explained to the person being
protecting confidential testing information:
assessed in advance. (9.10; APA, 2016)
• How will test information be stored and access
An important part to this principle is the need to limited to only authorized staff?
provide “an explanation of results.” That is, we • What are the procedures that will be used to
do not feel it is sufficient to mail test results to assure that the release of information is only
parents and allow them to call if they have ques- made with appropriate consent and to autho-
tions. We feel that it is imperative that the clinical rized individuals?
assessor “take reasonable steps” to discuss the • How can the confidentiality of test informa-
results of an evaluation with the parent and child, tion be ensured during the scoring process?
preferably in person, and allow both the parent • How can discussions of case material be lim-
and child to ask questions and fully understand ited to only those situations in which privacy
the results and their implications. This explana- can be assured?
tion includes a discussion of any diagnosis made
as a result of the testing. One final issue related to the confidentiality of
In Chap. 16, we provide recommendations for test material is when a clinical assessor should
how to communicate test results to parents, not release information, even with the consent of
teachers, and children. As we note in this discus- the person being tested, or in the case of children
sion, this description needs to be tailored to par- and adolescents below the age of consent, even
ents’ educational level and to the child’s with parental consent. The most important of
developmental level. For example, we do not rec- these instances is to “protect a client/patient or
ommend that young children necessarily be others from substantial harm or misuse or mis-
given “diagnostic terms” like Attention-deficit/ representation of the data or the test” (9.04; APA,
Hyperactivity Disorder, until they can adequately 2016). A common example of this is when the
understand what such a term means. In fact, for testing is outdated and may lead to erroneous
young children, the use of such terms could actu- interpretations of the child’s current functioning.
ally hurt the child’s understanding of the testing Another example is when a client requests that
results. Instead, describing the child as having a raw test material (e.g., child or parent’s directs
weakness in his ability to keep his mind on things responses to test questions or stimuli) to be
and stay seated that is making it hard to learn in released to someone who is not qualitied to accu-
school would be more understandable to a young rately interpret the results.
child than the use of a diagnostic term.
As noted in our discussion of informed consent, One final ethical issue related to clinical assess-
it is critical that clinical assessors make clear to ments does not involve individual clients but
both the child being tested and his or her parents relate to protecting the integrity of the test mate-
how the confidentiality of results will be pro- rial themselves. “Testing materials” refer to
tected and what limits to this confidentiality are manuals, instruments, protocols, and test ques-
required by law. The issue of confidentiality is tions or stimuli that are used in standardized
worth additional comment because: assessments. It can be challenging to maintain
70 4 Ethics in Testing: Rights, Responsibilities, and Fairness
the security of test content given that many Most clinical assessors are very familiar with
sources may request access to records, such as the concept of making “accommodations” in
requests from parents, lawyers, or patients tests for persons with disabilities, whereby test
themselves to see the actual record forms used administration procedures are modified to adapt
for the evaluation. Some internet sites even pro- to a person’s disability in order to obtain a more
vide sample items that are analogous to items accurate estimate of a person’s performance. An
found on popular tests, such as the MMPI-2 or example would be using a Braille test form or
the Rorschach. However, such test content large print answer sheets for someone who may
should be shielded from would-be test takers. be visually impaired. Another example would be
Without such security, the content could be mis- using voice-administered questions on a person-
used by unqualified persons, resulting in harm ality inventory for a child who has a reading dis-
to others. Also, if clients have access to test ability. Many such accommodations are intuitive
material prior to testing, this could invalidate and enhance the accuracy of the test results.
their performance on the material under the However, the difficulty comes in when deciding
standardized testing situation. Thus, when faced if any accommodations break the standardiza-
with questions about test security, clinical asses- tion of the tests in way that would decrease the
sors should: accuracy of the test scores and resulting inter-
pretations. That is, many tests require strict
• Explain the problems associated with release adherence to administration procedures, such as
of items for the ability to practice psychologi- uniform directions, specific time limits, and even
cal assessment with others and specific ways in which the room is arranged in
• Release test material only to other profession- order to interpret the tests. Such standardization
als qualified to interpret them properly for the allows the scores to be compared to the scores of
person making the request. others who were tested under the same condi-
tions. If these procedures are not followed, it
makes comparisons to the performance of others
Fairness tested under different conditions questionable.
Further, accommodations that inherently change
Fairness During the Testing Process the skill or ability being assessed also would
lead to erroneous interpretation of the test
A critical consideration in the ethical and accu- results. For example, allowing a child unlimited
rate conduct of clinical assessments is the issue time to complete a task measuring his or her
of “fairness.” The Standards provide an overall speed of processing information would obvi-
definition of fairness as: ously invalidate the results of the testing.
Responsiveness to individual characteristics and Many tests provide instructions for making
testing contexts so that test scores will yield valid accommodations for persons with disabilities and
interpretations for intended uses. (p. 50; AERA, provide evidence to support the comparability of
APA, NCME, 2014) scores and interpretations when these modifica-
tions are used. In these cases, assessors can con-
The Standards also distinguish between fairness fidently make evidence-based modification.
in how the person being tested is treated during However, in some cases, such evidence may not
the testing process and fairness in how the results be available. Thus, clinical assessors should con-
of testing are interpreted. With respect to sider two key questions when testing children
former: and adolescents with disabilities:
Regardless of the purpose of testing, the goal of
fairness is to maximize, to the extent possible, the • Can the testing procedures be modified to
opportunity for test takers to demonstrate their accommodate to the child’s disability in a way
standing on the construct(s) the test is intended to that will enhance and not reduce the accuracy
measure. (p. 50; AERA, APA, NCME, 2014)
of results?
Fairness 71
• If not, is the test appropriate for this child, been validated in research for the specific group
given his or her disability? of persons that includes the person being tested.
In this section, we cover some important con-
Accommodating to the needs of individual cepts that specifically relate to this form of test
children and adolescents in the clinical assess- validity and how to avoid making biased inter-
ment process goes beyond testing persons with pretations from test results.
disabilities. Similar considerations should be
made when testing individuals from diverse lin- Test Bias Differential item functioning (DIF) is
guistic, cultural, ethnic, racial, and socioeco- said to occur when equally able test takers differ
nomic backgrounds as well. Again, the same two in their probabilities of answering a test item
questions are imperative to consider: correctly as a function of group membership.
This can lead to different groups of individuals
• Can the testing procedures be modified to having mean score differences on the test, such
accommodate to the child’s background in a as children from families with less education
way that will enhance and not reduce the scoring higher on a measure of conduct prob-
accuracy of results? lems or African-American adolescents scoring
• If not, is the test appropriate for this child, higher on a measure of extraversion. Generally,
given his or her background? such mean differences on tests are not consid-
ered a meaningful indicator of bias (Reynolds &
Such modifications can range from testing the Kaiser, 1990). Instead, differential test function-
child in the language in which he or she is most ing (DTF) is usually considered evidence for
proficient to other modifications that enhance the test bias, and this refers to evidence that the
child’s comfort with who is conducting the test group differences in test scores result in differ-
and with the setting in which the test is ences in how well the test measures what it is
administered. supposed to measure for the different groups.
Again, this can be framed in terms of differences
in the validity of how well the test measures the
airness and Lack of Measurement
F construct of interest, and, as we discussed in
Bias Chap. 2, there are many types of validity that
can be influenced by DIF.
The next issue critical for fairness in clinical Content validity bias occurs when the con-
assessment is whether the test measures the tent of a test seems inappropriate or, perhaps,
construct of interest in the same way across even offensive to a group of individuals. This is
different groups of individuals. Stated another usually tested by having individuals from differ-
way, this issue focuses on whether the interpre- ent groups review the items of test and judge
tations that can be made from the test results are whether any items are inappropriate or insult-
equally valid across different groups of individ- ing. Construct validity bias exists when the
uals. If the validity of the interpretations are not measurement of a trait differs across groups.
equally valid across groups, this is evidence of One of the most popular methods used to study
“test bias” or “measurement bias” (Kamphaus, construct validity bias is factor analysis. For
2001). Thus, this issue is a specific instance of example, one would want to test whether the
the broader principle of evidence-based inter- factor structure of the test of personality is
pretations that was discussed previously in this invariant (i.e., does not change) across sex, lan-
chapter, whereby assessors only make interpre- guage, race, or ethnicity. For example, van de
tations from test scores that have been validated Looij-Jansen et al. (2011) demonstrated invari-
in research. This makes it more specific by ance in the five-factor structure (i.e., emotional
focusing on the fact that assessors should only symptoms, conduct problems, hyperactivity-
make interpretations from test scores that have inattention, peer problems, prosocial behavior)
72 4 Ethics in Testing: Rights, Responsibilities, and Fairness
in the Strengths and Difficulties Questionnaire assessors, should regularly review these guide-
across gender, age, parental level of education, lines and engage in discussions with colleagues
and ethnicity in a sample of 11,881 students in when questions come up as to how to apply them
the Netherlands. Predictive-validity bias occurs in practice. In this chapter, we provide a brief
when scores on a test predict important out- overview of some of the key principles that we
comes differently for different groups of indi- feel are especially important for the clinical
viduals. In the example provided in which a test assessment of child and adolescent personality
showed that African-Americans scored higher and behavior. However, of even greater impor-
on a measure of extraversion, predictive-validity tance, we use these principles throughout this
bias would be demonstrated if the scores on the text to guide our discussion of, and the recom-
scale also predicted more friendships, greater mendations we provide for, all parts of the
social skills, and greater positive affect for assessment process.
Caucasian youth than for the African-American
adolescents. Chapter Summary
Table 4.2 Multicultural Guidelines: An Ecological Approach to Context, Identity, and Intersectionality (APA, 2017)
Some key implications for the clinical assessment of child and
Guideline adolescent personality and behavior
Guideline 1. Psychologists seek to recognize and Avoid overgeneralized or simplistic categories or labels of
understand that identity and self-definition are sociocultural groups
fluid and complex and that the interaction between Recognize how a person’s identity is shaped by cultural
the two is dynamic. To this end, psychologists forces
appreciate that intersectionality is shaped by the Recognize that there can be important generational
multiplicity of the individual’s social contexts differences in cultural identity, reflected in differences
between children and their parents that could influence the
interpretation of test results
Guideline 2. Psychologists aspire to recognize and Recognize that behavior that may be normative in one
understand that as cultural beings, they hold sociocultural context may be considered pathological and
attitudes and beliefs that can influence their maladaptive in another
perceptions of and interactions with others as well Examine preexisting beliefs and assumptions about persons
as their clinical and empirical conceptualizations. from different sociocultural groups
As such, psychologists strive to move beyond Recognize how their and their clients’ sociocultural identities
conceptualizations rooted in categorical and experiences could influence interactions during the
assumptions, biases, and/or formulations based on assessment process
limited knowledge about individuals and Recognize the great heterogeneity within individuals in a
communities sociocultural group
Engage in ongoing training, education, and personal
reflection on multicultural issues
Guideline 3. Psychologists strive to recognize and Recognize that language describing cultural identity in
understand the role of language and conversations with diverse clients and in test reports
communication through engagement that is summarizing results of an evaluation conveys perceptions of
sensitive to the lived experience of the individual, and feelings about a particular group
couple, family, group, community, and/or Recognize that there can great variations within cultural
organizations with whom they interact. groups as to the preferred terminology use to refer to their
Psychologists also seek to understand how they sociocultural group
bring their own language and communication to Recognize that the meaning of labels and terminology
these interactions referring to sociocultural groups can change over time
Guideline 4. Psychologists endeavor to be aware of Consider the influence that resource-rich and resource-poor
the role of the social and physical environment in environments can have on a person when interpreting test
the lives of clients, students, research participants, results
and/or consultees
Guideline 5. Psychologists aspire to recognize and Recognize the influence that the experience of disparities in
understand historical and contemporary opportunities can have on a person’s psychological
experiences with power, privilege, and oppression. functioning
As such, they seek to address institutional barriers Consider how experienced racism can influence a person’s
and related inequities, disproportionalities, and psychological functioning
disparities of law enforcement, administration of Consider how experienced racism could influence a client’s
criminal justice, educational, mental health, and attitudes and behaviors during the assessment process
other systems as they seek to promote justice, Consider the influence of acculturation (i.e., the degree of
human rights, and access to quality and equitable identification with the dominant culture) when interpreting
mental and behavioral health services test results
Guideline 6. Psychologists seek to promote Consider how fear and distrust of institutions could influence
culturally adaptive interventions and advocacy mental healthcare utilization
within and across systems, including prevention, Identify and advocate for community services that are
early intervention, and recovery culturally adaptive
Guideline 7. Psychologists endeavor to examine Engage in dialog with practitioners from different countries
the profession’s assumptions and practices within to enhance a global perspective
an international context, whether domestically or Recognize one’s own cultural and political perspectives and
internationally based, and consider how this avoid imposing them on clients of different nationalities
globalization has an impact on the psychologist’s Consider the impact of mass trauma (e.g., natural disaster,
self-definition, purpose, role, and function attempted genocide, economic crises) on persons’
psychological functioning
(continued)
74 4 Ethics in Testing: Rights, Responsibilities, and Fairness
Table 4.2 (continued)
Some key implications for the clinical assessment of child and
Guideline adolescent personality and behavior
Guideline 8. Psychologists seek awareness and Recognize that a persons’ cultural identity can change over
understanding of how developmental stages and their life span
life transitions intersect with the larger Consider how developmental changes in cultural identity
biosociocultural context, how identity evolves as a may be influencing the results of psychological testing
function of such intersections, and how these Recognize that a person’s cultural identity and its influence
different socialization and maturation experiences on adjustment can be influenced by specific historical events
influence worldview and identity Recognize that stressors associated with cultural identity may
vary depending on the stage of development
Guideline 9. Psychologists strive to conduct Recognize that assessment tools often have limited validation
culturally appropriate and informed research, across diverse ethnic groups and, thus, have the potential to
teaching, supervision, consultation, assessment, mischaracterize and miss mental health needs of minority
interpretation, diagnosis, dissemination, and sociocultural groups
evaluation of efficacy Carefully consider whether tests that are administered are
free from measurement bias for the sociocultural group of the
person being assessed
Monitor effectiveness of services in a way that can detect
potential differences in their acceptability and efficacy across
sociocultural groups
Guideline 10. Psychologists actively strive to take Interpret psychological tests in light of both risk and
a strength-based approach when working with resilience factors that are evident in a person’s sociocultural
individuals, families, groups, communities, and context
organizations that seeks to build resilience and Recognize that resilience may be defined and expressed in
decrease trauma within the sociocultural context distinct ways across sociocultural contexts
Recognize that behavior thought to reflect resilience in one
cultural context may be considered undesirable in another
children and adolescents. Within this basic fied goals and objectives. As discussed in the
framework, however, evaluations must be tai- chapter on psychometrics, we noted that it was
lored to the needs of the individual case. The erroneous to think in terms of an assessment
critical developmental issues, the most relevant technique or battery being valid or invalid.
areas of adjustment to assess, and the most Results of the evaluation can be valid for specific
important elements of a child or adolescent’s interpretations. Therefore, what interpretations
environment will all vary from case to case. As a one anticipates making at the end of the evalua-
result, it is inappropriate to develop specific tion should guide the selection of tests for the
guidelines for designing evaluations. Instead, in assessment battery. For example, if an assess-
this section, we attempt to provide a framework ment is primarily intended to determine school
for designing assessments that can be tailored to placement, then the focus of the evaluation will
most assessment situations. not be to determine whether or not a psychiatric
diagnosis is warranted but to determine whether
or not the child meets the eligibility requirements
Planning the Evaluation of the school system in which the child is
enrolled. The assessor may feel that more infor-
Clarifying the Referral Question mation is needed to make appropriate recommen-
dations to meet a child’s psychological and
A crucial part of planning any evaluation is hav- educational needs than is required by these crite-
ing enough information prior to beginning testing ria. However, enough information to determine
to make at least some initial decisions on the eligibility should be part of the assessment, if this
structure and content of the assessment process. is the primary referral question. In our experi-
This is not to say that the assessment process ence, it is not uncommon for an otherwise sound
should be so structured from the outset that and competently conducted evaluation to be use-
changes after testing is underway are not possi- less for the specific purposes for which a child
ble. However, obtaining crucial information was referred.
before the first testing session enhances the likeli- There are many other examples of how the
hood that one will provide a focused and appro- intended use of the assessment information will
priate assessment. Almost every testing agency, determine which measures should be used. This
whether clinic, school, hospital, or private practi- may be as broad as defining what areas need to be
tioner, has some established intake process that covered for a certain purpose (e.g., some residen-
provides preliminary information on the child or tial treatment centers require a personality assess-
adolescent being tested. There is no single best ment prior to acceptance) to as specific as
way to structure the intake procedure. However, requiring certain tests (e.g., some school systems
there are some pieces of information that should require specific tests to be given for special edu-
be obtained routinely in any intake process, in cation placement). The assessor should not give a
addition to any business information (e.g., name, test that, in his or her professional judgment, is
address, phone number, insurance coverage) that inappropriate for a particular use or is inappropri-
is required by the agency. ate for a particular client. However, if, at the time
of referral, the intended use of the assessment is
clarified, and there is some question as to how
Clarifying the Purpose of Testing appropriate certain requirements are for a given
case, the assessor can attempt to address these
The most important piece of intake information issues before beginning the evaluation.
for planning an evaluation is the intended pur- Often the person or agency referring a child or
pose of the evaluation. A major flaw in many adolescent for testing is not sure how the test
clinical assessments is a lack of focus. From the results will be used. Instead, the child is referred
outset, an evaluation should have clearly speci- because the agency is unsure of the nature of a
Planning the Evaluation 77
assessments of children’s emotional and behav- assessed. Because this is so important, a signifi-
ioral adjustment need to be sensitive to a number cant focus of later chapters (which provide
of important developmental issues. First, a basic reviews of specific testing instruments) is a
tenet of developmental psychopathology is the description of the instruments’ norm-referenced
importance of taking a “process-oriented” scores. It is evident from these reviews that the
approach to conceptualizing children’s adjust- adequacy of these scores can vary across devel-
ment. As a result, it is important that clinical opmental stages (e.g., having a very limited nor-
assessments not only involve a standardized and mative sample for older adolescents). In addition
comprehensive assessment of a child’s behav- to specific tests, some testing modalities may be
ioral and emotional adjustment, but it is also more or less appropriate depending on the devel-
important that they include an assessment of the opmental level of the child. For example, in the
developmental processes (e.g., temperamental chapter on structured interviews, we discuss
tendencies, family context) that may be related to research suggesting that the traditional child self-
the child’s current pattern of adjustment. This report format on these interviews may be unreli-
more comprehensive assessment requires that able before age 9.
assessors maintain a current knowledge of the
research related to the type of problems they
encounter in their evaluations so that they have an Determining the Relevant
adequate understanding of the processes that may Psychological Domains to Assess
be involved in the development of these
problems. A fairly ubiquitous finding in research on child-
This research provides the assessor with some hood psychopathology is the high degree of over-
initial hypotheses regarding what might be lead- lap or comorbidity in problem behaviors (Jensen,
ing to or maintaining the problems that led to a 2003). That is, children with problems in one
child’s referral for testing. Measures are selected area of emotional or behavioral functioning are at
so that these hypotheses can then be tested in the high risk for having problems in other areas of
evaluation. For example, there is research to sug- emotional and behavioral functioning, as well as
gest that there are distinct subgroups of children problems in social and cognitive arenas. In addi-
with conduct problems, some of whom react very tion, a key assumption of a developmental
strongly to peer provocation and emotional stim- approach to understanding children’s adjustment
uli and some of whom show a lack of reactivity to is that all outcomes are influenced by multiple
emotional cues, leading them to ignore the poten- interacting processes. As a result, most evalua-
tial consequences of their behavior on others tions of children and adolescents must be fairly
(Frick, 2012). Solely assessing the child’s level comprehensive to ensure that all areas that could
and severity of conduct problems and making a be relevant to treatment planning are assessed. In
diagnosis of “Conduct Disorder,” without assess- planning an evaluation, one should consider the
ing the child’s affective and interpersonal style, most likely comorbidities associated with the
would not allow one to distinguish between these referral problem and the most likely factors that
different subgroups who may require different can lead to such problems and design the evalua-
approaches to treatment (Frick, 2012). tion to provide an adequate assessment of these
Second, it is important for the psychologist to areas. From the referral information, one may
consider the developmental stage of the child to also gather some clues as to how intensive the
be assessed in designing an assessment battery. assessment of these potentially important
For example, it is important, when selecting tests domains should be.
for a battery, to determine whether the tests pro- For example, consider a referral of a 7-year-
vide good norm-referenced scores for the devel- old boy who is having significant problems of
opmental stage of the child or adolescent being being organized, acts very impulsively, and is
Designing the Evaluation 81
always leaving his seat at school. Your initial this text (Chap. 12) is devoted to the assessment
hypothesis may be that the child has attention- of a child’s family environment. However, what
deficit hyperactivity disorder (ADHD), and you constitutes a family for a child is becoming
will design an evaluation to test this hypothesis increasingly diverse, and the intake can yield
and to test the many different processes (e.g., some preliminary information on the family
poor executive functioning) that could lead to structure (e.g., marital status of parents, degree of
this disorder (see Chap. 17). In addition, you contact with nonresident parents, other adult
know from research on ADHD that approxi- caretakers in the home) that provides the assessor
mately 30% of children with the disorder have with some clues as to the best method of structur-
a co-occurring learning disability (Kimonis, ing an evaluation of the family context.
Frick, & McMahon, 2014). Therefore, you con-
sider how you will assess for this likely comor-
bidity. However, in the initial intake, the child’s Other Practical Considerations
mother states that her son has no real problems
academically, other than losing his assignments In an important clinical endeavor like psycho-
frequently, and, in fact, he has only made two logical testing that can have important conse-
Bs on his report cards since entering school. quences for a child, one does not like to consider
Based on this piece of information, you may such mundane factors as time and expense in
decide not to conduct an intensive evaluation of designing the evaluation. Clearly, these factors
a potential learning disability unless, during the should not outweigh what is in the best interest of
course of the evaluation, you discover other the child being tested. However, sometimes these
evidence of learning problems. factors are unavoidable, and often expediency is
in the best interest of the child. For example, an
adolescent who has an impending court date for a
creening Important Psychosocial
S juvenile offense may need to have an evaluation
Contexts completed before that date to help in determining
the most appropriate placement and the most
Research has indicated that children’s behavior is appropriate services. One would take care to not
heavily influenced by factors in their psychoso- be so influenced by expediency that treatment
cial environment. Therefore, an important con- decisions are misguided by poor assessment
sideration in planning an evaluation is determining results. But one must consider what can mean-
what aspects of a child’s environment should be ingfully be obtained within the time frame avail-
assessed (e.g., teaching styles of specific teach- able and possibly make as part of the evaluation a
ers, affective tone of family interactions) and how recommendation for what additional testing
these should be assessed (e.g., naturalistic obser- might be beneficial as time allows.
vations, behavior rating scales). However, which How much to weigh cost and time constraints
contexts are most relevant will vary from child to will vary from case to case. However, we feel that
child. The intake information should provide one should ask two questions in designing any
enough information so that an evaluation can be evaluation:
planned in which (1) informants from each of a
child’s relevant contexts provide information on 1. What is the essential information needed to
a child’s functioning and (2) the contexts that answer the referral question(s)?
seem to have the most impact on a child’s func- 2. What is the most economical means of obtain-
tioning can be assessed in greater detail. ing this essential information without com-
One of the most influential contexts for the promising the usefulness of information?
majority of children is the family. A chapter in
82 5 Planning the Evaluation and Rapport Building
good at, what things you might not be so testing, adolescents may be asked many
good at. To do this, we are going to do a lot
of different things together. First I am going personal questions. They must be warned
to ask you to do some reading and math of these questions and informed as to how
problems with me. Then I will ask you to the information from the testing will be
fill out some questionnaires that will tell me conveyed to other people. This is very
how you feel about different things, how
you get along with kids in your class, and threatening to most adolescents, and the
how you get along with people in your fam- explanation should be sensitive to this
ily. Finally, I am going to show you some issue. Fourth, the majority of adolescents
pictures and ask you to tell me some stories referred for testing do not see the need for
about them. Before we start each of these
activities, I will tell you what we’re going to such testing and don’t want to be there. A
do and how to do each thing. I promise to major flaw we often see in presenting test-
give you a chance to ask me any questions ing to adolescents is that the assessor tries
you have about each task. I have already to cajole the adolescent into being happy to
talked to your mother about how things go
at home and I am going to ask your teacher be there and into appreciating the potential
to fill out a questionnaire about how she benefits of testing. Clearly, the potential
sees you at school. After I get all the infor- benefits of testing should be discussed with
mation, I should understand you a little bet- the adolescent in an attempt to enhance
ter and I will then talk about what I found
with you and your parents. Alanna, it is very motivation. However, this approach often
important that you understand that I’m not has a minimal effect on motivation, and
looking for things that are wrong with you. often one must simply acknowledge to the
My guess is that you are like most kids. You adolescent that you understand that he or
have things that you’re good at, some things
that you’re not so good at, and that there are she is not happy about being there, but if
things you like and other things you don’t you work together, you will get through it
like. I am just trying to get a good picture of quickly and relatively painlessly. The fol-
all these different parts of you. lowing is a sample explanation of psycho-
logical testing provided to a 16-year-old
male:
Jeff, I want to explain what we will be doing
Box 5.5 Explaining Testing to an Adolescent today and, please, feel free to ask me any
questions about what I say. You probably
There are several crucial issues that one know that your parents are concerned about
your behavior. They have seen some
must keep in mind when explaining testing changes in you recently and they want to
to an adolescent. First, adolescents spend a know if they can be doing anything more to
great deal of energy trying to convince peo- help you. I understand that you are not wild
ple that they are no longer children. about being here, but if we work together,
maybe we can see if there is anything that I
Therefore, one must be very careful not to can recommend to help you or at least put
come across as condescending to them. your parents’ minds at ease. I work with a
Second, because of the importance of peers lot of people your age who don’t want to be
in adolescence, adolescents are very con- here at first, but end up getting a lot out of
the experience. I will start out just asking
cerned with fitting in. Coming in for psy- you about some of the things that have been
chological testing may be viewed as a going on with you lately to get your view on
threat to this by making them feel different things. I have already talked to your parents
from other adolescents. Therefore, the about their views of what’s going on. I also
have some questionnaires for you to com-
explanation should attempt to normalize plete about your feelings, your behaviors,
the testing as much as possible. Third, pri- and your attitudes. Some of these questions
vacy is a major issue for adolescents. In (continued)
86 5 Planning the Evaluation and Rapport Building
Although many evaluations are conducted at the of their child’s adjustment and (2) that the evalu-
request of a parent, there are also many situations ation will be personalized for the individual
in which a child is referred by others (e.g., school, child. If parents are immediately asked to fill out
court), and, in these situations, building rapport rating scales or administered a structured inter-
with the child’s parent is critical. In these circum- view as part of a standard evaluation, they often
stances, the assessor must allow the parent to develop the impression that the assessor is more
express his/her views on the need for the evalua- interested in administering tests than in actually
tion very early on in the testing process. The understanding their child’s needs. As one would
assessor need not necessarily agree with these expect, such an impression is very damaging to
views, but the assessor should convey to the par- the development of rapport. Instead, we typically
ents a sincere interest in understanding their recommend starting clinical assessments with
views in order to build a working relationship parents with the very simple question, “what con-
with them. cerns you most about your child?”. This conveys
Even for parents who have instigated the to the parent that your primary concern is with
referral for testing, the assessor should be aware helping their child and that their opinion of what
of the potential threat to a parent’s self-esteem is going is important and valued.
that many testing situations present. For many
parents, acknowledging that their child might
have any type of disability is quite traumatic and he Basics of Building Rapport
T
can evoke a sense of failure. Also, parents often with Teachers
are struggling with feelings of blame for their
child’s problems and may be concerned that test- It is becoming increasingly clear that evaluations
ing will confirm their potential role in their of children must involve information from teach-
child’s difficulties. An assessor should be sensi- ers (Loeber, Green, & Lahey, 1990). The degree
tive to these dynamics and allow the parents to of teachers’ involvement varies considerably
express their concerns at some point during the depending on the focus of the evaluation.
testing. Additionally, the parents should be sup- However, many assessors who are not used to
ported in their role of getting help for their child. working in school settings find themselves ill-
For example, an assessor might tell the parents equipped to collaborate with teachers to conduct
how lucky their child is to have parents who are psychological evaluations (Conoley & Conoley,
willing to go the time, effort, and expense to 1991).
obtain help for him or her and not just let things In the introduction to the concept of rapport,
get worse. This helps to reframe the testing situa- we defined the basic ingredient to rapport build-
tion as one that could increase the parents’ self- ing as exhibiting an attitude of respect toward the
esteem, rather than one that is a threat to their client or informant. Although many psycholo-
self-concept. gists work hard at respecting and developing rap-
Several reasons were given for starting with port with parents and children, often this respect
structured tasks in testing children in an effort to is lost when dealing with other professionals,
enhance rapport. In our experience, the opposite such as teachers. A key way to demonstrate this
is true in rapport building with parents. Even attitude is by respecting the importance of teach-
prior to obtaining specific background informa- ers’ time. Matching phone calls to planning
tion from a parent, it is important to let the parent times, eliminating all but the most essential work
discuss his or her concerns about the child in an for the teacher, and always personally thanking
unstructured format. The unstructured clinical the teacher for his or her efforts in the evaluation
interview is discussed in more detail in a later are very simple, yet important, rapport-building
chapter. However, placing such an interview at strategies.
the start of the evaluation conveys to the parent If a teacher is to be sent a portion of the assess-
(1) a genuine concern with his or her perceptions ment to complete (e.g., rating scales), it is impor-
Conclusions 89
tant for the assessor to call the teacher and this chapter, we have tried to highlight some of
personally request the teachers’ participation in the important issues in rapport building with chil-
the evaluation, acknowledging and thanking the dren and adolescents of various ages. We have
teacher for his or her efforts, rather than simply also tried to make some practical recommenda-
sending the material to the teacher via the child, tions that address these issues.
parent, or mail. Such a call is a professional cour-
tesy that greatly enhances the collaborative effort. Chapter Summary
It sets the tone for the teacher being involved in
the evaluation as a valued professional who has 1. The first step in planning an evaluation is to
much to offer to the assessment of the child. clarify the reason for referral both in terms of
the purpose of testing and the types of behav-
ior that led to the referral.
Conclusions 2. Two important decisions that a clinical asses-
sor should make prior to starting any evalua-
In this chapter some nonspecifics of the clinical tion is whether or not a formal evaluation is
assessment of children were discussed. That is, a warranted and whether he or she is the most
successful evaluation is not simply a matter of appropriate person to conduct the evaluation.
appropriately administering and interpreting psy- 3. In addition to competently administering
chological tests. It is also dependent on an asses- tests, clinical assessors must create the appro-
sor’s ability to provide an appropriate context in priate environment within which the evalua-
which the testing takes place. tion can take place.
The first major issue discussed was the impor- 4. Building rapport with a child refers to efforts
tance of good planning. A good evaluation is at developing a collaborative and supportive
focused and goal-oriented. The purpose of the relationship with the child for the purpose of
evaluation and the intended uses of the assess- conducting the evaluation.
ment results will have a major impact on how the 5. Building rapport with other important people
assessment is structured. Enough information who will be involved in the evaluation (e.g.,
should be available prior to actual testing so that parents, teachers) is also critical to the assess-
the assessor has some initial hypotheses that will ment process.
be tested in the evaluation. 6. A thorough and sensitive informed consent
The second part of the chapter focused on procedure can play a major role in showing
rapport-building strategies with all participants in respect to the child client and his or her par-
the evaluation. Developing a collaborative, ents and thereby can greatly aid in the estab-
respectful, and trusting working relationship is lishment of rapport.
crucial to a successful evaluation. Being able to 7. An explanation of the testing procedures with
develop rapport is a skill that often takes years of a child must be sensitive to a large number of
practical experience to develop fully. However, in motivational and developmental issues.
Self-Report Inventories
6
The eventual user of any of the omnibus inven- attempts to gauge the child’s own perceptions
tories discussed will have to spend considerable and feelings about school, parents, peers, and his
time with the test manuals and additional read- or her own behavioral problems. The estimated
ings and seek supervised interpretation practice. minimum reading level of the SRP is the second
What has become more clear in recent research is grade (Reynolds & Kamphaus, 2015). The SRP
that despite their relative lack of validity for the may also be read to children in order to assure
assessment of very young children, self-report comprehension of the items and should be rou-
inventories can provide invaluable information as tinely considered for children under the age of 9.
to the youth’s perception of his/her functioning It takes approximately 20–30 minutes to admin-
and the factors that ameliorate or exacerbate his/ ister according to the authors. There are also
her problems. Spanish language versions of the child and the
Whether a self-report inventory has a place in adolescent SRP that were developed concurrently
a psychological assessment will depend on many with the English language versions of the SRP
factors, including the client’s developmental (see Reynolds & Kamphaus, 2015) (Table 6.1).
level, presenting problem, and the purpose of the
assessment. As with any measure, there are limi- Scale Content The child and adolescent ver-
sions of the SRP both include four adaptive
tations to the use of self-report scales. Specifically,
one must consider the client’s comprehension of scales. The child version has 11 clinical scales,
items and the potential for response sets includ- with the adolescent version containing the same
ing socially desirable response tendencies. The clinical scales and an additional Sensation
instruments reviewed below vary in the degree to Seeking scale. As with its predecessors, the scales
which they have appeared to take these factors were developed using a combination of rational,
into account. Therefore, the burden rests with the theoretical, and empirical approaches to test
development. The BASC-3-SRP continues the
clinician to select tools that will answer the refer-
ral question in a legitimate, comprehensive, and mixed response format initiated with the BASC-2
cost-effective manner. that includes both True-False items and
frequency-based items (i.e., Never, Sometimes,
Often, Almost Always). Scales on the BASC-3
Behavior Assessment System stem from the previous editions and were devel-
for Children Self-Report oped based on definitions of key clinical con-
of Personality (BASC-3-SRP; Reynolds structs for children and adolescents. Items were
& Kamphaus, 2015) developed based on these definitions, and covari-
ance structure analysis was used to enhance the
The Behavior Assessment System for Children, homogeneity of scale content while maintain the
3rd Edition-Self-Report of Personality (BASC-3- theoretical meaning of the respective scales
SRP; Reynolds & Kamphaus, 2015) is the most (Reynolds & Kamphaus, 2015). The SRP
recent iteration of the BASC family of rating includes scales that are typical of self-report
scales. The BASC-3 continues to rely on the measures (e.g., anxiety, depression) as well as
assessment approach utilized by its predecessors. some that are relatively unique (e.g., sense of
The BASC-3, like the BASC-2, includes a inadequacy, self-reliance, locus of control; see
college-age self-report form and a 14-item inter- scale definitions in Table 6.2). The BASC-3 sys-
view format to be used with 6- and 7-year-olds tem also includes critical items that may warrant
that consists of yes/no and open-ended questions. further follow-up (e.g., “threatens to hurt others,”
The focus of the below discussion will be on the “Says ‘I want to kill myself’”).
child (ages 8–11) and adolescent (ages 12–21)
forms. Five composites were constructed for the SRP
As with its previous versions, the BASC-3- using factor analysis: Emotional Symptoms
SRP is the component of the BASC-3 that Index (ESI), Inattention/Hyperactivity,
Table 6.1 Overview of self-report inventories
Reading Time to Norm
Inventory Ages Content level administer sample Reliability Validity MI
Omnibus Inventories
Behavior Assessment System 8–11; 137–189 items; emotional symptoms 2nd 20– E Internal consistency Good relations with E
for Children, 3rd Edition 12–21; index, inattention/hyperactivity, grade 30 minutes (good); test-retest similar tools; support
Self-Report of Personality 18–25 internalizing problems, personal across “several for factors
(BASC-3 SRP; Reynolds & adjustment school problems, weeks” (good)
Kamphaus, 2015) functional impairment; 3 validity
scales; 12 clinical scales, 4 adaptive
scales, 4 optional content scales
Achenbach Youth Self-report 11–18 112 items; internalizing, 5th 15– G Internal consistency Good validity E
(YSR; Achenbach & Rescorla, externalizing, competence; 8 scales grade 20 minutes (good); 7 month (clinical vs.
2001) test-retest non-clinical groups)
(moderate);
interrater (low to
moderate)
Minnesota Multiphasic 14–18 241 items; 9 clinical scales; 6 5th–6th 90 minutes G Internal Item overlap detracts F
Personality Inventory- validity scales; 25 specific problems grade consistency; from content validity;
Adolescent-Restructured Form scales; 5 personality test-retest; content scores do not match 4
(MMPI-A-RF; Archer, 2017) psychopathology scales scales better factor solution found
reliability than in research
clinical scales
Conners 3rd Edition, Self- 8–18 59 items; 39-item short form; 3 3rd 20 minutes E Internal consistency Good evidence of Ea
report (Conners-3; Conners, validity scales, 6 empirical scales, 1 grade and test-retest criterion-related and
Conners, 2008a) rational scale, 5 DSM-IV-TR (good) discriminant validity
symptom scales
Personality Assessment 12–18 264 items, including 17 critical 4th 30– Internal consistency Good relations with G
Inventory for items; 4 validity scales, 11 clinical grade 45 minutes and test-retest similar self-report
Adolescents(PAI-A; Morey, scales, 5 treatment consideration (good) tools
2007) scales, 2 interpersonal scales
Note: Norm sample evaluated as E Excellent, G Good, MI Degree of enhancement of multi-informant assessment, E Excellent, G Good, F Fair
a
We view the Conners-3 as providing an excellent means of assessment of externalizing problems; however, if an extensive assessment of internalizing or learning difficulties is
desired, the Conners Comprehensive Behavior Rating Scales (CBRS; Conners, 2008b) is recommended
93
94 6 Self-Report Inventories
Table 6.2 BASC-3-SRP scale definitions note: from Internalizing Problems, Personal Adjustment,
Reynolds and Kamphaus (2015)
and School Problems. The Inattention/
Scale Definition Hyperactivity composite includes the Attention
Anxiety Feelings of nervousness, worry, Problems and Hyperactivity scales. The School
and fear; the tendency to be
overwhelmed by problems Maladjustment composite includes the Attitude
Attention Tendency to be easily distracted Toward School and Attitude Toward Teacher
problems and unable to concentrate more scales. The Personal Adjustment composite
than momentarily assesses self-perceived personal strengths and
Attitude to school Feelings of alienation, hostility, includes the four adaptive scales (i.e., Relations
attitude to and dissatisfaction regarding
teachers school with Parents, Interpersonal Relations, Self-
Feelings of resentment and dislike esteem, and Self-reliance). The Internalizing
of teachers; beliefs that teachers composite consists of the remaining six scales
are unfair, uncaring, or overly (i.e., Atypicality, Locus of Control, Social Stress,
demanding
Anxiety, Depression, and Sense of Inadequacy)
Atypicality The tendency toward bizarre
thoughts or other thoughts and The ESI represents the scores for the six scales
behaviors considered “odd” with the highest loadings on an unrotated first
Depression Feelings of unhappiness, sadness, factor (i.e., Social Stress, Anxiety, Depression,
and dejection; a belief that Sense of Inadequacy, Self-esteem reversed
nothing goes right
scored, and Self-reliance reverse scored).
Hyperactivity Tendency to report being overly
active, rushing through work or According to the authors, this index is “the most
activities, and acting without global indicator of serious emotional disturbance,
thinking particularly internalized disorders” (p. 83).
Interpersonal The perception of having good
relations social relationships and
friendships with peers
Administration and Scoring The SRP is avail-
Locus of control The belief that rewards and able in two administration formats: a hard copy
punishments are controlled by form or online through Q-Global, a secure sys-
external events or other people tem. Scoring for the hard copy form can be
Relations with A positive regard for parents and accomplished by hand or online. The SRP has a
parents a feeling of being esteemed by
them
reasonable administration time given that there
Self-esteem Feelings of self-esteem, self- are 137 items on the SRP-C and 187 items on the
respect, and self-acceptance SRP-A. Although efforts were made to keep the
Self-reliance Confidence in one’s ability to item stems brief and at a second grade reading
solve problems; a belief in one’s level, some young people still will have difficulty
personal dependability and
decisiveness
using the SRP. This and other self-report invento-
Sensation The tendency to take risks and ries are not recommended for use with children
seeking seek excitement with significant cognitive deficits. Oral adminis-
Sense of Perceptions of being unsuccessful tration is possible and recommended for children
inadequacy in school, unable to achieve one’s below the age of 9, but even older children with
goals, and generally inadequate
reading comprehension problems may have dif-
Social stress Feelings of stress and tension in
personal relationships; a feeling ficulty understanding some items.
of being excluded from social
activities Validity Scales The BASC-3-SRP, like its pre-
Somatization The tendency to be overly decessors, includes three validity scales as well
sensitive to, experience, or
complain about relatively minor
as two indexes that alert the clinician to response
physical problems and patterns or inconsistency in responses (i.e., the
discomforts Response Pattern Index and Consistency Index,
respectively). The three validity scales are
Omnibus Inventories 95
Norming The BASC-3 SRP was normed on a Reliability The reliability of the SRP scales is
national sample of 300 children ages 8–11 and good as indicated by a variety of methods.
600 adolescents ages 12–18. The size of these Median internal consistency coefficients are gen-
norm groups is smaller than that used for the erally in the 0.80s (see Table 6.3). As shown in
development of the BASC-2-SRP. Equal num- Table 6.3, the lowest internal consistency was for
bers of males and females were included within the Somatization scale, which may make sense in
each of three age groups (i.e., 8–11, 12–14, that this scale consists of physical symptoms
15–18). Based on the occurrence of age group which may not be interrelated. Such was the case
differences on many subscales, these age groups for the BASC-2 SRP as well. Test-retest coeffi-
were used in the development of T-scores (i.e., cients taken at “several week” (Reynolds &
separate T-score distributions were used for Kamphaus, 2015; p. 226) are generally in the
8–11-year-olds from 12–14-year-olds). Reynolds 0.70s and 0.80s. However, coefficients for the
and Kamphaus (2015) report that within each adaptive scales (corrected r = 0.63–0.68) and for
age group, the sample was stratified according to the Sense of Inadequacy subscale (corrected
parent education level, race/ethnicity, and geo- r = 0.59) were somewhat lower in the child
graphic region of the United States based on data version. In the adolescent version, test-retest
from the 2013 US Census Bureau American coefficients were all 0.78 or higher for clinical
Community Survey (US Census Bureau, 2013). and adaptive subscales.
Omnibus Inventories 97
Table 6.3 BASC -3-SRP median internal Consistency Table 6.4 BASC -2-SRP composites and scale
members
Scale Coefficient
Anxiety 0.87 School maladjustment
Attention problems 0.86 Attitude to school
Attitude to school 0.82 Attitude to teachers
Attitude to teachers 0.83 Sensation seeking (adolescent only)
Atypicality 0.84 Internalizing problems
Depression 0.86 Anxiety
Hyperactivity 0.84 Depression
Interpersonal relations 0.81 Locus of control
Locus of control 0.78 Sense of inadequacy
Relations with parents 0.90 Somatization
Self-esteem 0.82 Social stress
Self-reliance 0.82 Personal adjustment
Sensation seeking 0.84 Relations with parents interpersonal relations
Sense of inadequacy 0.82 self-esteem
Social stress 0.87 Self-reliance
Somatization 0.75 Inattention/hyperactivity
Composites 0.90 Attention problems
School problems Hyperactivity
Inattention/hyperactivity 0.91
Internalizing problems 0.96
Personal adjustment 0.93 (RCMAS-2), Beck Youth Inventories-II, Beck
Emotional symptoms index 0.95 Depression Inventory-II, and Minnesota
Multiphasic Personality Inventory-Adolescent
Validity The BASC-3 manual provides an (MMPI-A).
extensive report of SRP exploratory and confir- In general, the SRP scales correlated moder-
matory factor analyses. The factor structure is ately to highly (i.e., r ≥ 0.50) with analogous
shown in Table 6.4 and mimics that reported for scales from the Achenbach Youth Self-report.
the BASC-2. The only exception is that the Similarly, scales (mainly pertaining to inattention
Somatization subscale did not load onto a higher- and hyperactivity) on the Conners-3 Self-Report
order factor in the child SRP. The school malad- were correlated 0.48–0.50 with their BASC-3
justment factor remains a relatively unique SRP counterparts. Interestingly, scores on the
contribution of the BASC self-report. The per- DREF-SRF showed moderate to high correlations
sonal adjustment factor is also unique in that it on all subscales of the BASC-3 SRP with the
provides a multidimensional assessment of adap- exception of Sensation Seeking and Self-esteem.
tation or potential strengths with an emphasis on Self- reports on the CDI-2 scales were correlated
self-perception and relationships with others. 0.29 (Negative Self-esteem) to 0.61 (Functional
The implications of this composite for long-term Problems) with the scores on the Depression scale
prognosis or overlap with common clinical con- of the child version of the SRP. The Anxiety and
cerns are unclear, however. Depression scales of the BASC-3 SRP were cor-
related 0.51–0.58 with their analogous scales on
The criterion-related validity of the SRP was the Beck Youth Inventories-II, and the Anxiety
evaluated by correlating it with the Achenbach scale of the SRP was correlated 0.42–0.69 with
Youth Self-Report, the Conners-3 Self-Report RCMAS-2 scale. In light of the breadth of the
Scale, the Dells Rating of Executive Function MMPI-A, there is a wide range of convergence
Self Rating Form (DREF-SRF; adolescent), between MMPI-A and SRP scales. Correlations
Children’s Depression Inventory-2 (CDI-2), between similar subscales are 0.50 and higher
Revised Children’s Manifest Anxiety Scale-2 (see Reynolds & Kamphaus, 2015).
98 6 Self-Report Inventories
As a whole, these findings indicate suitable ficulty in interpreting these composites given
convergent validity for BASC-3 SRP scales but their varied content (see Table 6.4), until further
also highlight a need for attention to differences research studies are available, primary SRP
in scale content in interpreting findings from interpretation should focus on the scale level.
each of these tools. The scales of the BASC-3 and its predecessors
In the norm sample, scores on BASC-3 scales are relatively straightforward to interpret
generally aligned closely with their analogous because the scale contents have some rational,
BASC-2 scales in terms of T-scores. The theoretical, and research basis with the caveat
12–14-year-old age group demonstrated the most that the clinician should still consider which
deviation from BASC-2 scores (i.e., up to items led to a client’s scale elevation rather than
4 T-score units) with the scores usually higher on relying on the name of the scale for
the clinical scales and lower on the adaptive interpretation.
scales on average on the BASC-3. Reynolds and
Kamphaus (2015) note that on the BASC-3, this Many items from the BASC-2 were retained
particular age group may appear to be f unctioning or slightly altered for the BASC-3-SRP, after an
worse on the BASC-3 than would have been initial survey of students as to the positive and
apparent on the BASC-2. negative behaviors they observe in others (see
At the time of this publication, the BASC-3 Reynolds & Kamphaus, 2015). This continuity
has not yet been the focus of independent peer- lends some confidence in scale interpretation for
reviewed research on its validity or clinical util- the clinician with experience using the previous
ity. The BASC-3 manual provides initial results version. Item-level factor-analytic results provide
on the correlates of the SRP scales; however, one empirical clue to the meaning of SRP scales.
much more research is needed, particularly with Some of the items with the most substantial fac-
clinical populations to help provide further tor loadings on each scale are identified in
understanding of the information garnered from Table 6.5. These symptoms can be linked to
these scales. background information in order to interpret SRP
results with greater certainty. However, the eleva-
Interpretation Although the composites of the tions of SRP scales may also lead the clinician to
SRP have some factor-analytic evidence to sup- follow-up on issues that might not have been
port their validity, there is little evidence as to raised during the gathering of background
the ultimate utility of the composites. The com- information.
posites may very well provide parsimonious and We propose the following steps in interpreting
clinically relevant information beyond the sub- the SRP (which can be applied to other omnibus
scales, but such studies to determine their utility rating scales as well):
simply have not been conducted yet. In the
BASC-3 manual, the authors provide results 1 . Check validity scales.
indicating that the Internalizing Composite and 2. Check critical items if available.
Emotional Symptoms Index show moderate to 3. Determine which scales are elevated.
strong correlations (0.38–0.80) with virtually all 4. Examine the items that appeared to lead to
scales and composites from the Achenbach scale elevations.
Youth Self-report. The Inattention-Hyperactivity 5. Integrate with ratings from other informants.
composite showed correlations with the ADHD- 6. Integrate with data from other tools (e.g.,
oriented scales on the Achenbach and on the interviews, intelligence testing).
Conners-3 Self-Report Scale (child version) 7. Establish connection between results with
that were r = 0.46 and higher. However, given clinical diagnoses and planned treatment/
the relative lack of research to date and the dif- intervention.
Omnibus Inventories 99
Table 6.5 SRP key symptoms as indicated by items with hamper the validity of interpretations garnered
the highest factor loadings per scale
based on unreliable scales. The reliability esti-
Scale Key symptoms mates shown in Table 6.3 can help a clinician
Anxiety Worried, stressed, feeling anxious, gauge the level of confidence in an obtained
nervousness
score. If, as an example, an adolescent’s scores
Attention Easily distracted, difficulty
problems concentrating, trouble paying are elevated primarily on scales that are less reli-
attention able, the results may be of less import. High
Attitude to Hates school, feels bored in school, scores on the reliable scales may be of greater
school wants to quit school concern. Among the SRP scales that inspire con-
Attitude to Feels that teacher is not proud of
fidence because of their reliability are Anxiety,
teachers him/her, disliking teacher, feeling
untrusted by teacher Depression, Social Stress, Attention Problems,
Atypicality Feels like someone is watching, and Relations with Parents. Scales with some-
hears things that others cannot, sees what lower, but still adequate, reliabilities include
things Locus of Control and Somatization. The applica-
Depression Depressed, feels without friends,
tion of a psychometrically influenced approach to
lonely, sad, feels that no one
understands him/her interpretation of the SRP is presented in the fol-
Hyperactivity Is told to be still, has trouble sitting lowing case study.
still, is told to slow down, has trouble
standing still in lines Strengths and Weaknesses The BASC family
Interpersonal Feels that nobody likes respondent
of rating scales, including its most recent self-
relations (reverse scored), feels that others
hate to be with him/her (reverse report iteration, the BASC-3-SRP, has numerous
scored), is made fun of by others strengths that make it a viable option for use with
(reverse scored), feels liked children and adolescents. Notable strengths
Locus of People get mad at respondent for no include:
control reason, blamed for things he/she
didn’t do
Relations with Parents like to be with him/her, feels
parents proud of parents, parents are easy to 1. A broader age range that is typically available
talk to for omnibus inventories designed for children.
Self-esteem Happy with self, confident, likes the Unique scales that are relevant to the milieus
way he or she looks
of children, such as attitude toward teachers
Self-reliance Reliable, dependable, can be counted
on, good at making decisions and schooling and parent-child relations.
Sensation Excited by dangerous things, likes to 2. A normative sample that is well-described and
seeking be dared to do something, likes to seems appropriately reflective of the US pop-
take risks ulation as of 2013.
Sense of Fails even when trying hard, wants to 3. Good reliability estimates, including some at
inadequacy do be better but is unable
Social stress Feels ignored by others, feels out of
the scale level that are improved relative to the
place around others, feels left out BASC-2 SRP (see Reynolds & Kamphaus,
Somatization Complains of upset stomach, feels 2015).
sick 4. Ease of administration and scoring.
5. Items account for heterogeneity of behaviors
and symptoms within domains.
6. Availability of a range of derived scores and
Some caution is needed in keeping a focus on norms for general, clinical, and gender-
interpretation at the scale level. In particular, referenced samples.
scales may have intuitive appeal and some rele- 7. Inclusion of validity scales that are intuitive
vant content, but questionable reliability would based on structure and content.
100 6 Self-Report Inventories
8. A clear link between the SRP and teacher and used as a self-report inventory, whereas the Adult
parent rating scales (BASC-3-TRS and Behavior Checklist (ABCL) provides ratings of
BASC-2-PRS) which enhances its use in one’s partner. Many of the domains assessed
multi-informant assessment. (e.g., internalizing problems) are similar to those
from the YSR with noteworthy differences such
To date, notable weaknesses of the BASC-3- as scales for substance use. In light of the devel-
SRP are: opmental periods of focus in this text, the adult
self-report scales of the ASEBA will not be dis-
1. Lack of independent research regarding the cussed in detail here, but an introduction may be
BASC-3. found at http://www.aseba.org/adults.html.
2. A smaller norm sample than previous versions
of the BASC. Scale Content Composite scores reflecting
3. Lack of case studies in the manual. externalizing and internalizing dimensions and a
4. Lack of validity evidence for the validity total composite are offered. The following clini-
scales and cutoffs used to indicate caution in cal scales contribute to these composites.
interpreting a response protocol.
5. As with previous versions, an excessive num- Withdrawn/Depressed – preferring to be alone,
ber of Self-Esteem scale items that deal with shy, sulks, sad, lacking energy, etc.
self-perceptions of personal appearance. In Somatic Complaints – nausea, headaches, dizzi-
addition, Self-Reliance items focus on ness, etc.
dependability to others rather than indepen- Anxious/Depressed – crying, fears, nervous, sui-
dence/autonomy. cidal ideation, etc.
Rule-Breaking Behavior – lying, substance
abuse, truancy, stealing, etc.
Achenbach System of Empirically Aggressive Behavior – teasing others, arguing,
Based Assessment Youth Self-Report fighting, destruction of property, etc.
(YSR; Achenbach & Rescorla, 2001)
Included in the Total Composite but not the
The Youth Self-Report (SRP; Achenbach & Internalizing or Externalizing composites are the
Rescorla, 2001) is one component of the larger following:
set of assessment instruments offered within the
Achenbach System of Empirically Based Social Problems – jealous of others, teased by
Assessment (ASEBA) that includes a parent rat- others, clumsy, etc.
ing scale, a teacher rating scale, an observation Thought Problems – strange behaviors, hoarding
scale, and a number of other measures. The YSR objects, sleeping less, hallucinatory experi-
is designed for ages 11 through 18 and obtains ences, etc.
adolescents’ reports about their own competen- Attention Problems – failing to finish assign-
cies and problems in a format similar to that of ments, immature, impulsivity, daydreaming,
the parent Child Behavior Checklist (CBCL) and etc.
Teacher Rating Form (TRF) from the ASEBA,
discussed in the next chapter. The rating scales Scales referred to as Social Competence are
within the ASEBA have considerable content also included that assess participation in a variety
overlap, which can be both an advantage and dis- of activities (e.g., sports) and social interactions
advantage of this system, depending on what the (e.g., friendships). A Social Competence com-
clinician is seeking from behavioral rating scales. posite is also offered.
Additional forms cover adults from ages 18 to The clinical scales are empirically derived via
59. Specifically, the Adult Self-Report (ASR) is factor analysis, and the competence scales are
Omnibus Inventories 101
rationally derived. Critical items (e.g., harming The sample included 1057 youth, 52% of whom
self, setting fires, etc.) are also available on the were male. Children who had reports of mental
YSR. In addition, six DSM-Oriented scales are health, substance abuse, or special education
available for the YSR. These scales were formed services were excluded, thus making this stan-
based on psychiatrists’ impressions of items (see dardization sample a “normal” sample, rather
Achenbach, Dumenci, & Rescorla, 2003) that than a “normative” sample. Most participants
theoretically map on to the DSM-related domains (i.e., 53%) were from a middle SES background,
being assessed (i.e., Affective Problems, Anxiety whereas 16% were from lower SES homes. The
Problems, Somatic Problems, Attention/ sample was 60% White, 20% African American,
Hyperactivity Problems, Oppositional Defiant 8% Latino, and 11% Mixed or Other. From
Problems, Conduct Problems). The DSM- these statistics, both African Americans and
Oriented scales of the ASEBA rating scales have individuals identifying as Latino(a) appear to be
been updated to reflect DSM-5. Research has underrepresented (United State Census Bureau,
supported the structure of the syndrome scales of 2001). Approximately 40% of participants were
the YSR across over 20 cultures (Ivanova et al., from the southern part of the United States with
2007) and convergence with indicators of behav- the Northeast, Midwest, and West each being
ioral and emotional dysregulation across 34 cul- represented by approximately 20% of the sam-
tures (Jordan, Rescorla, Althoff, & Achenbach, ple participants (see Achenbach & Rescorla,
2016). 2001).
Administration and Scoring The YSR is The derived T-scores for the YSR are normal-
designed to be self-administered and requires ized, which results in changing the shape of the
approximately 15–20 minutes (Achenbach & raw score distribution (i.e., reducing skewness).
Rescorla, 2001). The YSR uses a three choice Furthermore, the T-score distributions are trun-
response format: Not True, Somewhat/Sometimes cated, which limits the range of low scores on the
True, and Very True or Often True. Some items clinical scales and high scores on the competence
also provide space for additional information, scales. As a result, T-scores for the clinical scales
such as Describe: following “I store up too many are not recommended to be interpreted below a
things I don’t need.” The scale also ends with the value of 50. The transformation to reduce skew-
item, “Please describe any other concerns you ness and truncated score range both serve to
have.” make the T-score distribution for the YSR differ-
ent from original sample results. The intent of
Computer scoring is available, and an integra- this approach is to aid in interpretation of
tive computer program can be used to compare strengths and difficulties across domains.
results for several raters (e.g., a parent, two teach- However, this lack of reflection of sample charac-
ers, and the YSR). This option is discussed in teristics in the T-scores makes them of dubious
great detail by Achenbach and Rescorla (2001). value for research purposes in particular. For
The level of comparability facilitates the study of most research questions, raw scores would likely
interrater agreement in clinical or research set- be more appropriate than normalized T-scores.
tings and aids in clinical interpretation of con- These T-score characteristics also render some
verging evidence and discrepancies in reports of clinical questions unanswerable, such as “How
the child’s functioning. well-developed are his competencies?” or “Does
she truly have fewer problems than other children
Norming The design of the YSR norming sam- her age?” It should also be noted that unlike the
ple attempted to mimic the national population teacher- and parent-report measures of the
of school children for ages 11 through 18 in Achenbach system, the YSR does not include
terms of SES, geographic region, and ethnicity. gender-specific norms.
102 6 Self-Report Inventories
Reliability Internal consistency estimates are report forms of the Achenbach System (see Chap.
reasonable for the clinical scales falling between 7). Gomez and colleagues noted that the YSR
0.71 and 0.89. The internal consistencies of the tended to converge better with parent than with
composites are all above 0.90. Internal consis- teacher ratings. In addition, little divergence
tency estimates are somewhat lower for the com- between Rule-Breaking and Aggressive Behavior
petence scales (see Achenbach & Rescorla, were noted. Overall, the authors concluded that
2001). considering ratings across informants using the
Achenbach System would aid in diagnoses rele-
Short-term (i.e., approximately 1-week inter- vant to subscales (e.g., Anxious/Depressed and
val) test-retest coefficients are generally good, Withdrawn/Depressed scales in the diagnosis of
with only the Withdrawn/Depressed scale having anxiety or depressive disorders; Gomez, Vance,
a coefficient below 0.70. Seven-month test-retest & Gomez, 2014).
coefficients were adequate with coefficients gen- A predictive validity study of the YSR with a
erally in the 0.50 range. Coefficients for the sample of 2431 youth from Australia found that
Withdrawn/Depressed scale and Somatic scores on YSR DSM-Oriented internalizing
Complaints scale were somewhat lower. scales at age 14 were predictive of depression at
age 21 (Dingle, Alati, Williams, Najman, & Bor,
Validity The YSR manual does not report evi- 2010). Longitudinal studies such as this one,
dence of criterion-related validity, particularly in although relatively rare, speak to the practical
regard to the correspondence between the YSR utility of the YSR and perhaps other youth self-
and other measures of emotional and behavioral report measures. These tools may provide impor-
functioning. However, such information is avail- tant information for future or enduring risk of
able in conjunction with the publication of the psychopathology.
BASC-3-SRP, with YSR and analogous SRP sub-
scales demonstrating good correspondence Interpretation The Achenbach manual pro-
(Reynolds & Kamphaus, 2015). vides some case studies, yet it does not provide
interpretive guidance. This omission is remedied
Some differential validity data are presented, to some extent by the existence of articles on the
with the scales of the YSR consistently differen- YSR and by its amenability to interpretive
tiating between clinic-referred and non-referred approaches such as that described earlier in this
youth. Achenbach and Rescorla (2001) indicate chapter.
that differential validity is the driving force
behind content selection for the current YSR and A strength of the Achenbach approach to
its predecessors. The most recent YSR has six scale construction is the ease with which inter-
items that differ from the items in the previous pretations across informants because of the
YSR. Generally speaking, the validity evidence close item correspondence on the different
reported in the manual concerning the YSR is forms. As mentioned above, Achenbach and
inadequate. However, the YSR has been used Rescorla (2001) detail the approach by which
extensively in research in ways that can provide statistical comparisons across informants may
information regarding criterion-related or con- be made. To the extent that informants agree on
struct validity based on associations between the presence of a problem, the clinician may be
YSR scales and measures of other clinically rel- more confident that a problem warranting atten-
evant constructs. tion exists.
An independent study with clinic-referred However, it is possible to be too heavily influ-
children (Gomez, Vance, & Gomez, 2014) exam- enced by indexes of agreement in the interpretive
ined convergent and divergent validity of the process. For example, one might require agree-
YSR in conjunction with the teacher- and parent- ment across raters to make a diagnostic or other
Omnibus Inventories 103
construct. In this way, the MMPI-A-RF continues RC8, Aberrant Experiences Items on this scale
the MMPI tradition of emphasizing internalizing assess the respondent’s unusual sensory experi-
problems but with an additional emphasis on more ences or thoughts consistent with psychotic
varied and independent subscales. symptoms.
Restructured Clinical Scales Content The RC9 Hypomanic Activation Similar to Clinical
MMPI-A retained the clinical scales of the adult Scale 9 (Hypomania) from the MMPI-2 and
MMPI-2. Numerous volumes are available dis- MMPI-A, this scale assesses excitability and
cussing the structure, content, and correlates of “overactivation, impulsivity, aggression, and
these scales (e.g., Archer, 2005). The below-scale grandiosity” (Handel, 2016; p. 367).
description pertains to the Restructured Scale and
is derived from Handel (2016). Generally speak- Specific Problems Scales In the MMPI-A, con-
ing, the numbered clinical scales of the RF corre- tent scales served as an appealing alternative to
spond to those of the MMPI-A and MMPI-2. The the MMPI clinical scales because they were
exceptions are the Social Introversion Scale (Scale developed using a more rational/theoretical
10) and the outdated Masculinity- Femininity approach to test development (Williams , Butcher,
Scale (Scale 5) from the MMPI-A/MMPI-2 which Ben-Porath, & Graham, 1992) in which scale
do not have an analogous clinical scale in the content considered both empirical factor loadings
MMPI-A-RF. More information on each of these and homogeneity of content within scales.
scales is available from Archer (2017). Moreover, the content scales themselves connote
constructs that are perhaps deemed more clini-
cally relevant in most present-day settings. For
RCd: Demoralization Items on this scale cap- more information regarding the original develop-
ture an adolescent’s general unhappiness and life ment of the MMPI-A content scales, see Williams
dissatisfaction. et al. (1992). The MMPI-A-RF takes a similar
approach through the availability of Specific
RC1, Somatic Complaints These items assess Problem (SP) scales (Archer, 2017). These scales
physical health complaints. are summarized in Table 6.6.
RC2, Low Positive Emotions This scale covers The RF also includes what are referred to as
emotional symptoms connected to depression. the Personality Psychopathology Five (PSY-5)
Scales. These scales are designed to assess major
RC3, Cynicism As stated by Handel (2016), this dimensions of personality pathology and were
scale assesses “non-self-referential beliefs that originally devised from MMPI-A item content
others are bad and not to be trusted” (p. 367). and have since been revised with the MMPI-
A-RF items (Handel, 2016). The PSY-5 scales for
RC4, Antisocial Behavior The Antisocial the RF are referred to as Aggressiveness-Revised,
Behavior scale evaluates self-reported rule- Psychoticism-Revised, Disconstraint-Revised,
breaking and irresponsibility. Negative Emotionality/Neuroticism-Revised,
and Introversion/Low Positive Emotionality-
RC6, Ideas of Persecution Unlike the Cynicism Revised. Limited psychometric information is
Scale, items on this scale involve self-referential available on these scales to date.
thoughts that others are a threat.
Administration and Scoring A major improve-
RC7, Dysfunctional Negative Emotions This ment of the MMPI-A-RF is its reduced length and
scale negative affect, including anger, irritability, concomitant reduced administration time. For
and anxiety. most adolescents, the MMPI-A-RF should be able
to be completed within an hour (Handel, 2016).
Omnibus Inventories 105
Table 6.6 MMPI-A-RF Specific Problems Scales all); however, there is considerable variability
Scale Description on the reading level from item to item (Archer
Malaise (MLS) Sense of poor physical health et al., 2016). When in doubt, an examiner may
Gastrointestinal Nausea, poor appetite, etc. ask the child to read some items aloud to get
complaints (GIC)
some sense of the child’s ability to comprehend
Head pain complaints Headaches, neck pain
(HPC) the item content. The validity checks provide
Neurological Dizziness, loss of balance, another useful alert to possible readability
complaints (NUC) etc. problems.
Cognitive complaints Memory difficulties, problems
(COG) with concentration Validity Scales The MMPI series has a long tra-
Helplessness/ Negative beliefs about goals
dition of the use of validity indexes which contin-
hopelessness (HLP) and problems
Self-doubt (SFD) Lack of confidence, feelings
ues with the MMPI-A-RF. Brief descriptions of
of uselessness the validity scales follow and are taken from the
Obsessions/ Obsessions and compulsive descriptions offered by Handel (2016):
compulsions (OCS) behaviors
Stress/worry (STW) Preoccupation with time VRIN-r (Variable Response Inconsistency-
pressure and disappointments
Revised) The VRIN-r Scale provides an index of
Anxiety (AXY) Anxiety across situations,
nightmares variable responding. This validity check involves
Anger proneness Impatient, easily angered those item pairs that have either similar or oppos-
(ANP) ing item content. The score yielded by the VRIN-r
Behavior-restricting Behavioral inhibitions due to scale reflects the number of item pairs answered
fears (BRF) fears inconsistently. A high score may reveal a careless
Specific fears (SPF) Various specific fears
response style on the part of the client.
Negative school Negative beliefs about school
attitudes (NSA)
Antisocial attitudes Antisocial beliefs TRIN-r (True Response Inconsistency-
(ASA) Revised) This scale reflects the respondent’s
Conduct problems Stealing, behavioral fixed response set. In other words, the TRIN-r
(CNP) difficulties at home and scale consists solely of item pairs with opposite
school
content, and an elevated score may reveal a ten-
Substance abuse Current and past alcohol and
(SUB) drug use dency for the test subject to indiscriminately
Negative peer Negative peer affiliations answer True to the items. Conversely, a low
influence (NPI) TRIN-r score may reveal non-acquiescence.
Aggression (AGG) Physical aggression
Family problems Family conflict CRIN (Combined Response Inconsistency) The
(FML)
CRIN is simply a combined index of the VRIN-r
Interpersonal Unassertiveness and
passivity (IPP) submissiveness in and TRIN-r scales.
interpersonal interactions
Social avoidance Avoidance or lack of F-r (Infrequent Reponses) The F-r scale cap-
(SAV) enjoyment of social events tures the degree to which a respondent endorses
Shyness (SHY) Discomfort or anxiety around uncommon problems or deviancies (based on
others
Disaffiliativeness Dislike of others and of being
responses from the normative sample) such that
(DSF) around them the respondent may be presenting him/herself in
Adapted from Archer et al. (2016); Handel (2016) an unfavorable light.
Checks on the respondent’s reading compre- L-r (Lie) This scale is intended to capture
hension level are necessary. Readability a nalyses attempts by adolescents to put themselves in an
of individual items show adequate readability overly favorable light or report “uncommon vir-
(i.e., approximately fourth to fifth grade over- tues (Handel, 2016; p. 367).
106 6 Self-Report Inventories
her lack of confidence regarding school- Erin views herself as less attractive than
work may be unrealistic. others and does not view herself as having
many strengths.
Attention problems 68
Table 6.7 MMPI-A-RF median internal consistency reliability estimates for the normative sample
Clinical Boys Girls Specific Problems
Scale (N = 805) (N = 805) Scale Boys Girls
RCd 0.80 0.83 MLS 0.54 0.53
RC1 0.77 0.78 GIC 0.59 0.69
RC2 0.60 0.53 HPC 0.51 0.59
RC3 0.60 0.65 NUC 0.59 0.54
RC4 0.80 0.77 COG 0.55 0.61
RC6 0.64 0.66 HLP 0.59 0.61
RC7 0.63 0.66 SFD 0.56 0.60
RC8 0.59 0.55 NFC 0.52 0.53
RC9 0.45 0.52 OCS 0.39 0.45
STW 0.54 0.58
AXY 0.37 0.44
ANP 0.57 0.52
BRF 0.43 0.38
SPF 0.39 0.39
NSA 0.54 0.55
ASA 0.52 0.47
CNP 0.62 0.52
SUB 0.53 0.56
NPI 0.37 0.29
AGG 0.55 0.55
IPP 0.36 0.30
SAV 0.65 0.61
SHY 0.68 0.73
DSF 0.43 v 0.45
NOTE: From Archer et al. (2016)
ance is low as well” (p. 18) and that the validity result. An analogous situation would be to have
evidence concerning a measure sheds more light some intelligence test items included on several
on its ultimate utility. The hierarchy of describing subtests or composites that presumably reflect at
scale reliability, however, is useful in that there is least somewhat different abilities. It is difficult to
a relation between reliability and validity. The imagine, but what if several WISC-V Block
scales with median coefficients below 0.60 are Design items were allowed, because of their cor-
less likely to be the beneficiaries of substantial relations with the Verbal Comprehension sub-
validity evidence. For the MMPI-A-RF, the pre- tests, to be included in calculations of the Verbal
ceding charts suggest that the clinician could Comprehension Index? Such a move would prob-
have more confidence in the information obtained ably be greeted by skepticism, causing clinicians
from the restructured clinical scales than from to wonder about the distinction between the
the specific problems scales. Verbal Comprehension and Visual Spatial
In addition, as is typical for such scales, test- Indexes.
retest reliability coefficients differ from internal
consistency estimates. Test-retest reliability coef- Three higher-order factors were identified in
ficients are generally in the 0.50–0.70 range, with the initial development of the MMPI-A-RF:
one scale (i.e., BRF) demonstrated test-retest Emotional/Internalizing Dysfunction (EID),
reliability less than 0.24 and another (i.e., RCd) Thought Dysfunction (THD), and Behavioral/
having a test-retest coefficient of 0.82. Notably, Externalizing Dysfunction (BXD; Handel, 2016).
the time frame for test-retest analyses is not Limited information on the external correlates or
reported in the MMPI-A-RF manual. validity of these factors is available to date.
Test-retest coefficients are somewhat more However, unlike the MMPI-A, the factor-analytic
difficult to interpret, however, because it is results are used to derive scores that may then be
unclear whether some scales measure traits that interpreted in conjunction with other assessment
theoretically should be stable over at least short data. Additionally, composites labelled
periods of time. Regardless, test-retest data can “Externalizing” (consisting of NSA, ASA, CNP,
be of value when gauging changes from one eval- SUB, NPI, & AGG) and “Interpersonal” (consist-
uation to another. Indeed, instability of a scale ing of FML, IPP, SAV, SHY, DSF) are available.
may be desirable if one wants a measure that is
sensitive to change. Other Validity Evidence Because the develop-
Overall, the reliability estimates for the ment and initial analyses of the restructured form
MMPI-A-RF are more variable and lower than of the MMPI-A were relatively recent, there is
might be expected. Such variability requires a limited independent validity evidence for its
more discerning user who evaluates the reliabil- scales. In terms of convergent and divergent
ity of results on a scale-by-scale basis, which validity, the MMPI-A-RF manual reports the
would not be necessary with more uniform reli- intercorrelations among all of the RF scales over-
ability coefficients. all and separately for males and females. In addi-
tion, correlations between MMPI-A-RF scales
Validity One of the greatest challenges in the and the original MMPI-A are provided. Scales
interpretation of MMPI-A results was the fact were also evaluated in inpatient, residential treat-
that items may load onto multiple scales. The ment, and forensic settings. From these analyses,
MMPI-A-RF sought to reduce the overall length Archer et al. (2016) conclude that the validity
of the scale and, to some extent, the degree to scales, restructured clinical scales, and specific
which items might cross-load onto more than one problems scales function as expected.
scale. However, item overlap across scales
remains and issue with the MMPI-A-RF. The In addition, as summarized by Handel (2016),
result is that scales with common items would be the EID and BXD demonstrated moderate corre-
expected to have stronger intercorrelations as a lations across subscales of the YSR in forensic
110 6 Self-Report Inventories
samples. These broad factors differed relatively Some questions to ask oneself when interpret-
little in their magnitude of association with these ing scale elevations on the MMPI-A/MMPI-
scales. An important exception is that the EID A-RF could include:
factor correlated more strongly (difference of
approximately 0.20–0.30) with internalizing • How does an adolescent get a high score on
scales from the YSR, as would be expected. In a this scale? What are the behaviors, percep-
sample of youth from a residential treatment set- tions, and feelings assessed by this scale (i.e.,
ting, correlations with clinical scales with other the item content)?
available data on emotional and behavioral • How reliable are the scales that I wish to inter-
functioning were very small to moderate at best. pret as being of some clinical value in this
The correlation of the highest magnitude (0.37) case? As noted above, the internal consistency
was between Demoralization (RCd) and suicidal of some scales is quite poor.
ideation. • What does the pattern of scale elevations sug-
The MMPI-A and MMPI-A-RF provide a gest in terms of factor-analytic research?
relatively unique and quite widely known • How is this scale reflective of non-test-based
approach to the assessment of adolescent func- clinical symptomatology?
tioning. However, given the unique developmen- • Does external validity evidence exist (e.g., dif-
tal issues operating during this period, there are ferential validity studies) to support or refute
particular challenges with interpreting data from my interpretation of this scale?
the MMPI-A/MMPI-A-RF and similar tools. In • Has the validity study been independently
his summary, which is highlighted in Box 6.2, replicated?
Archer promotes an understanding of adolescent • For a particular case, how well do the scale
response patterns on the MMPI-A (and thus, evi- elevations converge with/diverge from other
dence regarding potential psychopathology) evidence collected in the course of the
within a developmental framework. As with any assessment?
assessment tool, it is critical to remember to place
the results in the client’s context. This context Interpretation of the MMPI-A, and now the
includes obvious environmental variables as well MMPI-A-RF, is also supported by practical and
as development itself as an influential contextual thorough books on the topic (e.g., Archer, 1997,
variable. Archer, 2017) as well as numerous workshops
and other continuing professional development
Interpretation Extensive interpretive guidance opportunities. The MMPI-A literature, while not
for each scale is available in the MMPI-A-RF nearly as extensive as that of the MMPI-2, is
manual (Archer et al., 2016). The MMPI-A-RF expansive.
also continues the long tradition of providing A sample case on the following page illus-
profile-based interpretations based on patterns of trates briefly how MMPI-A scale elevations can
elevations. However, a substantial change is an be used in conjunction with other assessment
emphasis on higher-order factor scores to guide information to aid diagnostic decisions and deter-
subsequent interpretation rather than focusing on mining the primary target(s) of intervention (See
the highest (or two highest) clinical scale eleva- Box 6.3 and Box 6.4).
tions in the MMPI-2/MMPI-A tradition. This
approach is more consistent with those of other Strengths and Weaknesses The MMPI-A’s
widely used rating scale systems discussed in this long history is simultaneously its greatest asset
chapter and in Chap. 7 and is also more aligned and liability. On the one hand, the volumes of
with the overall approach to integration of results MMPI research guide practice. In direct contrast,
discussed throughout this text. the original clinical scales are incongruent with
Omnibus Inventories 111
3. A particular issue is the duplication of items nalizing problems. The CBRS self-report form
on different scales which produces high inter- (for ages 8–18) provides an assessment of several
correlations, thus bringing into question the areas of behavioral, emotional, and academic
distinctiveness of measurement of individual functioning and is desirable if the clinician needs
constructs (Kline, 1995). information in more domains than externalizing
4. Length of the MMPI-A and MMPI-A-RF rel- problems.
ative to other self-report inventories, making
the practicality of administering the MMPI-A Scale Content The Conners-3 includes four
an issue. As noted above, the use of the RF content scales: Hyperactivity/Impulsivity,
may help mitigate this issue. Inattention, Aggression, Family Relations, and
Learning Problems. There are four DSM-5
In spite of the extraordinary amount of MMPI Symptom scales (i.e., ADHD Inattentive; ADHD
research available, much remains to be done, Hyperactive-Impulsive, Conduct Disorder, and
especially with the recent advent of the RF. We Oppositional Defiant Disorder). As noted above,
refer the reader to work of Archer (e.g., Archer, the Conners-3 SR includes screening items for
2005, 2017) who has greatly contributed to the depression and anxiety, as well as impairment
body of knowledge regarding MMPI-A validity items for home, school, and social relationships.
and interpretation. Like the BASC, the Conners-3 includes critical
items that may signal the need for further follow-
up. The critical items on the SR are specific to
Conners, 3rd Edition, Self-Report severe conduct problems (e.g., “uses a weapon,”
(Conners-3 SR; Conners, 2008a) “forced sex,” “trouble with the police”).
Consistent with its predecessors, the Conners-3
The Conners-3 SR (Conners, 2008a) is another includes a brief ADHD Index. This scale is based
updated version of a rating scale system with a on items that best differentiate ADHD from non-
long history of research and clinical use. clinical samples. The Conners-3 also includes
However, despite this long history, a self-report three validity scales: Positive Impression (or
instrument had largely been considered experi- “fake good”), Negative Impression (“fake bad”),
mental until the most recent rendition of the and the Inconsistency Index. Lastly, the
Conners rating scale system. The Conners-3 self- Conners-3 SR includes an open-ended item
report rating scale consists of 59 items which are assessing “strengths/skills.” Overall, the
written at approximately a third grade level. A Conners-3 scales and the existence of the validity
39-item Short Form also exists. scales are in line with the current state-of-the-art
It takes approximately 20 minutes to complete in rating scale systems, as well as providing per-
the Long Form of the Conners-3 SR. The inclu- haps more detailed assessment of externalizing
sion of a standard self-report form makes the symptoms than is typical for other rating scales.
Conners system competitive with other well-
known rating scale systems such as those high- Administration and Scoring The Conners-3
lighted in this chapter. It should be noted that the SR is designed for use with youth ages 8–18.
Conners-3 SR includes extensive assessment of Responses are made on a 4-point scale, where
externalizing problems, particularly ADHD, with 0 = not at all true (never, seldom) and 4 = very
screening of anxiety and depression. Depending much true (very frequent). The time frame for
on the referral issue, this design may be either responses is the last month prior to the assess-
ideal or less-than-ideal for the clinician. ment. Both hand scoring and computer scoring
We focus on the Conners-3 rather than its are available as are secure Internet administration
companion rating scale system, the Conners and scoring.
Comprehensive Behavior Rating Scales (CBRS;
Conners, 2008b), because the Conners-3 is rela- The profile form that is included with the
tively unique in its extensive evaluation of exter- response form is used to convert raw scores to
Omnibus Inventories 115
T-scores. The T-scores for the Conners-3 are with the exception of the Conduct Disorder scale
Linear T-scores, meaning that the scales maintain for females with coefficients ranging from 0.74
their distributions when converted to T-scores. to 0.79 and the Aggression scale for 8- and
Male and female norm-referenced scores are 9-year-old females (i.e., 0.75). Two- to four-week
obtained separately. In addition, each age has test-retest reliabilities were also good, with coef-
separate norms. The T-scores shown on the avail- ficients all 0.71 and higher (Conners, 2008a).
able profile forms are truncated such that T-scores
below 40 are not specified. Detailed step-by-step Validity Factor analysis was used extensively in
scoring procedures are available in the Conners-3 development of the Conners-3. Exploratory and
manual (Conners, 2008a). confirmatory factor analyses support a four-
factor model (i.e., Family Relations, Learning
Norming The original normative sample of Problems, Aggression, and Hyperactivity/
1000 cases was collected mostly in the United Impulsivity; Conners, 2008a). Thus, these con-
States, with “a limited amount of data” (Conners, tent scales are considered empirically derived,
2008a; p. 139) collected in Canada. Recruitment whereas the Inattention scale is considered theo-
was targeted to approximate the ethnic/racial dis- retical/rational. The SR content and DSM scales
tribution of the population according to US show moderate to high intercorrelations and
Census statistics. Data presented in the Conners-3 moderate correlations with analogous scales
manual indicate that accurate representation from the Conners-3 parent and teacher rating
across ethnic groups was attained. Only 8% of scales.
cases came from the western part of the United
States, indicating underrepresentation of this Criterion-related validity has been demon-
region. The majority (i.e., 70%) of cases in the strated through correlations with other rating
normative sample of the SR had parents with at scales (Conners, 2008a; Erford, Bardhoshi,
least some secondary education. As described Haecker, Shingari, & Schleichter, 2018).
further in Chap. 7, more recent normative data Specifically, in these analyses, correlations
were collected during an effort to better align between scales on the Conners-3 SR were mod-
subscale content with the DSM-5 (Conners, erately to highly correlated with analogous scales
2014). Normative data for this update were ana- on the BASC-2-SRP and ASEBA YSR. The one
lyzed from 3400 responses with efforts focused exception was a nonsignificant negative relation
on matching data from the 2000 US Census in between the Relations with Parents (higher scores
terms of race and ethnicity. It should be noted indicate better relations) scale on the BASC-2-
that changing demographics likely make norma- SRP and the Family Relations (higher scores
tive data collected based on prior census reports indicate worse relations) scale of the Conners-3
an underrepresentation of racial and ethnic SR for younger children (ages 8–11). This rela-
minorities. tion was significant and negative for older youth
(ages 12–18). Note that the BASC-2-SRP was the
In both the original and the updated normative most recent version of the BASC at the time that
samples, equal numbers of females and males the Conners-3 SR was developed.
were included in the normative sample at each Discriminant validity for the SR is also sup-
age. As noted above, norm-referenced T-scores ported based on comparisons of clinical and gen-
are provided separately for boys and girls, which eral populations. In particular, the Learning
limits interpretation to gender-only Problems, Inattention, Hyperactivity/Impulsivity,
comparisons. and Aggression scales all differentiated not only
the clinical sample from the general sample, but
Reliability Internal consistency coefficients for they also tended to differentiate among individu-
SR are good for both the content scales and DSM als within the clinical sample. For example,
symptom scales. Specifically, coefficients were scores on the Aggression scale were higher for
all above 0.80 in all age groups across genders, individuals diagnosed with disruptive behavior
116 6 Self-Report Inventories
Table 6.8 Internal Consistency of PAI-A Clinical Scales ries reviewed in this chapter. Thus, the clinician
and Subscales (Community Sample) needs to weigh this administration time in light
Internal consistency of the unique content of the PAI-A, as well as the
Scales and subscales coefficient fact that there are no analogous parent- or teacher-
Somatic complaints (SOM) 0.82
report forms for the PAI-A. Items are written at a
Conversion symptoms 0.73
(SOM_C)
fourth grade reading level (Morey, 2007) which
Somatization (SOM_S) 0.58 should allow for self-administration for most
Health concerns (SOM_H) 0.66 adolescents.
Anxiety (ANX) 0.86
Cognitive (ANX_C) 0.80 Linear T-scores are calculated based on the
Affective (ANX_A) 0.61 community sample norms. The scoring software
Physiological (ANX_P) 0.62
also includes a “clinical skyline” which enables a
Anxiety-related disorders (ARD) 0.71
clients’ profile of elevated scores to also be com-
Obsessive-compulsive 0.61
(ARD_O) pared with results from the larger clinical norm
Phobias (ARD_P) 0.47 sample.
Traumatic stress (ARD_T) 0.85
Depression (DEP) 0.86 Norming The norm sample included a commu-
Cognitive (DEP_C) 0.67 nity sample of 707 adolescents ages 12–18 years
Affective (DEP_A) 0.75 recruited to match 2000 US Census data.
Physiological (DEP_P) 0.64 Participants were selected to match the approxi-
Mania (MAN) 0.80
mate demographics of their age group according
Activity level (MAN_A) 0.59
to these census data. A clinical sample of 1160
Grandiosity (MAN_G) 0.68
Impulsivity (MAN_I) 0.76
adolescents mostly recruited from outpatient
Paranoia (PAR) 0.81 mental health settings was also utilized for deri-
Hypervigilance (PAR_H) 0.67 vation of clinical norms. (Morey, 2007).
Persecution (PAR_P) 0.80
Resentment (PAR_R) 0.67
Schizophrenia (SCZ) 0.81 Reliability Average subscale internal consis-
Social withdrawal (SCZ_S) 0.76 tency coefficients of 0.79 (community sample)
Thought disorder (SCZ_T) 0.78 and 0.80 (clinical sample) are reported. Average
Psychotic experiences 0.62
(SCZ_P)
scale test-retest reliability with an average of an
Borderline features (BOR) 0.85 18-day interval is 0.78 (Morey, 2007). As shown
Affective instability (BOR_A) 0.68 in Table 6.8, internal consistencies of scales and
Identity problems (BOR_I) 0.63 subscales in the community sample were gener-
Negative relationships 0.66 ally adequate, with the exception of the
(BOR_N) Somatization, Phobias, and Self-Harm scales.
Self-harm (BOR_S) 0.57 Inter-item relations for such constructs might be
Antisocial features (ANT) 0.87
expected to be lower, as they may tap distinct
Antisocial behaviors (ANT_A) 0.78
physical complaints, sources of phobic symp-
Egocentricity (ANT_E) 0.69
Stimulus-seeking (ANT_S) 0.71
toms, and self-harm behaviors, respectively, that
Alcohol problems 0.79 are not necessarily interconnected.
Drug problems 0.77
From Morey, 2007)
Validity Validity was initially evaluated through
Administration and Scoring Because of its correlations with similar scales from numerous
length, the administration time of the PAI-A is measures (i.e., MMPI-A, BDI, NEO-Five Factor
longer than some of the other self-report invento- Inventory, Personality Inventory for Youth,
118 6 Self-Report Inventories
Adolescent Psychopathology Scales, Symptom Strengths and Weaknesses The PAI-A has
Assessment-45, College Adjustment Scales, some notable strengths and weaknesses.
Clinical Assessment of Depression, Adolescent
Anger Rating Scale, Conners’ Adult ADHD Rating Strengths include:
Scale, Beck Depression Inventory, State Trait
Anxiety Inventory). In summary, PAI-A scales cor- 1. Its unique inclusion of domains of personality
related moderately to higher (i.e., 0.50 and higher) pathology that are often important consider-
with similar subscales in other measures. Some ations in adult clinical assessments
exceptions were somewhat lower correlations 2. Large clinical norm sample
involving PAI-A Somatic Complaints and its sub- 3. Good validity scales and inclusion of critical
scales with broader indices of internalizing prob- items for further follow-up
lems, as well as the Anxiety-Related Disorders 4. Provision of a suitable alternative in terms of
(ARD) scales and measures of anxiety from other content to longer or more dated youth person-
measures which were low to moderate. Of note, ality inventories
validity of the Treatment Consideration Scales was
likewise investigated via correlations with scales Weaknesses include:
from the other listed measures assessing similar
constructs (e.g., PAI_A Aggression with indices of 1. The PAI-A is lengthy relative to many behav-
anger, delinquency, conduct problems, and low ior rating scales, particularly if the unique
agreeableness). In these analyses, relations demon- domains covered are not central to a client’s
strated moderate convergence/divergence in referral question.
expected directions (see Morey, 2007). 2. Relatively small community norm sample.
3. Limited independent research on its psycho-
metric properties and clinical utility.
Interpretation Extensive interpretive guidance
is available through the scoring software and in
the PAI-A manual. However, as emphasized Single-Construct Personality
throughout this text, it is important that the clini- Inventories
cian integrate all assessment evidence from all
methods and informants in drawing conclusion In addition to a continually growing body of
about the client and developing recommenda- omnibus self-report rating scales, a number of
tions. Because the PAI-A has easy-to-interpret single-construct or single-domain scales exist.
validity scales, we recommend starting with These scales are typically oriented toward older
examining scores on those scales to determine if children and adolescents and toward an addi-
clinical scale elevations are interpretable. Next, tional assessment of internalizing problems.
critical items should be reviewed with follow-up Some examples of these scales are reviewed in
as needed for any such endorsements. Then, later chapters when we discuss the assessment of
attention should be paid to clinical scale eleva- specific constructs.
tions, followed by any such scores on the nega- Given the need to approach assessment com-
tive treatment indicator scales. Lastly, prehensively, particularly in regard to assessing
convergence with/divergence from other infor- for comorbidity, we recommend first administer-
mation collected by other methods and infor- ing interviews and rating scales that evaluate a
mants should be considered (see Chap. 15 for variety of domains. As information indicates
further information on integration of assessment problems in a particular domain or more infor-
information for interpretation). mation on a construct would aid in decision-
Conclusions 119
making and recommendations, well-validated evidence will have to be gathered for all mea-
single-construct inventories may be useful sures in order to approximate the higher stan-
(Boxes 6.3 and 6.4). dards of practice required today and to developed
a truly evidence-based approach to assessment.
The current state of self-report personality assess- 1. In the past, parent and teacher reports were
ment continues to improve. In particular, a vari- routinely been preferred over the use of self-
ety of assessment systems have developed report inventories in child personality
methods that allow for relatively easy interpreta- assessment.
tion across self-, parent, and teacher reports. 2. Decisions to use self-report inventories
Personality assessment through self-report rat- should be based, in part, on the child’s devel-
ings continues to lag behind assessment of intel- opmental level, the constructs being
ligence and achievement testing in many ways. assessed, and the purpose of the assessment.
Subscale reliabilities are often still too low to 3. In general, older children and adolescents
support diagnostic decisions and may be influ- are more useful informants than younger
enced by a child’s developmental level and read- children.
ing comprehension. In some cases, subscale 4. Self-reports of covert conduct problems and
reliabilities are too low to support hypothesis internalizing symptoms may be particularly
generation. Only a few of the instruments dis- informative.
cussed herein have been empirically checked for 5. The SRP is one of many components of the
item bias, and few have used modern statistical BASC-3. The SRP attempts to gauge chil-
methods such as structural equation modeling dren’s perceptions and feelings about school,
and latent trait methodology as is commonly parents, peers, and their own behavior
done for intelligence and achievement tests. problems.
Moreover, even the revered MMPI-A and its (a) Six composites are available for the
restructured form (MMPI-A-RF) fail to satisfy SRP: the Emotional Symptoms Index
some basic psychometric standards (see (ESI), Inattention/Hyperactivity,
Table 6.7). Internalizing Problems, School
Improvements have been made in norming Maladjustment, Personal Adjustment,
and the inclusion of reliability and validity data and Functional Impairment.
in test manuals. Some of the noteworthy contri- (b) The SRP includes three validity scales.
butions have been the inclusion of potentially (c) The SRP was normed with on a national
informative self-perception and content scales on sample of 300 children and 600
the BASC-3-SRP, the restructuring of the adolescents.
MMPI-A, the numerous cross-cultural studies (d) The reliability of the SRP scales is good
with the YSR, the extensive assessment of vari- as indicated by a variety of methods.
ous externalizing problems on the Conners-3,
(e) Until clinical experience and further
and the development of the PAI-A. Psychologists research studies are available, initial
clearly have more and better options than in years efforts at SRP interpretation should
past, and youth self-report of emotional, social, focus on the scale level rather than the
and behavioral functioning is valuable for obvi- composite level because these item
ous reasons. Still, much validity evidence remains pools have some rational, theoretical,
to be gathered, particularly with so many mea- and research basis.
sures being relatively new to the market. 6. Relatively few independent studies have
Relatively few independent validity studies are been conducted with the most recently
available for some of these scales. Much more update ASEBA-YSR. However, numerous
120 6 Self-Report Inventories
studies have demonstrated that the previous 4. A thorough evaluation of the ado-
version of the YSR produces similar scores lescent’s self-appraisal due to the
across cultures and consistent sex differ- variety of items presented.
ences across cultures. (a) Weaknesses of the MMPI-A/MMPI-
(a) The YSR does not include validity
A-RF include:
scales. However, it includes an assess- 1. Uncertainty as to the correspon-
ment of clinical and adaptive domains dence of scale elevations to pathol-
that are relevant for most purposes of ogy in the adolescent period of
child assessment. development (Archer, 2005).
7. The MMPI-A has strong roots in the original 2. A limited research base for the
work of Hathaway and McKinley (1940). A MMPI-A-RF itself.
restructured form (MMPI-A-RF; Archer, 3. Duplicating items on different
2017) has recently been developed and aims clinical scales on the MMPI-A,
to capture more clinically relevant content in which produces high intercorrela-
a shorter inventory than the MMPI-A. tions, thus compromising the dis-
(a) The MMPI-A-RF has nine clinical tinctiveness of measurement from
scales, five validity scales, and several individual scales
specific problems scales. 8. The self-report form of the Conners-3 is the
(b) The MMPI-A-RF is much shorter than current version of a long-standing rating
the MMPI-A but does not have analo- scale system. Until now, the self-report form
gous parent- and teacher-report forms to had largely been considered experimental.
aide in multi-informant assessments. (a) The Conners-3 SR has good initial valid-
(c) Several internal consistency coefficients ity evidence. It offers an extensive
for MMPI-A-RF are below minimum assessment of ADHD and disruptive
standards. behaviors but a limited assessment of
(d) Strengths of the MMPI-A/MMPI-A-RF internalizing problems.
include: 9. The PAI-A provides a unique self-report
1. The familiarity of the MMPI sys- assessment of personality and behavior in
tem to a large group of devoted users that it includes subscales for core clinical
2. The existence of a substantial constructs and several subscales assessing
research base on the MMPI-A that personality pathology.
informed the development of the 10. A number of single domain self-report rating
MMPI-A-RF. scales, particularly for internalizing prob-
3. Interpretive flexibility because of lems, are in existence. Examples of these
the numerous scales available. instruments are discussed in later chapters.
Parent and Teacher Rating Scales
7
item content of these measures, parents respond Loeber, Green, Lahey, & Stouthamer-Loeber,
based on how previously vague notions of hyper- 1991; Tripp, Schaughency & Clarke, 2006).
activity, depression, nervousness, and the like However, parents are still considered necessary
have been operationalized in the rating scale, pre- and useful in providing information about inat-
sumably reducing the impact of their own idio- tention and hyperactivity (Tripp et al., 2006) and
syncratic perceptions of their child’s symptoms. in documenting effects of treatment for ADHD
Ratings by parents also ensure increased ecologi- (Biederman, Gao, Rogers, & Spencer, 2006). In
cally valid assessment of children’s behavior. addition, parents may be in a unique position to
However, all rating scales, including parent understand the antecedents of a child’s disruptive
ratings, can be accused of bias and susceptibility behaviors, as they can observe their child more
to response sets (Wirt et al., 1990). Even biased closely than a teacher who works with several
reporting, however, can be of value. If, for exam- children simultaneously. Parents have also been
ple, parent ratings provide very different results discussed as particularly important observers and
when compared to the ratings of others, the clini- informants of child anxiety and depression
cian can develop some important insights into (Klein, Dougherty, & Olino, 2005; Silverman &
child or family functioning. If a child’s father Ollendick, 2005).
rates his son as having significantly more behav- Evidence is mixed on the degree to which the
ioral problems than the mother rates the child as child’s age influences agreement in ratings across
having, the clinician can explore the potential informants (see De Los Reyes & Kazdin, 2006
reasons behind the ratings. A straightforward and Chap. 14). Parents, in particular, may be use-
explanation may be that the father is doing the ful informants of a child’s functioning through-
majority of the child care. This information could out youth, although there may be discrepancies
be important to acquire if the presumption had between their reports and the reports of others.
been that the child’s mother was providing most The information used in conjunction with parent
of the caretaking. reports may vary with age. More specifically,
parents are obviously vitally important sources of
information for young children in such areas as
Factors Influencing Parent Ratings conduct problems, whereas the children them-
selves would not be reliable (and, thus, not valid)
As discussed in more detail in Chap. 15, some informants. Teachers, however, could offer useful
research has indicated that parental, specifically perspectives of the young child’s social, aca-
maternal, distress may influence ratings of child demic, and behavioral functioning. For adoles-
functioning in a negative way (i.e., higher reports cents, parents may still provide valid and useful
of problems/symptoms). Although the issue of information, but their knowledge of the child’s
whether maternal distress is directly influential conduct problem behaviors may be more limited,
on ratings of a child’s symptoms is far from set- as the behaviors may sometimes occur outside of
tled, it stands to reason that stressful home envi- the parent’s awareness. The adolescent—pro-
ronments would be positively correlated with vided that he or she is willing to provide such
parent reports of child symptoms (either as a pre- information—would be the most knowledgeable
sumed cause or presumed effect of the child’s informant of these behaviors, and the teacher’s
functioning). contribution would presumably diminish.
The construct being evaluated and the child’s Nevertheless, the parent’s awareness, or lack
developmental level are two additional factors thereof, regarding the adolescent’s conduct
that may influence parents’ reports. Teachers would be an important clinical consideration.
have traditionally been considered superior to Finally, parent ratings are more likely to attri-
parents as reporters of a child’s ADHD-like bute the child’s problems to dispositional factors
symptoms that are often most evidence in the in the child, whereas youth self-reports are more
classroom (Loeber, Green & Lahey, 1990; likely to indicate the family environment as a
Evaluating Children via Teacher Ratings 123
factor in need of intervention (see De Los Reyes tivity, it is not surprising that the primary assess-
& Kazdin, 2006). Thus, informants (including ment instruments for children’s school behavior
parents, teachers, and children) may base their have been teacher-completed behavior rating
ratings of a child’s functioning on the attributions scales. In addition to suggestions for appropriate
that they make regarding the genesis and mainte- use of rating scales in general discussed previ-
nance of the child’s problems. ously, there are several considerations for inter-
That parent ratings may be influenced by fac- preting information from teachers that warrant
tors that are not necessarily directly tied to the special attention.
child’s actual functioning does not render parent
ratings questionable. Instead, it calls to mind the
many potential variables to consider in under- Factors Influencing Teacher Ratings
standing the child’s presenting problems—an
understanding that is critical for case conceptual- As described above for parents, the usefulness of
ization and subsequent recommendations for teacher information may vary depending on what
intervention. type of behavior is being assessed. However, lit-
tle recent research has examined the utility of one
informant over another. Traditionally, teachers
valuating Children via Teacher
E have been considered the best source of informa-
Ratings tion on a child’s attention problems and overac-
tivity because they have the opportunity to
Although teachers have traditionally been con- observe the child in a situation with high demands
sidered an important source of information on for sustained attention and inactivity. Early
children’s academic performance, they often are screening by teachers can indeed predict a variety
not used in the assessment of children’s behav- of behavioral and social difficulties in school in
ioral and emotional functioning. However, know- later years, but it is unclear the extent to which
ing how a child behaves in the classroom is they predict later behavioral problems that occur
important for several reasons. First, school is a outside a classroom or school setting (Racz et al.,
setting in which the child spends several hours a 2013).
day. Therefore, a child’s adjustment to the school The usefulness of teacher information may
setting can have a dramatic impact on his or her also vary according to the age of the child.
overall psychological functioning. Second, the Children in early elementary school frequently
multiple demands of the school environment have one teacher who observes a child across sev-
(e.g., to stay seated, to follow demands of adults, eral class periods, if not the entire school day. In
to interact with classmates) present many chal- contrast, high school teachers frequently have
lenges to the child, challenges that may not be students for one class period during the day and
present in other settings. Third, the demands of consequently likely have less knowledge about
the school setting change as a child progresses each specific student’s functioning.
through school (e.g., demands for organization, A final issue in interpreting teacher rating
the importance of social acceptance). Therefore, scales is understanding the frame of reference or
understanding school-related problems that are standard used by teachers. As discussed previ-
unique to a given period can provide clues to spe- ously and through recent evidence (see Splett
cific problems in adaptation that a child or ado- et al., 2018), a number of characteristics of a
lescent might experience. rater, the person being rated, and the assessment
Based on these considerations, there is context can influence his or her judgment of the
increasing interest in assessing a child’s behav- intensity, quality, and/or frequency of a child’s
ioral and emotional functioning in the school set- behavior. In the case of teacher ratings, one char-
ting. Given the many advantages of behavior acteristic of teachers that can influence their rat-
rating scales, such as time efficiency and objec- ings is their experience with many children of the
124 7 Parent and Teacher Rating Scales
same age. Experience allows the teacher to make this broad overview of the selected scales is not
some internal normative comparison of a child’s designed to replace information provided in the
behavior with the behavior of other children the technical manuals that accompany these instru-
teacher has taught. This internal norm is a double- ments, which should be reviewed by any user of
edged sword. It often gives the teacher a unique the scales.
perspective of knowing both the individual child The parent and teacher rating scales reviewed
and knowing what behaviors are age-appropriate. in this chapter are highlighted in Tables 7.1 and
However, some teachers, such as teachers who 7.2, respectively. Commonly used multidomain/
work in special education classrooms, may have multisyndrome/omnibus measures, as opposed to
a skewed base of comparison that could influence single-construct measures, are the focus of this
their ratings. That is, their ratings of a child’s chapter. Although these scales are discussed in
behavior may be influenced by a comparison of isolation to balance clarity and specificity, it
the child to other disturbed children. should be recalled that they should be used as
Despite these cautions and limitations, we feel part of comprehensive assessment using multi-
strongly that teacher ratings are an essential ele- method/multitrait/multisource assessment meth-
ment of a comprehensive clinical assessment of ods that are discussed in various chapters of this
children’s behavioral and emotional functioning. book. As noted, in Chap. 15, teacher and parent
Carlson and Lahey (1983), in an early review of ratings only typically produce moderate correla-
teacher ratings, reported that most of the teacher tions. Although such discrepancies present chal-
rating scales available at that time suffered from lenges to the assessment process and may be
significant psychometric problems in develop- cause for some concern regarding the validity of
ment and inadequate norming. As a result, the ratings, they may also signal true differences in
available scales were severely limited in their the child’s functioning across settings. Therefore,
usefulness for clinical evaluations. Fortunately, parent and teacher ratings can be viewed as pro-
since that 1983 review, there have been numerous viding unique information toward answering a
advances in teacher rating scales, with many of referral question that, used with other evidence,
the inadequacies of earlier scales eliminated or at can inform useful recommendations. The integra-
least greatly reduced. tion of components and information from differ-
ent informants and methods is discussed in the
context of interpretation in Chap. 15 and in sub-
verview of Omnibus of Parent
O sequent chapters that address specific
and Teacher Rating Scales syndromes.
Behavior Checklist 6–18 externalizing, competence; 8 grade 20 minutes school-age version); 1-week (clinical vs.
(CBCL; Achenbach & problem scales; test-retest (good); interrater nonclinical groups);
Rescorla, 2000; 5–6 DSM-oriented scales (moderate for preschool limited info on
Achenbach & Rescorla, version; good for school-age relations with other
2001) version) tools; limited info on
current version
Conners third Edition, 6–18 110 items; 43-item Short fForm; 4th–5th 20 minutes E Internal consistency, 2- to Good evidence of E
Parent Rating Scale 3 validity scales; 6 empirical grade 4-week test-retest, and criterion-related and
(Conners-3; Conners, scales; 1 rational scale; 4 DSM-5 interrater reliability all good differential validity
2008a, 2014)a scales
Note: Norm sample evaluated as E = excellent; G = good; F = fair; MI = degree of enhancement of multi-informant assessment; E = Excellent; G = Good
a
We view the Conners-3 as providing an excellent means of assessment of externalizing problems; however, if an extensive assessment of internalizing or learning difficulties is
desired, the Conners Comprehensive Behavior Rating Scales (CBRS; Conners, 2008b) is recommended
125
126
that span the preschool (2–5 years), child Table 7.3 BASC-3-PRS scale definitions and key symp-
toms as indicated by items with the highest factor load-
(6–11 years), and adolescent (12–21 years) age
ings per scale
ranges. The PRS takes a broad sampling of a
Activities of Skills associated with performing
child’s behavior in home and community daily living everyday tasks; “Sets realistic
settings. goals”; “Needs help putting clothes
on”; “Needs help bathing”
Content As with its predecessors, the BASC-3- Adaptability Ability to adapt to changes in the
PRS was developed using both rational/theoreti- environment; “Adjusts well to
changes in routine”; “Adjusts well
cal and empirical means in combination to to changes in plans”; “Recovers
construct the individual scales. The benefit of this quickly after a setback”
approach is that the resulting scales have rela- Aggression Tendency to act in verbally or
tively homogenous content. The uniqueness of physically hostile or threatening
manner; “Hits other children
the scales was also enhanced by not including (adolescents)”; “Throws or breaks
items on more than one scale. Table 7.3 provides things when angry”; “Is overly
descriptions and item examples for each scale. aggressive”
There are four composites: Externalizing Anxiety Tendency to be nervous, fearful, or
Problems, Internalizing Problems, Adaptive worried; “Worries about making
mistakes”; “Worries”; “Is easily
Skills, and a Behavioral Symptoms Index that stressed”; “Is fearful”
includes some internalizing and externalizing Attention Tendency to be easily distracted or
scales (i.e., Atypicality, Attention Problems, problems have difficulty concentrating; “Is
Hyperactivity, Aggression, Depression, and easily distracted”; “Has trouble
concentrating”; “Pays attention
Withdrawal). Two types of scales are included at
when spoken to” (reversed scored)
each age level: clinical and adaptive. Clinical Atypicality Tendency to behave in “odd”
scales of the PRS are designed to measure emo- manner; “Acts strangely”; “Does
tional, behavior, and social problems much like strange things”; “Does weird
other measures discussed below in that behav- things”
Conduct Tendency to engage in antisocial
ioral excesses (e.g., hitting others, getting out of
problems and rule-breaking behavior;
seat) are the focus of assessment. The PRS also “Breaks the rules”; “Lies”;
includes critical items that are thought to warrant “Disobeys”
follow-up or clinical attention in their own right. Depression Feelings of unhappiness, sadness,
These items (e.g., “Has a hearing problem,” or stress; “Is easily upset”; “Is
irritable”; “Changes mood quickly”
“Bullies others”) are not necessarily indicative of
Functional Ability to communicate ideas and
the most severe pathology; instead, they may be communication express oneself clearly;
worthy of further questioning or recommenda- “Communicates clearly”;
tions by the clinician. “Responds appropriately when
asked a question”; “Is able to
describe feelings accurately”
The adaptive scales measure behavioral defi- Hyperactivity Tendency to be overly active and
cits (e.g., “adjusts well to changes in family act without thinking; “Has poor
plans”) or skills that are associated with good self-control”; “Acts out of control”;
adaptation to home and community (see “Acts without thinking”; “Is unable
to slow down”
Table 7.3). Each of the parent forms of the
Leadership Possessing skills needed to
BASC-3 includes seven optional content scales: accomplish goals, ability to work
Anger Control, Bullying, Developmental Social with others; “Gives good
Disorders, Emotional Self-Control, Executive suggestions for solving problems”;
“Is good at getting people to work
Functioning, Negative Emotionality, and
together”; “Is highly motivated to
Resiliency. As with the BASC-3-SRP (see Chap. succeed”
6), the content scales for the BASC-3-PRS were (continued)
constructed via theoretical and empirical meth-
128 7 Parent and Teacher Rating Scales
ods. These scales were initially developed in the These various norm-referenced comparisons
BASC-2 system and are based on current theo- are more than typically offered for such scales.
retical and empirical perspectives on important The general national norming sample is advised
domains of youth functioning (Reynolds & as the starting point for most purposes (Reynolds
Kamphaus, 2015). There exists very little & Kamphaus, 2015). Of course, depending on
research on these scales, yet their labels and item the question, the clinician may opt for gender or
content are intriguing and warrant further investi- clinical norms. Gender norms may be useful in
gation of their reliability, validity, and clinical trying to convey a child’s current level of func-
utility. Initial analyses indicate that the PRS con- tioning to others, such as parents. In other words,
tent scales possess good (i.e., 0.79 and higher) one might present the child’s scores relative to
internal consistencies (Reynolds & Kamphaus, the general population and then emphasize how
2015). the child compares to other boys/girls on areas of
concern to provide a more complete picture.
Administration and Scoring The PRS uses a However, too much information may cause con-
four-choice response format (i.e., never, some- fusion for some parents. Gender-based norms
times, often, almost always) with no space may also help answer some specific research
allowed for parent elaboration. According to the questions, i.e., correlates of inattention and
authors, each age level of the scale takes about hyperactivity among girls and among boys.
10–20 minutes for parents to complete. The clinical norms may be particularly useful
in comparing a child who is in an inpatient or
A variety of derived scores and interpretive residential setting to others with relatively severe
devices are offered. Linear T-scores are available difficulties. That is, it may be understood that a
for all scales and composites, meaning that the child is functioning poorly compared to most
original distributions for these indices in the other children his/her age (i.e., elevations on gen-
norming sample were maintained. Other scores eral norms), but it may be informative to consider
available include percentile ranks, confidence how the child functions (e.g., “How severe are his
bands, and statistical methods for identifying conduct problems?”) in comparison with other
high and low points in a profile. T-scores can also children with emotional and behavioral difficul-
be obtained for three norm groups. Both hand ties for treatment planning purposes.
scoring and computer entry scoring are available The general norm sample for the PRS included
for the PRS. 600 preschoolers, 600 children, and 600 adoles-
cents. Cases were collected at test sites in 44
Behavior Assessment System for Children, Third Edition 129
states. Across age groups, the PRS sample closely larly holds for the composites and for the exter-
matches US Census statistics (US Census Bureau, nalizing problem scales. Generally, the
2013) in terms of parent education level, race/ internalizing problem scales (e.g., Anxiety,
ethnicity, and geographic region (see Reynolds & Depression, Somatization) show moderate to
Kamphaus, 2015 for more details). The clinical high correlations with analogous scales from
norming sample for the PRS included responses other measures (see Reynolds & Kamphaus,
from 755 parents, with the most common classi- 2015).
fication of cases being ADHD or ASD.
Interpretation The same logical interpretive
Reliability The median reliability coefficients steps that were outlined for the BASC-3-SRP
provided in the manual suggest good evidence of (discussed in Chap. 6) also apply to the BASC-3-
reliability for the individual scales and compos- PRS. Briefly, the clinician should:
ites. All scales and composites have median reli-
ability estimates of 0.72 and above, with the
exception of Activities of Daily Living. The 1. Assess response validity using validity
BASC-3 manual also provides information on 1- indexes and informal means (e.g., inspect for
to 7-week test-retest reliability and interrater reli- a high number of items with no response)
ability between parents. Test-retest reliability 2. Inspect critical items and follow-up as
coefficients were 0.80 and higher. Interrater reli- appropriate
ability was generally high (>0.60), with the 3. Interpret scores on clinical scales and com-
exception of Anxiety (0.54) and Adaptability posites, with particular attention to elevations
(0.58) on the child form. (T-scores of 65–70 and higher) on clinical
scales and low scores (T-scores of 35 and
below) on adaptive scales
Validity The PRS appears to have at least the 4. Attend to items that appeared to have led to
same, if not better, coverage of important and rel- scale elevations (or low adaptive scores)
evant content as its predecessor. The PRS assesses 5. Integrate with ratings from other parents or
a variety of externalizing behaviors and appears other informants
to have improved content validity within certain 6. Integrate data with information from other
scales (e.g., Aggression). In addition, as with the assessment tools (e.g., interview, behavioral
previous versions, the BASC-3 enjoys consider- observations, intelligence testing) to make
able factor analytic support for the PRS and diagnoses
retains a three-factor model consisting of exter- 7. Set objectives for treatment/intervention
nalizing problems, internalizing problems, and
adaptive skills. The strongest measures of the As was the case with the SRP, we again rec-
externalizing factor are the Aggression, Conduct ommend a focus on interpretation at the clinical
Problems, and Hyperactivity. The Internalizing and adaptive scale level, as the reliabilities of the
factor is marked by strong loadings of the PRS scales are generally good and elevations on
Depression, Anxiety, and Withdrawal scales with scales are more specifically informative than
moderate loadings by Atypicality and would be the case for elevated composite scores.
Somatization. Adaptive Skills scales that load There is quite limited independent research
highly on this factor are Adaptability, Activities information available to date on the BASC-3-
of Daily Living, Functional Communication, PRS. However, one study focused on the pre-
Leadership, and Social Skills. school version of the available screening form
(Behavioral and Emotional Screening System;
Criterion-related validity analyses produced BESS) demonstrated good construct validity
consistent associations between the PRS and with some differential item functioning across
other parent rating scales. This pattern particu- ratings of children from different racial/ethnic
130 7 Parent and Teacher Rating Scales
backgrounds (DiStefano, Greer, & Dowdy, place that information in the context of other
2019). Such issues in the screener and full ver- information obtained in the overall BASC system
sions of the BASC-3 deserve further attention. (e.g., self-report scale, parent rating scales, class-
Nevertheless, the combined rational and empiri- room observation system). As with the PRS,
cal approach to scale development has intuitive there are three forms of the BASC-3-TRS: pre-
appeal for use in clinical situations. Clinicians school (ages 2–5), child (6–11), and adolescent
are still urged to keep abreast of the research lit- (12–21). The three forms contain behavioral
erature discussing the strengths and limitations of descriptors that are rated by the teacher on a
any assessment tool. 4-point scale of frequency, ranging from “never”
to “almost always.” The three forms have 105
Strengths and Weaknesses The BASC-3-PRS items for the preschool version, 156 for the child
has a number of apparent strengths and weak- version, and 165 for the adolescent version.
nesses. The strengths include:
Table 7.4 Composites and scales of BASC-3-TRS and withdrawal, which aid in the assessment of
Composite Scales emotional difficulties. Fourth, the BASC-3-TRS
Externalizing problems Aggression includes items that screen for learning problems
Hyperactivity and difficulties with study habits that often
Conduct problems accompany emotional and behavioral problems
Internalizing problems Anxiety
in children.
Depression
Somatization
School problems Attention problems Administration and Scoring The BASC-TRS
Learning problems takes approximately 10–20 minutes to complete.
Adaptive skills Adaptability The cover of the record provides instructions to
Functional communication the teacher for completing the form and space for
Leadership recording background information about the
Social skills child and teacher (e.g., age, gender, type of class,
Study skills length of time in class). Both hand scoring and
Behavioral symptoms Hyperactivity computer scoring are available. Norm tables in
Index Aggression
the BASC-3 manual are provided so that any of
Depression
four sets of norms can be used: general, male,
Attention problems
Atypicality female, and clinical (see above for discussion of
Withdrawal the uses of these different types of norms). Both
T-scores and percentile ranks are listed for each
set of norms, with linear T-scores again being uti-
content of the BASC-2. The BASC-3-TRS con-
lized for the TRS. As with the other BASC-3 rat-
sists of 5 composites (i.e., Behavioral Symptoms
ing scales, the BASC-3-TRS scoring software
Index, School Problems, Externalizing Problems,
highlights critical items (e.g., “I want to kill
Internalizing Problems, Adaptive Skills) across
myself”) that are clinically important or that war-
all age ranges, with 11 scales in the preschool
rant further follow-up.
version and 15 scales in the child and adolescent
versions. The scales grouped into the compos-
ites—except for the Behavioral Symptoms Index
Norming The norming group for the TRS
which includes the Hyperactivity, Atypicality,
included 500 preschoolers, 600 children (ages
Depression, Aggression, Attention Problems, and
6–11), and 600 adolescents (ages 12–18) with
Withdrawal scales—are provided in Table 7.4.
equal sex distributions in all age groups.
The TRS also has the same optional content
Respondents were recruited from sites through-
scales as those provided for the PRS (see above).
out the United States. As described previously for
Limited information is available on the clinical
the PRS, the sampling procedures for obtaining
utility of the content scales, although the BASC-3
the normative sample were designed to closely
manual provides detailed information on their
mirror US Census statistics in terms of parent
reliability and criterion-related validity.
education, race/ethnicity, and geographic region
(see Reynolds & Kamphaus, 2015). Details
The content coverage of the BASC-3-TRS regarding the 611-member clinical sample for the
scales has several unique features relative to TRS are also provided in the BASC-3 manual.
other teacher rating scales. First, it provides com-
prehensive coverage of several areas of adaptive
behavior. Second, the BASC-3-TRS continues Reliability The manual for the BASC-3-TRS
the strategy of including separate scales for (Reynolds & Kamphaus, 2015) provides evidence
hyperactivity and attention problems, which aids on three types of reliability: internal consistency,
in the differentiation of presentation subtypes of test-retest reliability, and interrater reliability. With
Attention-Deficit Hyperactivity Disorder. Third, very few exceptions, the scales of the BASC-3-
there are separate scales for anxiety, depression, TRS proved to be quite reliable in the normative
Behavior Assessment System for Children, Third Edition 133
sample. More specifically, median internal consis- are available for the PRS and SRP) provide
tency coefficients tended to average well above another initial point of interpretation. Critical
0.80 across all age groups, and all were 0.77 or items should be reviewed promptly, because
higher. Similarly, test-retest reliability over these items tend to be clinically important indica-
1–7 weeks was high with coefficients 0.79 or tors that deserve careful follow-up assessment.
higher. Finally, the consistency of ratings between
two teachers was tested in samples of preschool- The reliability estimates at the clinical and
age children (n = 71), school-age children adaptive scale level of the TRS are good; there-
(n = 116), and adolescents (n = 80), with moderate fore, we again recommend focusing interpretation
to high reliability estimates emerging across age mainly at the scale, rather than composite, level
groups and scales (i.e., coefficients from 0.51 to since more specific information is available on the
0.84). Of note, interrater reliability of teacher rat- TRS clinical and adaptive scales. Interpretations at
ings for the adolescent TRS was substantially the item level must be made quite cautiously due
improved in the BASC-3 relative to the BASC-2 to the low reliability of individual items. It is often
(see Reynolds & Kamphaus, 2004, 2015). informative to see which items led to a child’s or
adolescent’s elevation on a given clinical scale.
For example, for a child with an elevation on the
Validity The TRS is closely, but not exactly, Adaptability scale, it may be informative to deter-
aligned with the item content of the PRS. The TRS mine if this elevation was largely due to problems
also has additional scales (i.e., Study Skills, specifically within interpersonal domains or due to
Learning Problems) that seem particularly relevant more general problems in adapting to changes in
for use with a teacher rating scale. The BASC-3 routine. Finally, interpretation at the scale level for
manual provides factor-analytic support for the the parent form is a viable early step in interpretation
construct validity of the scales and composites of (see above); therefore, interpretation at the scale
the TRS. In addition, initial research on the TRS level of the TRS facilitates integration of informa-
shows generally high correlations with analogous tion across parent and teacher ratings. In addition,
scales from other teacher rating scales. However, considering individual items within elevated scales
the correspondence to analogous scales is slightly on both rating forms may help determine the
lower for internalizing of problems than for the source of consistencies and inconsistencies across
indices of externalizing problems (see Reynolds & parent and teacher ratings, further informing case
Kamphaus, 2015). A limitation of the latest ver- conceptualization and recommendations.
sion of the TRS is its lack of independent research
to date on its validity and utility. Strengths and Weaknesses Like its companion
parent rating scale, the BASC-3-TRS has a num-
ber of strengths and weaknesses.
Interpretation As with the SRP and PRS, the
BASC-3-TRS includes validity scales that pro- Notable strengths include:
vide a useful and efficient first point of interpreta-
tion. More specifically, it contains a “fake bad” 1. It is part of a multi-method, multi-informant
index (F), which helps to assess the possibility system that aids in a comprehensive clinical
that a teacher rated a child in an overly negative evaluation with item content that covers
pattern. Therefore, interpretation of this scale in important problematic and adaptive domains
particular should be the first step in the interpre- of classroom behavioral and emotional
tative process, keeping in mind that a high score functioning.
on the F index may actually indicate highly prob- 2. The assessment of adaptive functioning and
lematic functioning. Therefore, this validity academic problems are more expansive than is
index should be interpreted in the context of other typical for teacher rating scales.
assessment data. The Consistency Index and the 3. The improved reliability coefficients for this
Response Pattern Index available for the TRS (as version of the TRS.
134 7 Parent and Teacher Rating Scales
Table 7.5 Sample content of CBCL (6- to 18-year-old DSM-Oriented scales were slightly revised
version) syndrome scales
for DSM-5. Specifically, there is now an Autism
Anxious/Depressed—Cries a lot, is fearful, feels Spectrum Problems scale for the preschool ver-
worthless or inferior, talks of suicide
sion (along with Depressive Problems, Anxiety
Withdrawn/Depressed—Would rather be alone, shy/
timid, sad Problems, Attention-Deficit Hyperactivity
Somatic Complaints—Feels dizzy and constipated, has Problems, and Oppositional Defiant Problems).
headaches, nausea, nightmares The school-age version does not have the Autism
Social Problems—Dependent, lonely, gets teased, Spectrum Problems scale but includes Conduct
doesn’t get along with other kids Problems and Somatic Problems in addition to
Thought Problems—Can’t get mind off of certain
the other DSM-5 scales noted for the preschool
thoughts, sees things, harms self, strange behavior
Attention Problems—Can’t concentrate, daydreams, version. The Anxiety Problems and Somatic
impulsive, can’t sit still, fails to finish things Problems DSM-Oriented scales have been
Rule-breaking Behavior—Drinks alcohol, lacks guilt, slightly altered. The Depressive Problems,
breaks rules, sets fires, prefers to be with older kids Attention-Deficit Hyperactivity Problems,
Aggressive Behavior—Argues a lot, destroys others’ Oppositional Defiant Problems, and Conduct
things, gets in fights, mood changes, threatens people
Problems DSM-Oriented scales remain
From Achenbach and Rescorla (2001)
unchanged due to the criteria for their pertinent
diagnoses remaining the same from DSM-4 to
for the 1½- to 5-year-old version are Emotionally DSM-5. Items comprising the revised scales were
Reactive, Anxious/Depressed, Somatic determined in consultation with experts from 30
Complaints, Withdrawn, Sleep Problems, societies as reported at www.aseba.org/dsm5.
Attention Problems, and Aggressive Behavior html.
(Achenbach & Rescorla, 2000).
On both versions there is a Total Problems Administration and Scoring The CBCL is eas-
score (the most global score available on the ily administered in 15–20 minutes. The CBCL is
CBCL) as well as composites for Internalizing somewhat unique in that adaptive behavior is
Problems and Externalizing Problems (see assessed with a combined fill-in-the-blank and
Achenbach & Rescorla, 2000, 2001). The CBCL Likert scale response format. In addition, some of
also includes competence scales that are designed the problem behavior items (e.g., “trouble sleep-
to discriminate significantly between children ing,” “speech problem,” “fears certain animals,
referred for mental health services and non- situations, or objects other than school”) ask the
referred children (Achenbach & Rescorla, 2001). parent to elaborate on or describe the problem
The competence scales assess Activities, Social, endorsed. This format is advantageous in that it
and School competencies, although these scales allows the parent to respond in an open-ended
are not available on the preschool version of the format. Clinicians can gain access to qualitative
CBCL. information of value using this format. Open-
Unique to the ASESBA system, including the endedness, however, also has a disadvantage: It
CBCL, is extensive work and discussion on the may extend administration time and requires
application of these tools to multicultural set- more decision-making on the part of the parent.
tings. This work includes the derivation of scales
for obsessive-compulsive problems, sluggish Hand scoring and computer scoring are avail-
cognitive tempo, and post-traumatic stress that able for the CBCL. The CBCL offers normalized
were first developed using multicultural norms. T-scores as the featured interpretive scores.
The reader is referred to Achenbach and Rescorla Percentile ranks are also provided. T-scores are
(2010a, 2010b) for detailed discussion for the use available for all problem scales and the three
of the ASEBA measures in the context of multi- composites, Externalizing, Internalizing, and
cultural assessment. Total, as well as for the Competence scales.
Achenbach System of Empirically Based Assessment 137
The advantages and disadvantages of using SES background, 49% from middle SES, and
normalized versus linear T-scores are debatable, 17% from a lower SES background. Again, 40%
as we discuss in Chap. 2. On the one hand, the of these respondents were from the southern
advantage of comparable percentile ranks across United States (Achenbach & Rescorla, 2000).
scales was recognized by the MMPI-A author Children were excluded from the sample if
team (see Chap. 6). Normalized scores, however, they had “received mental health or special edu-
clearly change the shape of the many skewed raw cation classes during the previous 12 months”
score distributions forcing the T-score distribu- (Achenbach & Rescorla, 2001; p. 76). As result,
tion to take a shape that it does not actually take this standardization sample should be considered
in the general population (see Chap. 2). In addi- a “normal” sample, rather than a “normative”
tion, the reporting of T-scores on the CBCL is sample. Separate clinical norms are not offered
truncated for the syndrome scales, such that low for the CBCL.
scores are reported simply as T ≤ 50. For the
Competence scales, the distribution is truncated Reliability The CBCL has good evidence of
above a T-score of 65 and below a T-score of 35. reliability with internal consistency coefficients
ranging from 0.78 to 0.97 on the syndrome scales
Norming The norming of the school-age CBCL and somewhat lower internal consistency on the
was based on a national sample of 1753 children Competence scales (i.e., 0.63–0.79; Achenbach
aged 6 through 18 years. This sample was & Rescorla, 2001). On the preschool version of
collected in 40 states and the District of Columbia the CBCL, the internal consistency coefficients
(Achenbach & Rescorla, 2001). Relevant stratifi- for the syndrome scales and Composites ranged
cation variables such as age, gender, ethnicity, from 0.66 to 0.95.
region, and SES were recorded in an attempt to
closely match US Census statistics on these vari- A 2-year interval test-retest study of 67 chil-
ables. The respondents were mothers in 72% of dren yielded coefficients ranging from 0.45 for
the cases and fathers in 23% of the cases (5% of Somatic Problems to 0.81 for Aggressive
the cases used “others”). Sixty percent of the Behavior for the school-age CBCL. These coef-
respondents were classified as White, with ficients are indicative of strong reliability in light
Hispanics appearing to be somewhat underrepre- of the lengthy interval and the expected natural
sented (9%). Fifty-one percent of cases were instability in some of these areas over time.
from a middle SES background, 33% were from Lastly, mother-father interrater agreement on the
an upper SES background, and 16% were from a CBCL was generally good on the school-age ver-
lower SES background. Forty percent of respon- sion with all coefficients except the Activities
dents were from the southern part of the United scale being 0.63 or higher. The interrater agree-
States (see Achenbach & Rescorla, 2001). From ment on the preschool version was lower, with
this sample, separate norms were developed for coefficients ranging from 0.48 to 0.67 (see
ages 6–11 and 12–18, with each of these groups Achenbach & Rescorla, 2000). No reliability
further delineated by gender. data have been reported for the revised DSM-5-
Oriented scales.
The norming sample for the preschool CBCL
version for ages 1½–5 was also recruited in an Validity Much of the validity evidence reported
attempt to match US Census statistics on the by the authors of the CBCL focuses on the ability
same variables. This sample consisted of 700 of scales to differentiate clinical from nonclinical
respondents (76% mothers, 22% fathers, 2% samples. Results of these analyses indicated good
“other”). Fifty-six percent of respondents were differential validity across scales for both males
White, 21% African-American, 13% Latino, and and females. Some evidence suggests that the
10% mixed or others. In the preschool norming CBCL does a good job of ruling out attention
sample, 33% of respondents were from an upper problems, anxiety, or oppositionality (Eiraldi,
138 7 Parent and Teacher Rating Scales
Power, Karustis, & Goldstein, 2000) but would criterion-related and convergent validity of the
obviously need to be combined with other infor- CBCL, including internationally, it remains a
mation to rule in related diagnoses. The CBCL strong option as a parent-report rating scale,
Attention Problems scale along with the same despite its lack of a recent update.
scale from the Teacher Report Form (see below)
was also able to differentiate between children Interpretation Interpretation of the CBCL is
diagnosed with ADHD combined type and other bolstered by many articles by Achenbach and
clinic-referred children (Jarrett, Van Meter, colleagues devoted to its clinical use dating back
Youngstrom, Hilton, & Ollendick, 2018). to McConaughy and Achenbach’s (1989). More
Additional evidence indicates that factor struc- specifically, McConaughy and Achenbach (1989)
ture of clinician-completed CBCLs mirrors that provide an assessment methodology for the iden-
of parent-completed CBCLs and relates well to tification of severe emotional disturbance in the
other indicators of child functioning (Dutra, schools. Their multiaxial empirically based
Campbell, & Westen, 2004). That is, the CBCL assessment model proposes five axes for such
appears to cover important domains in a manner assessment situations: (1) parent reports
that is psychometrically sound and that can aid in (Achenbach, CBCL), (2) teacher reports
synthesizing information that a clinician has (Achenbach Teacher Report Form), (3) cognitive
gathered and/or observed. assessment, (4) physical assessment, and (5)
direct assessment of the child (i.e., Achenbach
The factor structure of the CBCL continues to Direct Observation Form and/or Youth
raise some conceptual issues regarding the con- Self-Report).
tent validity of the scales. For example, it is
unusual for depression and anxiety items to be For interpreting the CBCL specifically,
included on the same scale. In addition, the because there are no established validity scales
Attention Problems scale includes items that and because there are some scales that include
appear more indicative of hyperactivity and heterogeneous content, we recommend more
impulsivity than inattention (see Table 7.5). High attention to item-level interpretation than we
scores on these scales still require a great deal of have for other measures. That is, the clinician
clinical judgment as to what characteristics led to should pay close attention to scale elevations and
the high scores and should be the focus of further draw most conclusions at the scale level; how-
attention. ever, it would behoove practitioners to determine
As described above, the DSM-5-Oriented any concerning aspects of the parent’s item
scales rely on content validity (i.e., judgments by response style that would render the protocol
experts as to which items assess the relevant invalid. Additionally, interpretation should not
diagnostic criteria) to date. Validity studies as stop at the scale labels; rather, one should inspect
well as basic and applied research investigations the items that led to the scale elevations to deter-
on the syndrome scales of the CBCL are legion. mine the best way to describe the child’s difficul-
Overall evidence on its validity is good. In par- ties. Fortunately, the scoring methods available
ticular, the CBCL has been touted for its ability for the CBCL make linking item responses to
to screen for a variety of problem areas and in its scale elevations a straightforward process.
strong convergent and divergent validity (Scholte,
Van Berckelaer-Onnes, Van der Ploeg, 2008). A Strengths and Weaknesses The CBCL has
detailed summary of the validity evidence on the many strengths that continue to make it a popular
CBCL is not possible here. However, given the choice for clinicians. Noteworthy strengths
tremendous amount of research supporting the include:
Achenbach System of Empirically Based Assessment 139
1. A strong research base on the CBCL and its which is labeled the “Caregiver-Teacher Report
predecessors Form” (Achenbach & Rescorla, 2000). The item
2. Its popularity among professionals which
content of the TRF is very closely matched to the
facilitates communication about its results CBCL item content.
3. Several writings that provide interpretive
guidance above and beyond that provided by Content The school-age version of the TRF
the manual includes several background questions (e.g.,
4. Improved approach to assessing competence “How long have you known this pupil?”, “How
and availability of new DSM-Oriented scales well do you know him/her?”), a teacher’s rating
that are aligned to DSM-5 criteria of a child’s academic performance, and a four-
5. Some response flexibility in that parents are item screening of a child’s adaptive behavior
asked to elaborate on their answers to some with scoring on a 1–7 scale (e.g., “How hard is
items he/she working?”, “How appropriately is he/she
behaving?”, “How much is he/she learning?”,
Weaknesses of the CBCL include: “How happy is he/she?”). The preschool TRF
includes the same background questions but
1. Lack of validity scales which are now com- does not include the other items. These back-
mon among behavioral rating scales ground questions have associated norms and fall
2. Lack of close correspondence between the
under the “Adaptive Functioning” domain of the
empirically derived scales and common diag- TRF. The major portion of the TRF consists of
nostic criteria along with limited research on 100 items for the preschool version and 113
the new DSM-Oriented scales items for the school-age version. These items
3. Heterogeneous content within some scales describe problematic behaviors and emotions
that the teacher rates as being not true, some-
The CBCL continues to be a preferred choice what true/sometimes true, or very true/often
of many child clinicians due to its history of suc- true of the child. The problem behavior items
cessful use and popularity with researchers. The cover a broad array of both internalizing (e.g.,
continuing development of the CBCL database anxiety, depression, somatic complaints) and
bodes well for its future. The most recent ver- externalizing (antisocial behavior, aggression,
sions of the CBCL would benefit from research oppositionality) behaviors. As with the CBCL,
aimed at assessing the construct validity of the the TRF includes DSM-Oriented scales (6 for
new DSM-Oriented scales for DSM-5. Such the school-age version; 5 for the preschool
research efforts are necessary to define further version).
the degree of confidence that a clinician can place
on specific scales for making differential diag-
nostic decisions. Administration and Scoring The TRF takes
approximately 15–20 minutes to complete. The
instructions to the teacher are printed on the front
eacher Report: Teacher Report Form
T of the answer sheet. Scoring of the TRS can be
(TRF) done by hand using the TRF Profile Sheets with
separate profile sheets available for boys and
The Achenbach Teacher Report Form (TRF; girls. However, a computer-scoring system is
Achenbach & Rescorla, 2001) is designed to be available that greatly facilitates scoring by auto-
completed by teachers of children between the matically calculating raw scale scores and con-
ages of 6 and 18 for the school-age version and verting them to norm-referenced scores
between 1½ and 5 for the preschool version appropriate for the child’s age and gender.
140 7 Parent and Teacher Rating Scales
Both the Profile Sheets and the computer- represented Whites and African-Americans well
scoring program provide raw scores and norm- (i.e., 48% and 36%, respectively), with only 8%
referenced scores for several scales. As with the of the sample identified as Latino.
ASEBA CBCL, a Total Problem score, which is For both versions of the TRF, the normative
an overall indicator of a child’s classroom adjust- sample excluded children who had received men-
ment, and two broadband scores consisting of tal health or special education services within the
Internalizing and Externalizing behaviors are preceding 12 months. Therefore, as with the
included. These broad dimensions are further other rating scales in the ASEBA system, the
divided into the eight syndrome scales, which are sample should be considered a normal compari-
again similar to those included on the ASEBA son group, rather than one that is normative and
CBCL. representative of the general population.
The TRF allows for raw scores on all scales
to be converted to T-scores and percentile ranks Reliability Achenbach and Rescorla (2000,
based on the standardization sample. The 2001) provide three types of reliability informa-
T-scores are normalized standard scores. That tion on the TRF. Internal consistency estimates
is, the raw score distributions are transformed to were provided for the syndrome scales.
a normalized distribution. This procedure allows Coefficients indicated good internal consistency,
T-scores on all scales to have similar distribu- ranging from 0.72 to 0.95. For the preschool ver-
tions and corresponding percentiles based on the sion, coefficients were quite variable, ranging
assumptions of a normal distribution. However, from 0.52 for Somatic Complaints to 0.96 for the
it assumes that the dimensions assessed by the Aggressive Behaviors scale. Test-retest reliabil-
scale should be normally distributed in the gen- ity over an average of a 16-day interval is pre-
eral population, an assumption that is question- sented on a sample of 44 children in the age range
able because most children tend to cluster in the of 6–18. The test-retest coefficients were gener-
normal (e.g., fewer problems) end of the distri- ally high (i.e., .80s and higher), with the excep-
bution. Norm-referenced scores are based on tion of the Withdrawn/Depressed scale (r = 0.60)
gender- and age-specific norms. In addition, as and the Affective Problems scale (r = 0.62). Four-
with the ASEBA CBCL, T-scores on the TRF month test-retest reliability was variable, ranging
are truncated, such that the lowest score pro- from 0.31 (Affective Problems) to 0.72
vided is ≤50. (Hyperactivity-Impulsivity). The 8-day test-
retest reliability for the preschool version of the
Norming The norming sample for the school- TRF was somewhat variable, ranging from a
age version of the TRF consisted of 2319 chil- coefficient of 0.57 for Anxiety Problems to 0.91
dren, 72% of whom were White. Fourteen percent for Somatic Complaints (the scale with the low-
were identified as African-American, and 7% est internal consistency). Three-month test-retest
were identified as Latino; thus, both groups reliability for the preschool TRF was variable
appear to be underrepresented in this sample. The with coefficients ranging from 0.22 for Somatic
TRF normative sample appears to be geographi- Complaints to 0.71 for Emotionally Reactive.
cally representative. Thirty-eight percent of chil-
dren in this sample were from an upper SES Correlations between ratings from two differ-
background, 46% from a middle SES back- ent teachers are provided for 88 children
ground, and 16% from a low SES background. (Achenbach & Rescorla, 2001). The correlations
were modest across scales, with the mean coeffi-
For the preschool version, the norming sample cient for the full sample being 0.49 for the
consisted of 1192 children. This sample was geo- Competence scales and 0.60 for the syndrome
graphically diverse, but only 10% of the sample scales. Similar analyses for 102 preschoolers
came from a lower SES background. The sample revealed an overall mean interrater correlation
Achenbach System of Empirically Based Assessment 141
coefficient of 0.62, with a range of 0.21 (Somatic that used a more explicit guiding theory for scale
Complaints) to 0.78 (Aggressive Behavior). development. For example, the Attention
Problems scale consists of items traditionally
Validity The validity data reported in the manu- associated with inattention (e.g., difficulty con-
als (Achenbach & Rescorla, 2000, 2001) mainly centrating) but also includes items associated
focus on the ability of the scales to differentiate with immaturity, overactivity, poor school
non-referred from clinical samples within gen- achievement, and clumsiness. Therefore, it is
der. The TRF scales generally show such differ- imperative that the items that led to a clinical
ential validity. However, the Somatic Complaints scale elevation be viewed to understand the
scale of both versions does not appear to consis- meaning of the elevation. For example, a child
tently differentiate the two groups (Achenbach & may show an elevation on the Attention Problems
Rescorla, 2000, 2001). As with the CBCL, the scale because of problems with immaturity,
TRF has been the object of numerous follow-up clumsiness, or academic problems, or a child
studies pertaining to its criterion-related validity, may have an elevation due to problems of inat-
including across several countries (e.g., Ang tention and/or hyperactivity. However, because
et al., 2012; Bean, Mooijaart, Eurelings- of the unreliability of individual items, this item-
Bontekoe, & Sprinhoven, 2007; Kendall et al., level analysis should be conducted only when
2007). Research has also supported the predictive there is an elevation on the clinical scale. The
validity of the TRF at age 3 in predicting prob- Attention-Deficit Hyperactivity Problems DSM-
lems, especially externalizing problems, at age 5 Oriented scale can be quite helpful in this type of
(Kerr, Lunkenheimer, & Olson, 2007). The TRF, scenario as it is more closely aligned with charac-
like the CBCL, was based heavily on its well- teristics indicative of ADHD.
researched predecessor. Interpretation of the Thought Problems scale
on both the TRF and CBCL versions of the
ASEBA deserves several cautionary notes. This
Interpretation Independent research on the scale has an especially heterogeneous content,
reliability and validity of the adaptive function- consisting of items describing obsessions (e.g.,
ing component of the TRF is limited, although “Can’t get mind off certain thoughts”), compul-
one study found teacher ratings on the adaptive sions (e.g., “Repeats acts over and over”), fears
component of the TRF to be related to (e.g., “Fears certain animals, situations, or
neuropsychological indicators of good inhibitory places”), and psychotic behaviors (e.g., “Hears
control (Vuontela et al., 2013). Nevertheless, sounds or voices that aren’t there”), many of
interpretation of these scales should be done cau- which are fairly ambiguous (e.g., strange behav-
tiously, if at all. Subsequently, although it may be ior, strange ideas). For this scale, it should be
useful to next consider the TRF Total Problems apparent that item-level interpretation and inte-
score and composites, more specific information gration with other information collected during
can be gleaned from interpretations of the eight the assessment are crucial before drawing
syndrome scales and the six DSM-Oriented conclusions.
scales. The DSM-Oriented scales have been One final note is in order for interpreting TRF
updated to reflect behaviors/symptoms that align scales. Because the norm-referenced scores of
with DSM-5 criteria. the TRF are based on a normal sample and not a
normative sample, it is recommended that a more
The reliability of these scales (with the excep- conservative cutoff score be used than would be
tion of the Somatic Complaints scale) is good. the case for other rating scales. More specifically,
However, because the initial development of the one may want to limit conclusions to only those
TRF item pool was done in an attempt to be athe- over a T-score of 65 or even 70 (as opposed to
oretical, the item content of the TRF scales tends 60). However, any elevations, regardless of the
to be more heterogeneous than other rating scales degree of the elevation, should still be considered
142 7 Parent and Teacher Rating Scales
in conjunction with other assessment results similarly to the BASC and ASEBA systems in
(e.g., parent report, self-report, history, observa- that it includes a number of clinically relevant
tions, etc.). The sample case that follows gives a domains for which normative scores are derived.
brief example of an interpretive approach using The parent rating scale is designed for ages 6
both the CBCL and TRF. through 18. The Long Form contains 110 items,
and Short Form includes 45 items. There is also a
Strengths and Weaknesses The TRF remains ten-item Global Index Form. The Conners-3-P
one of the most widely used of the teacher- takes 10–20 minutes to complete, depending on
completed behavior rating scales. In addition to which form is used. The following discussion
its popularity and familiarity with a large num- will focus on the Long Form.
ber of professionals, the strengths of the TRF As noted in Table 7.1, we recommend the
include: Conners Comprehensive Behavior Rating Scales
(Conners, 2008b) for an assessment that covers
externalizing, internalizing, and academic issues
1. The large research literature on the TRF good as opposed to the Conners-3. However, as the
correspondence between the TRF and other information below indicates, the Conners-3 is
indicators of child functioning, particularly on unique in its detailed evaluation of ADHD and
externalizing behaviors. other externalizing issues.
2. The inclusion of DSM-Oriented scales aids
the clinician in interpreting teacher reports in Scale Content The Conners-3-P includes five
terms of diagnostic categories. empirically derived scales: Hyperactivity/
3. An expansive section that allows for teachers Impulsivity, Executive Functioning, Learning
to elaborate on strengths/competencies and Problems, Aggression, and Peer Relations. An
areas of concern. Inattention scale developed theoretically is also
available, as are five Symptom scales for ADHD
Some weaknesses of the TRF include: (i.e., Predominantly Inattentive Presentation,
Predominantly Hyperactive-Impulsive
1. An underrepresentation of Hispanics in the Presentation), ODD, and CD that have been
normative sample updated for DSM-5. The Conners-3-P includes
2. The exclusion of children with mental health screening items for depression and anxiety, as
or special education services in the normative well as impairment items for home, school, and
sample, indicating that such children are not social relationships. Like the BASC-3, the
represented Conners-3 includes critical items that may signal
3. The questionable reliability and validity of the the need for further follow-up. These critical
Somatic Complaints scale items are particularly geared toward severe con-
4. A relatively limited assessment of adaptive duct problem behaviors (e.g., uses a weapon, is
functioning cruel to animals). The Conners-3 also includes a
brief ADHD Index. This scale is based on items
that best differentiate ADHD from nonclinical
Conners, Third Edition (Conners-3; samples. As described in Chap. 6 for the
Conners, 2008a, 2014) Conners-3 SR, the Conners-3-P has three validity
scales: Positive Impression (or “fake good”),
Parent Rating Scale Negative Impression (“fake bad”), and the
Inconsistency Index. Two open-ended questions
The Conners-3 (Conners, 2008a) Parent Rating regarding other concerns and particular strengths/
Scale (Conners-3-P) is the most recent revision to skills are also included. Detailed information on
a widely known behavior rating scale system. the generation and selection of items is provided
The Conners Parent Rating Scale is designed in the Conners-3 manual (Conners, 2008a).
Conners, Third Edition 143
Administration and Scoring The Conners-3-P Reliability Internal consistency coefficients for
uses a four-choice response format where 0 = not the content scales of the Conners-3-P are all 0.80
at all true (never, seldom) and 3 = very much true and higher, and many are 0.90 and higher for the
(very often, very frequently). A Spanish transla- overall sample. The Peer Relations scale for
tion is available. Both handing scoring and com- females had somewhat lower coefficients (i.e.,
puter scoring are available. Raw scores are 0.72 for 6- to 9-year-olds; 0.78 for 10- to 13-year-
transformed to linear T-scores, meaning that each olds). Two- to four-week test-retest coefficients
scale maintains its natural distribution in the con- were good (i.e., all higher than 0.70). Interrater
version to norm-referenced scores. Separate reliability for the parent form was also good, with
norms are used for males and females, as is the adjusted rs all 0.74 and higher for the content
case for the other versions of the Conners-3. scales (see Conners, 2008a).
Norms are also computed by age.
reliability as well as convergence with the 41 items). Two ten-item Index Forms are also
ASEBA CBCL and a structured diagnostic inter- available. The following discussion will focus
view (Morales-Hidalgo, Hernández-Martínez, primarily on the Long Form of the Conners-
Vera, Voltas, & Canals, 2017). 3-T. As with the self-report and parent forms of
the Conners-3, we discuss the Conners-3-T
Interpretation As discussed in Chap. 6, because of its relatively unique focus on ADHD
Conners (2008a) provides a clear step-by-step and behavioral problems. The companion teacher
approach for interpreting ratings on the various rating scale from the Conners CBRS (Conners,
forms of the Conners-3. This approach involves 2008b) provides a more extensive assessment of
(a) examining the validity scales; (b) evaluating broader domains of functioning. Therefore, the
scale elevations; (c) examining the overall profile selection of one set of rating scales versus the
of scores (i.e., determining the constructs that other in the Conners family should be dictated by
seem to be represented across elevations); (d) the purpose of the evaluation.
item-level interpretation, including critical items
and screener items; and (e) integration with other Content The Conners-3-T has some item over-
assessment information. lap with the Conners-3-P, but there are also
unique items in each form. The same 4-point
response scale used for the self-report and parent-
Strengths and Weaknesses Some of the report versions of the Conners-3 is also used for
strengths of the Conners-3-P are: the teacher-report scale. As noted above, the
scales are the same as those for the parent rating
scale, including validity scales, impairment
1. Availability of complementary teacher and
items, and critical items.
self-report forms that provide a comprehen-
sive assessment of externalizing problems
2. Good reliability and validity evidence Administration and Scoring The Conners-3-T
3. Brevity of Short and Index Forms can be completed in 10–20 minutes or less if the
Short Form is used. The Conners-3-T has both
Some characteristics that may be considered hand scoring and computer scoring formats that
weaknesses are: allow for easy calculation of norm-referenced
scores. As with the self-report and parent-report
1. Limited assessment of internalizing problems forms, only gender-specific T-scores can be cal-
and adaptive functioning (an issue that is culated. The scores are linear T-scores and are
addressed through use of the Conners CBRS) based on each age group, which allows it to cap-
2. Uniform negative wording of items, which
ture potential variability in discrete developmen-
may result in a negative response set tal stages.
Teacher Rating Scale Norming The norming process for the Conners-
3-T was essentially the same as that used for the
The Teacher Rating Scale (Conners-3-T) in the Conners-3-P and Conners-3 SR. Specifically, the
Conners-3 system is very similar to the parent updated Conners-3-T norm sample consists of
rating scale. In fact, the teacher rating scale is for 3400 teachers from the United States and Canada.
youth of the same ages as the parent rating scale Selection of the norm sample involved 50 boys
and includes the same scales. The Long Form of and 50 girls at each age in the age range captured
the teacher rating scale is slightly longer than that by the Conners-3-T who matched 2000 US
of the parent rating scale (i.e., 115 items), Census data on ethnicity/race (Conners, 2014). In
whereas the Short Form is slightly shorter (i.e., addition, a clinical sample of 2143 ratings was
Conners, Third Edition 145
collected. A “reasonable spread” (Conners, 2014) demonstrated good differential validity in that
on parent education and geographic regions of scales were elevated for individuals from a clini-
the United States and Canada are reported. cal sample relative to a general sample and scale
scores tended to differ within the clinical sample
in intuitive ways. For example, ADHD scale
Reliability Internal consistency coefficients for scores tended to be higher for youth diagnosed
the teacher report version Conners-3 were quite with ADHD than for youth with other difficulties
high. Specifically, the coefficients for each of the who did not have ADHD diagnoses (see Conner,
content scales were 0.90 or higher. Two- to four- 2008a).
week test-retest reliability coefficients were also
good, ranging from 0.72 to 0.83 (see Conners, Additional validity information, particularly
2008a). Lastly, and perhaps particularly impor- for the Short Form and the Index Forms, is avail-
tantly for teacher ratings, interrater reliability able in the Conners-3 manual (Conners, 2008a).
coefficients for pairs of teacher raters were mod- Independent validity research has not yet been
erate to high. The Peer Relations scale had the conducted.
lowest adjusted coefficients (i.e., 0.52), whereas
the Hyperactivity/Impulsivity scale had the high- Interpretation At the very least, the Conners-
est coefficient (i.e., 0.77). It should be noted that 3-T appears to be useful as a screening for prob-
the lower coefficients from these analyses may lems in classroom adjustment, particularly in
reflect less-than-ideal rater agreement, or they terms of learning or externalizing problems, and
may reflect real differences in a child’s behavior as part of a comprehensive assessment battery.
from one classroom context to another. Additional The recommended approach for interpreting the
analyses, particularly in determining whether Conners-3-T mirrors that described for the
teacher agreement might change as a function of Conners-3 SR and Conners-3-P.
the child’s age, are needed.
2. Lack of independent research on reliability child may have problems. That is, the HSQ and
and validity. In particular, more research is SSQ were not designed to assess specific behav-
needed on teacher interrater reliability. iors but to assess situations in which problem
3. The normative sample is not quite as diverse behaviors can occur. Therefore, these measures
as that for the parent and self-report forms of provide an indication of the specific situations in
the Conners-3, yet it still is diverse in terms of which the child may demonstrate particular dif-
race/ethnicity. ficulty or impairment.
Both measures were designed in the same
manner. The respondent (parent or teacher) rates
Sample Impairment-Oriented whether or not the child has any problem in a
Scales given situation and then, if yes, rates the severity
of the problem on a 1–9 scale. These measures
As can be determined from the previous review, may be used with a variety of clinical problems,
omnibus rating scales can provide invaluable as the respondent can be directed to respond as to
information about a variety of domains of child whether the child “has problems” in the situa-
functioning. This information, however, tends to tions provided.
couch functioning in terms of severity and/or fre- The psychometric development of both mea-
quency of problems. Rating scales typically stop sures is limited. Normative information is avail-
short of providing an indication as to what extent able from Altepeter and Breen (1989) as well as
the problems interfere with the child’s function- Barkley and Edelbrock (1987) but is dated.
ing. Information on impairment is often left to However, norm-based comparisons may not
the clinician to infer based on interview or other represent the best use of these tools.
information. However, this information is no less Alternatively, gathering information about the
important for case conceptualization and treat- specific circumstances under which a child
ment planning. In addition to assessing for exhibits behavioral problems and the relative
impairment via structured or unstructured inter- severity of those problems may provide useful
views, one may employ an inventory to gather and unique information for treatment planning
such information in a time-efficient manner and and progress monitoring.
then follow-up accordingly. A brief discussion of The HSQ has demonstrated high internal con-
some such inventories follows. sistency and good 2- to 4-week test-retest reli-
ability, with the SSQ also demonstrating strong
internal consistency and moderate convergence
ome Situations Questionnaire (HSQ)
H with the HSQ (DuPaul & Barkley, 1992).
and School Situations Questionnaire Interrater agreement for the SSQ was tested in a
(SSQ) sample of 46 students ages 8–17. The correlation
between teachers was 0.68 for the number of
The content of the Home Situations Questionnaire problem situations and 0.72 for the mean severity
(HSQ; Barkley & Edelbrock, 1987) and the score (Danforth & DuPaul, 1996). Barkley and
School Situations Questionnaire (SSQ; Barkley Edelbrock (1987) reported numerous significant
& Edelbrock, 1987) is markedly different from correlations between the SSQ and rating scale
the other rating scales reviewed in this chapter. measures of externalizing behavior problems and
Rather than having items that describe different evidence that the SSQ differentiates children
types of child behaviors, these measures include with ADHD from children without
situations (e.g., “While playing alone,” “When ADHD. However, for both the HSQ and SSQ
visitors are in the home,” “During individual desk criterion-related validity, evidence is more diffi-
work,” “At recess,” “On the bus”) in which a cult to operationalize, as these measures have a
Sample Impairment-Oriented Scales 147
different focus than ratings of symptoms or prob- ment). A patent, teacher, or interviewer is asked
lems. Importantly, the HSQ or similar assessment to rate the child on this scale where deciles are
tools might provide incremental validity in the accompanied by a descriptor (e.g., 51–60, “some
clinical assessment of externalizing problems in noticeable problems”). Previous studies have
that they assess situations in which the child is demonstrated some evidence of reliability and
impaired to complement ratings of symptomatol- validity. A cut score is commonly used in studies
ogy (Barry, Pinard, Barry, Garland, & Lyman, of child psychopathology (e.g., CGAS 70 or
2011). In fact, the DSM-5 criteria for ODD now below identifies a clinical case). Although this
includes a severity specifier based on how many measure has not been updated for quite a while, it
situations the child shows the symptoms of the is still widely used in research, particularly to
disorder, based on research showing that this pre- monitor the effectiveness of intervention strate-
dicted severity of the disorder better than simply gies (e.g., Chia et al., 2013; Greaves & Salloum,
counting symptoms (Frick & Nigg, 2012). In 2015; Youngstrom et al., 2013).
addition, situations in which the child has diffi- The psychometric properties of the CGAS
culties, as indicated on the HSQ and SSQ, can have been well-studied (see review by Schorre &
assist the clinician in appropriately designing and Vandvik, 2004). Of course, the accuracy of
prioritizing intervention strategies. CGAS ratings (as is the case for all ratings)
A revised version of the HSQ was developed depends heavily on the rater’s knowledge of the
for use with children with autism spectrum disor- child’s functioning in a variety of spheres
der who have co-occurring behavioral problems (Weissman, Warner, & Fendrich, 1990). Can par-
(Chowdhury et al., 2016). Most revisions were ents, for example, provide useful ratings of their
made to include more demand-specific situations child’s school and peer functioning as is required
(e.g., when told to brush teeth, when told to put by the CGAS? Schorre and Vandvik (2004) call
on clothes) that may be particularly problematic for increased consistency in how clinicians assess
for youth with ASD who may have trouble with and then rate impairment. Certainly, consistency
such tasks or with transitioning into those tasks. in conceptualizing constructs such as attention
Support was demonstrated for a Socially problems or depression aids in communication
Inflexible factor and a Demand-Specific factor. and treatment planning. Such could also be the
One- to two-week test-retest reliability was mod- case for assessing impairment caused by these
erate, with convergent and divergent validity problems.
cited based on correlations between HSQ scores An additional consideration is whether the
and ratings of behavioral problems and adaptive best approaches to assessing impairment are
behavior by parents and clinicians (Chowdhury already embedded in rating scales such as those
et al., 2016). reviewed in this chapter. That is, clinical eleva-
tions on standard ratings scales may, then, pro-
vide an indicator of impairment. However, Mash
hild Global Assessment Scale
C and Hunsley (2005) concluded that “Assessments
(CGAS) of children and adolescents need to focus not
only on specific disorders and problems but also
Another example of an assessment of impairment on specific impairments that may occur in the
takes a different approach. The Child Global absence of a diagnosable disorder” (p. 368).
Assessment Scale (CGAS; Shaffer et al., 1983) is Therefore, it is quite likely, and quite important,
an adaptation of an adult scale designed to assess that measures of impairment will continue to see
overall level of impairment at home, in school, or more use in clinical practice and perhaps should
with friends. The scale extends from a low of 1 be an additional consideration during clinical
(extremely impaired) to a high of 100 (no impair- interviews (Chap. 13).
148 7 Parent and Teacher Rating Scales
Conclusions
define and detect the problems of focus. It
has the disadvantage of being utilized by Parent and teacher rating scales are now common
the very person or persons trying to imple- methods for assessing child problems. The qual-
ment and demonstrate the effectiveness of ity of parent and teacher rating scales has
their therapeutic strategies. Research has improved considerably. Scales routinely have
increasingly addressed the implications of national normative samples and provide expan-
this approach that could bias such assess- sive information about their reliability and valid-
ments (e.g., Lambert et al., 2003), but rela- ity. In short, rating scales are a time-efficient and
tively little research has been done to reliable method for obtaining assessment infor-
actually test the extent of such biases. Even mation from parents and teachers.
less is known about teacher assessment of We focused primarily on global scales that
changes in behavioral and emotional func- assess multiple domains of functioning because
tioning during and following interventions. the nature of childhood problems is such that
An exception would be single-case designs dysfunction in one domain is often associated
tracking behavioral changes resulting from with problems in other areas of functioning. Our
classroom interventions; however, many review of rating scales was not intended to be
times, these interventions are evaluated by exhaustive but was designed instead to focus on a
school psychologists or other mental health few commonly used scales to illustrate what we
professionals rather than teachers. Some feel are some crucial areas to consider in evaluat-
specific assessment tools (some of which ing scales for use in a clinical assessment. Also,
are reviewed in this text) that might be used our overview was not intended to replace a care-
in school settings to provide ongoing prog- ful reading of the technical manuals of these
ress monitoring are discussed by scales but to highlight some of the important fea-
Bohnenkamp, Glascoe, Gracey, Epstein, tures of the scales that might influence their use
and Benningfield (2015). Particular atten- in clinical assessments.
tion is needed to how well the chosen tools
assess the outcomes of interest for a par- Chapter Summary
ticular student.
We recognize that the preponderance of 1. Concerns about the validity of child self-
our attention is devoted to initial assess- reports in young children have made parent
ment of referral concerns rather than and teacher rating scales commonplace in
assessment of change. Nevertheless, an modern child assessment practice. These tools
increased research based on individualized tend to be a very efficient means of gathering
as well as large-scale approaches to evalu- clinically relevant information.
ating interventions is imperative. 2. Research has indicated that the construct
Discussions of evidence-based assessment being evaluated and the child’s developmental
must continue to include approaches for level influence ratings provided by parents
assessing treatment progress that are broad and teachers and even the usefulness of such
such as the Y-OQ/Y-OR-SR, and as well as ratings.
methods that can be modified for specific 3. The Behavior Assessment System for Children
cases and/or specific settings. (BASC-3) Parent Rating Scales (PRS) and
Teacher Rating Scales (TRS) have three forms
of similar items that span the preschool (2–5),
150 7 Parent and Teacher Rating Scales
child (6–11), and adolescent (12–21) age (a) The Conners-3 has a limited assessment
ranges. The PRS takes a broad sampling of a of anxiety, depression, and adaptive
child’s behavior in home and community set- functioning.
tings, whereas the TRS does the same for the (b) The Conner-3 system is new as of this
school setting. writing; therefore, research is limited.
(a) The PRS and TRS were developed using 6 . A new trend in parent and teacher rating
both rational/theoretical and empirical scales is the development of scales that focus
means in combination to construct the on the degree of impairment associated with
individual scales. a child’s problems in adjustment. Examples
(b) The BASC-3 measures include a rela-
of such measures are the Home Situations
tively comprehensive assessment of adap- Questionnaire (HSQ) and School Situations
tive functioning. Questionnaire (SSQ). More research is
(c) To date, research on the current versions needed on these instruments. Nevertheless,
of the PRS and TRS is limited. they represent a significant improvement
4. The Achenbach CBCL and TRF have long particularly since the lack of assessment of
been considered one of the premier rating impairment has been a historic weakness of
scale measures of child psychopathology. rating scales.
(a) The CBCL and TRF continue to be a pre- 7. Assessment of change is an important consid-
ferred choice of many child clinicians due eration in interventions with children, yet the
to its history of successful use and popu- availability of formal tools for evaluating
larity with researchers. change in behaviors/symptoms is quite limited.
(b) The CBCL and TRF include only gender- In many instances, the rating scales discussed
specific norms. in this chapter are not suitable for determining
5. The Conners-3-P and Conners-3-T are
short-term change, and the clinician should
designed for ages 6 through 18. They both consider idiosyncratic approaches to measur-
provide an extensive assessment of external- ing change that are meaningful for specific cli-
izing problems and have good reliability and ent behaviors/symptoms of concern.
validity evidence thus far.
Behavioral Observations
8
difficulty in assessing many behaviors that are In the following section, we discuss basic
very infrequent (e.g., cruelty to animals, halluci- issues in the development and use of observa-
nations, panic attacks) or by nature covert (e.g., tional systems. As was mentioned earlier in this
stealing, lying). In most cases, one would not chapter, many clinical assessors use informal
ethically want to contrive a situation that would observational techniques in their assessment bat-
prompt such behaviors, and the behaviors are tery without establishing a well-defined system.
often too infrequent to be observed naturally Unfortunately, the information obtained from
occurring in the child’s environment. such systems is difficult to interpret. Compared
A final issue in the use of behavior observa- to behavior rating scales, there are much fewer
tions is the fact that observations are limited to standardized observational systems that are read-
the assessment of overt behaviors. They do not ily available for clinical use that have well-
provide an effective means for assessing the cog- established psychometric properties. Therefore,
nitive, affective, and motivational components of the next section focuses on basic considerations
a child’s functioning (Mash & Terdal, 1988). in designing an observational system for one’s
This does not negate the importance of having a own clinical use. Following this discussion, some
good assessment of a child’s overt behavior in examples of observational systems that are com-
making diagnostic and treatment decisions. mercially available or that have been tested in
However, it has become increasingly clear that research are reviewed.
overt behavior is only one piece of a complex
puzzle. Research in several areas of child psycho-
pathology has supported the importance of intra- Basics of Observational Systems
psychic variables for both assessing (e.g., Frick,
2016) and treating (e.g., David-Ferdon & Kaslow, Defining Target Behaviors
2008) children and adolescents.
In summary, direct observations are affected The basic components of observational systems
by some factors that often preclude their use in can be broken down into the what, where, how,
many clinical settings and limit the usefulness of and by whom of the system. The first part of
the data obtained. We spent a great deal of time developing a system of direct observation
reviewing the factors that affect behavioral obser- involves defining what behaviors one wishes to
vations, not because of a bias against this form of observe. Defining the behaviors of interest first
assessment but because we have found that asses- involves deciding on the level of analysis one
sors sometimes ignore these issues. We feel that a wishes to use (Barrios, 1993). Specifically, the
clinical assessor should be aware of these issues level of analysis can be at the level of isolated
in deciding whether or not direct observations behaviors, at the level of constellations of behav-
should be included in an assessment battery and iors (syndromes), or at the level of interactions
should consider these issues when interpreting within a social unit. As an example of the social
observational data. However, these limitations unit level of analysis, many observational sys-
should not be considered any greater than those tems allow for the recording of how a child
associated with other assessment techniques, and behaves in response to parental behavior and how
the limitations must be weighed against some a parent responds to a child’s behavior (Bagner,
very important advantages of direct observation Harwood, & Eyberg, 2006). Also, the example
(e.g., elimination of reporter bias and ready trans- given previously demonstrates an observational
lation into environmental interventions). In sum- system that focuses on behaviors by fellow stu-
mary, direct observations can be an integral part dents and teachers that can influence the on-task
of many assessment batteries, but, as is the case behavior of a student with ADHD (Carroll et al.,
for all assessment techniques, they also have lim- 2006). Because these systems allow one to docu-
itations in the information they provide in ment events (stimuli) that elicit a behavior and
isolation. responses to the behavior that may help to
154 8 Behavioral Observations
Box 8.1 A Hypothetical Example of Simple A-B-C Observational System of an 8-Year-Old Boy (B)
After the level of analysis is chosen, one must The target behaviors in Table 8.1 are simply
operationally define what behaviors, what con- lists of behaviors from several domains that can
stellation of behaviors, or what antecedents/con- be assessed by observational systems. In order to
sequences will be observed within this window. be reliable, coding systems must have very
These definitions are made prior to beginning a explicit definitions of each behavior. This is nec-
direct observation and must be specified in objec- essary to reduce the possibility that the observer
tive and understandable terms in order to reduce will use some subjective and idiosyncratic defini-
the potential for bias and increase the reliability tion of the behavior, thereby making interpreta-
of the observation. Some examples of target tions from the observations difficult. Without such
behaviors used in observational systems are definitions, the primary advantage of direct obser-
described in Table 8.1. vations, objectivity, is severely compromised.
1. The voice quality of the word or phrase Source: Summarized from the manual
is the primary distinguishing element for the Dyadic Parent-Child Interaction
for coding whine. Coding System (Eyberg & Robinson,
2. Each whined sentence constitutes a sep- 1982) with the authors’ permission
arate whine. Whined phrases separated
from one another by a pause of 2 sec-
onds or longer are coded as separate
whines. Setting
visits to observe a child or adolescent interacting social competence of a child’s behavior in each
with his or her parents. Also, for some behaviors, of the imagined situations. An example of one of
there may not be a way of obtaining unobtrusive the role-play situations used in this study is
observations in a child’s natural environment. As a included in Box 8.4.
result, the level of reactivity would be so high that
the data would be meaningless. In addition to
these more practical considerations, sometimes Box 8.4 A Role-Play Situation from the
there is a need to exert more control over the situ- Dodge Study of Social Competence
ation than is possible in a natural setting. For
example, one may wish to observe a child’s activ- Situation #2
ity level in a free play situation by determining Let’s pretend that I’m playing blocks with
how many times a child passes from one part of a some of my friends after lunch. We’re
room to another. To code this reliably, one can building a really neat house. You come in
divide the room into sections with tape and then the schoolroom and see us. Pretend that
code the number of times a child crosses over a you really want to play blocks with us.
tape divider (Milich, 1984). This type of control What do you do and say?
(e.g., dividing the playroom into grids) may not be
feasible in a child’s natural environment. Scoring:
For these reasons, it is sometimes necessary or
desirable to conduct analog observations in a (a) Was the child role playing?
laboratory or clinic. Analog refers to the creation (b) How competent was the response?
of a contrived setting that approximates the Score:
natural environment. Dividing a clinic playroom 8 – Complimentary or evaluative
into grids to observe a child’s activity level is one remark with a request to play.
example of an analog setting. However, the key (e.g., Boy, that’s neat. Can I
to these observations is how well the analog situ- play?)
ation approximates the natural environment. 6 – A simple request to play: I’d
Staying with our example, it would be imperative ask, “May I please play?”
that the playroom be similar to a play area that a 4 – Rhetorical question or evalua-
child would be in outside of the clinic (e.g., with tive remark: “What are you
age-appropriate toys available). There are many doing?” or “That’s neat.”
other examples of analog settings for behavioral 2 – Suggestion for different activ-
observations, but each involves the basic compo- ity: Want to play a game?
nent of simulating a child’s natural environment 0 – Aggressive responses such as I’d
in a clinic setting. knock the blocks over, subject
Sometimes it is not feasible to have the clinic would say nothing or sit down at
setting approximate the natural setting. In these desk without speaking: subject
cases, children may be asked to imagine them- didn’t know what to do, or didn’t
selves in a situation, and their behavior is answer.
observed in this role-play situation. An area in
0 2 4 6 8
which role-play observations have been fre- Very Somewhat Neither Somewhat Very
quently used is the assessment of children’s incompetent in- competent competent competent
social competence (e.g., Bornstein, Bellack, & competent nor
Hersen, 1977; Dodge, McClaskey, & Feldman, incompetent
the degree of social withdrawal of a child, one Table 8.2 Steps in training and monitoring observers
may divide an observational period into Step Description
60-second intervals and record whether or not a Orientation Informing observers of the importance
child was engaged in interactions with other chil- of objective assessment in the
understanding and treating of
dren at the end of each interval. childhood disorders. Informing
observers of their duties and
responsibilities, in particular their
Selecting the Observer independent, unbiased, and faithful
recording of the behavior of interest
Education Instructing observers in the response
After determining what behaviors will be definitions and recording scheme
observed, where the observations will take place, through the use of written materials,
and how the observations will be conducted, one filmed illustrations, and live
demonstrations
still must determine who is best suited to conduct
Evaluation Assessment of observers’ knowledge
the observation. Having someone who is in the of the response definitions, coding
child’s natural setting (e.g., teacher or parent) system, and recording scheme through
conduct the observation is often useful in natural- the use of written and oral
istic observations because it helps to maintain examinations. Representation of
materials until observers are
unobtrusiveness. However, it is often difficult to thoroughly acquainted with all aspects
teach people in the child’s natural environment of tracking and recording of the
how to use a coding system and to ensure that it behaviors of interest
is being used appropriately. Application Graduated implementation of the
observation system across a range of
In order to exert more control over the obser-
situations, beginning with analog ones
vational methodology, many observational sys- and ending with actual setting of
tems require rigorously trained and monitored interest. Transition from one situation
observers. Barrios (1993) provides a summary to the next contingent upon observers
achieving a criterion level of
of the steps required in training and monitoring
agreement and accuracy
observers (see Table 8.2). As one can see from Recalibration Assessment of the accuracy and
this summary, using specially trained observers agreement of observers’ recordings in
is quite costly. Such stringent methodology is the setting of interest. Identification
feasible only when a large number of children and correction of any breakdowns in
the fidelity of observers’ recordings
are being observed with the same observational
Termination Questioning observers as to the merits
method. Therefore, this methodology may not of the observation system. Informing
be optimal in many clinical settings. One type observers of their contributions to the
of observation that is frequently used in clinical understanding and treating of the
behaviors of interest. Reminding
assessments involves training a person to
observers of the need to maintain
observe his or her own behavior. The same steps confidentiality
of selecting target behaviors, determining the Reproduced with permission from Barrios, B. A. (1993).
setting for the observation, and determining Direct observation. In T. H. Ollendick and M. Hersen
how the target behaviors will be recorded are (Eds.), Handbook of child and adolescent assessment
followed. However, in self-monitoring, the child (pp. 140–164). Boston, MA: Allyn & Bacon
is trained to record his or her own behavior.
Although self-monitoring has been used largely One limitation in these self-monitoring pro-
with adults, children have used self-monitoring cedures is the need for the child to keep some
systems to monitor such diverse behaviors as record of their observations (i.e., diaries), often
classroom attending, class attendance, talking noting the behaviors as soon as they occur to
out in class, room cleaning, aggression, and avoid any inaccuracies resulting from having the
inappropriate verbalizations, substance use, and child recall their behavior and emotions. For
frequency of intense mood states (Bower, 1981; example, having a child recall their mood states
Mash & Terdal, 1988). at a previous point in time could be influenced
160 8 Behavioral Observations
by his or her current mood and/or by events that child’s self-monitoring of their emotions and
happened after the event that is being recorded behaviors. However, with all self-monitoring
(e.g., a classmate apologized for being mean to procedures, including those using electronic dia-
the child; Bower, 1981). Further, the child may ries, research suggests that their accuracy is
forget to code their emotions or behaviors during dependent on the level of training given to the
a certain time and thus engage in “backfilling,” child, the use of a clear and simple observational
whereby he or she goes back and fills in diary system, the use of an outside monitor to track the
entries all at one time, again leading to poten- accuracy of their recording, and the use of some
tially inaccurate observations (Takarangi, Garry, system for reinforcing for the consistency and
& Loftus, 2006). accuracy of their recording (Keller, 1986). Also,
To overcome some of these limitation in paper while self-monitoring techniques are cost-
diaries, an increasingly popular method of self- effective and less intrusive than many other
monitoring is the use of electronic diaries, which forms of behavioral observation, research also
capitalizes on the newer and almost universally indicates that children change their behavior as
availability of portable electronic devices, most they become more aware of it through self-mon-
notably cell phones (Piasecki, Hufford, Sohan, & itoring (Keller, 1986; Shapiro, 1987). Whereas
Trull, 2007). Specifically, apps on the child’s cell this change in behavior may be a beneficial
phone can be used that provide a signal to the aspect of self-monitoring in a treatment program,
child when to code the target behaviors in real this reactivity limits the usefulness of self-moni-
time (i.e., when it is actually occurring) and not toring as a means of obtaining objective informa-
allow the child to “backfill” data from observa- tion on a child’s behavior for assessment
tions that may be missed. Further, these apps can purposes.
relay information to the assessor on the child’s
level of compliance, so that problems can be
addressed quickly. Examples of Observational Systems
One example of this electronic diary technique
was used by Whalen et al. (2006), who had 52 In contrast to behavior rating scales, there are few
children (mean age of 10.58) report on their observational systems that are widely used, have
moods, behaviors, and social contexts every standardized procedures, and are readily avail-
30 minutes during nonschool hours. The children able for clinical use. In this section, we review
carried a portable electronic device (PED) that some notable exceptions. The goal of this over-
beeped every 30 minutes to signal it was time to view is not to provide an exhaustive list of obser-
respond to certain questions about their emotions vational systems but to provide a carefully
and behaviors. Twenty-seven of the children were selected list of observational techniques that vary
diagnosed with ADHD, and the results showed in terms of target behaviors, settings of observa-
that these children had more behavioral problems, tion, method of data collection, and degree of
negative mood, and conflict with parents. training needed to reliably use the observational
Similarly, Suveg, Payne, Thomassin, and Jacob system. We feel that, even if one does not choose
(2010) used PEDs in studying the emotional states to use one of the specific systems discussed here,
of 38 children ages 7–12. Children were prompted these systems provide concrete examples of some
to indicate the intensity of their current emotions of the issues discussed in this chapter and there-
four times per day for 1 week. Sixty percent of the fore can serve as a guide for the development and
prompts were entered as intended; these momen- use of other observational systems. Also, some
tary ratings were associated with several potential additional observational systems that focus on a
predictors of emotional problems. specific area of adjustment are reviewed in other
Thus, these electronic diaries seem to a prom- chapters.
ising method for enhancing the accuracy of a
Examples of Observational Systems 161
scale (0 = “no occurrence”; 3 = “definite that is designed for use in a classroom setting. It
occurrence with severe intensity or 3 or more is part of the comprehensive BASC-3 system,
minutes duration”). Items are rated by the exam- which includes parent and teacher behavior rat-
iner immediately after the testing session. Like ing scales and a child self-report form, all of
the DOF, the TOF has a strong overlap in items which have been discussed in previous chapters.
with other measures in the ASEBA system. Thus, The SOS defines 65 specific target behaviors that
a Total Problem, an Internalizing, and an are grouped into 13 categories (4 categories of
Externalizing composite can all be obtained from positive/adaptive behaviors and 9 categories of
the TOF. Also, it includes five narrow band problem behaviors). The 13 categories and exam-
scales: Withdrawn/Depressed, Language/ ples of target behaviors in each category are pro-
Thought Problems, Anxious, Oppositional, and vided in Table 8.3.
Attentional Problems.
The TOF has a normative sample of 3943 chil- Table 8.3 Behavioral categories from the Student
dren between the ages of 2 and 18, most of which Observation System of the BASC-3
were obtained during the standardization of the Example of specific
Stanford-Binet Intelligence Scales-5th Edition BASC-3 category/definition behaviors
Response to teacher/lesson Follows directions
(Roid, 2003). In general, the TOF scales shows
(appropriate academic behaviors Raises hand
good test-retest reliability (.53–.87) over a period involving teacher or class) Contributes to
of 10 days, adequate interrater reliability (.42– class discussion
.73), and good internal consistency (.74–.94) Waits for help on
assignment
(McConaughy, 2005). Scores on the TOF are
Peer interaction (appropriate Plays with other
moderately correlated with corresponding parent interactions with other students) students
completed ASEBA (.27–.43) and teacher com- Interacts in
pleted ASEBA (.26–.38) scales (McConaughy, friendly manner
2005). Also, the TOF scales have differentiated Shakes hand with
other student
children with ADHD from normal control chil- Converses with
dren (McConaughy, 2005). others in
In summary, the TOF and DOF are both time- discussion
efficient observational systems that require mini- Work on school subjects Does seatwork
mal observer training and fit into a multi-method (appropriate academic behaviors Works at
that student engages in alone) blackboard
assessment system. Both observational systems Works at computer
have information showing some basic levels of Transition movement Puts on/takes off
reliability. They both also provide norm- (appropriate non-disruptive coat
referenced scores, although the sample for the behaviors while moving from Gets book
one activity to another) Sharpens pencil
DOF norms is very limited. Also, both systems Walks in line
have proven to be useful for discriminating nor- Returns material
mal from clinic-referred children. However, their used in class
ability to differentiate within children with emo- Inappropriate movement Fidgeting in seat
tional and behavioral problems is less clear. (inappropriate motor behaviors Passing notes
that are unrelated to classroom Running around
work) classroom
Sitting/standing on
ehavioral Assessment System
B top of desk
for Children-Student Observation Inattention (inattentive Daydreaming
behaviors that are not disruptive) Doodling
System Looking around
room
The BASC-3-Student Observation System (SOS; Fiddling with
Reynolds & Kamphaus, 2015) is a commercially objects/fingers
available, short (15-minute) observational system (continued)
Examples of Observational Systems 163
(e.g. animal, dark, stranger, heights, blood), then b ehavioral components (i.e., a child’s avoidance
requiring the child to approach the feared stimuli of a feared stimulus) of childhood anxiety.
in graduated steps.
BATs provide explicit and objective criteria
for observing a child’s behavioral reaction to the Conflict Negotiation Task
feared stimulus, such as how closely the stimulus
is approached, the number of steps in the gradual The Conflict Negotiation Task was designed to
approach that are taken, or the time spent touch- assess peer interactions, especially those interac-
ing or handling the phobic stimulus. In quantify- tions that may be associated with childhood
ing these responses to anxiety-producing stimuli, depression (Rudolph, Hammen, & Burge, 1994).
BATs provide a measure of the severity of a Children are observed with an unfamiliar partner
child’s anxiety and can help document changes of the same age and gender. The system uses a
brought about by treatment (Vasay & Lonigan, task involving three points of potential conflict of
2000). interest between the child and his or her partner.
Southam-Gerow and Chorpita (2007) provide First, the child dyads are placed in a situation in
a good summary of the advantages and disadvan- which they are to build structures with colored
tages of BATs. They describe the primary disad- blocks to match either of two models. They are
vantage of BATs as the absence of a single informed that whoever constructs an identical
standardized BAT. Instead, there have been model would win a prize. The dyad is given a set
numerous different BATs developed that vary of blocks to share, but the number of blocks is
widely on the number of steps in the graduated only sufficient to build one complete model.
approach, the types of instructions given to the Second, after 10 minutes of observation, the
child, and how the feared stimulus is presented. dyad is asked to decide on how to distribute two
As a result, it is impossible to compare the find- prizes of unequal value. Third, the dyad partici-
ings across studies, and therefore, it is difficult to pates in a 5-minute interview, and the child who
develop a significant body of knowledge on the received the less valuable prize is allowed to
reliability, validity, and normative base for any of choose a new one.
the BATs. In addition, a hallmark of the BATs is The interactions during this task are coded by
the rigorous control over how the feared object is trained observers. Using an event recording sys-
presented to the child and how the child tem, behaviors are rated on a 7-point scale
approaches it. This degree of control may prevent (0 = not at all present; 4 = moderately present;
the behavior observed in the contrived setting 7 = to a large degree present) and all ratings are
from generalizing to the child’s natural environ- scaled such that high scores indicate more nega-
ment. One final limitation of BATs is that they are tive peer interactions. The behaviors are grouped
most commonly used to assess specific fears and into four composites. Two composites are related
are more difficult to design for assessing more to broad dimensions of social behavior displayed
generalized anxiety (Vasay & Lonigan, 2000). by the child being assessed. Conflict-resolution
On the positive side, however, BATs are rela- competence includes persistence in problem-
tively simple and time-efficient in their adminis- solving efforts, positive assertiveness, positive
tration. Many of the BATs have been shown to conflict management, and general social compe-
have good inter-observer agreement with mini- tence. Emotional regulation includes conflict
mal observer training and to be sensitive to treat- exacerbation, positive affect, and negative affect.
ment effects (Barrios & Hartmann, 1988; The third composite consists of a dyadic quality
Southam-Gerow & Chorpita, 2007). Also, scores code based on ratings of conflict or friction within
from BATs are correlated with subjective ratings the dyad, collaboration, problem-solving compe-
of fear and with phobia diagnoses (Vasay & tence of the dyad, and mutuality/reciprocity.
Lonigan, 2000). Most importantly, they provide Finally, a peer response code is based on the rat-
one of the only methods of assessing the ings of the peer’s behavior toward the assessed
Examples of Observational Systems 165
child, including general response valence (nega- 6-minute period and then coded a second child
tive or positive) to the target child, discomfort according to a prearranged schedule.
and embarrassment in response to the target There were 18 target behaviors of five types
child, and emotional state at the end of the that were defined for the observation system (see
interaction. Table 8.4). A complex event recording system
Rudolph et al. (1994) reported on the use of was used for the observations. Each time a target
this system in a sample of 36 children (20 girls behavior was observed, the observer coded the
and 16 boys) between the ages of 7 and 13. The time, the context (structured vs. unstructured),
four composites from this observational system the target behavior observed, and the peer target
showed very high correlations between raters (number of child). Observers received extensive
(.88–.92), and the behaviors within each compos- training over a 4-week period and, with this train-
ite were highly intercorrelated (.82–.97). Most ing, the observational codes showed quite high
importantly, when the 36 children were divided inter-observer agreement (Dodge, 1983). Across
into those high on a measure of depression and the 18 target behaviors, 15 behaviors showed
those low on this measure, depressed children inter-observer agreement of 65% or better. The
were rated as significantly less competent on all three behaviors that showed poor inter-observer
four of the composites. Specifically, children agreement were watching peers, norm-setting
high on depression were observed to be signifi- statements, and supportive statements.
cantly worse in their conflict-resolution compe-
tence, in their emotional regulation, in mutuality
Table 8.4 Target behaviors in the Dodge Observational
and cooperation during the interaction, and in the System of Peer Interactions
negative valence placed on the interactions by Behavior category Target behaviors
their partners (e.g., their partner being more Solitary active Solitary play
uncomfortable in the interactions). The authors Watching peers
point out that these data support the contention On-task behavior
that children who score high on depression have Off-task behavior
Interactive play Cooperative play
significant interpersonal difficulties. Furthermore,
Aggressive play
the findings suggest that observational systems of Inappropriate play (e.g.,
children’s interactions with peers can be useful standing on table)
for assessing important aspects of these social Verbalizations Social conversations with
difficulties. peers
Norm-setting statements (e.g.,
rule making)
Hostile verbalizations (e.g.,
odge’s Observation of Peer
D insults, threats)
Interactions Supportive statements (e.g.,
compliments, offers of help)
Exclusions of peers from play
Dodge (1983) developed a direct observation Extraneous verbalization (e.g.,
system that also assesses several components of laughs, cheers)
peer interactions. However, this observation Physical contact with Hits
peers Object possession (e.g.,
system focuses on behaviors associated with
grabbing an object from peer)
acceptance in a peer group. Dodge developed Physically affectionate
his system in sample of 5-, 6-, 7-, and 8-year- behavior (e.g., holding hands,
old boys. Children were observed in 60-minute hugging)
play groups of eight boys each by three observ- Interactions with Social conversation with adult
adult group leader leader
ers who were stationed behind a one-way mir- Reprimanded by adult group
ror. The boys wore numbered T-shirts to aid in leader
quick identification by the observer. Each Source: Dodge, K. A. (1983). Behavioral antecedents of
observer coded the behaviors of one boy for a peer social status. Child Development, 54, 1386–1399
166 8 Behavioral Observations
Family Interaction Coding System of the interactions between a child and other fam-
ily members. The behavior of the child and the
One of the most common uses of behavioral obser- person(s) with whom the child interacts is coded
vations is to observe parent-child interactions, in sequence. After initial coding, many of the
especially those associated with childhood con- child’s behaviors are summarized in a rate-per-
duct problems (Frick & Loney, 2000). For exam- minute variable that combines both frequency
ple, Patterson and colleagues at the Oregon Social and duration of the behavior. However, the most
Learning Center developed a direct observational frequently used score from the FICS is the Total
system designed to assess children’s conduct prob- Aversive Behavior (TAB) score, which is a sum
lems in the home and to assess the interactional of the number of aversive behaviors that occurred
patterns in which the conduct problems are often during the observational session.
embedded. The Family Interaction Coding System Patterson (1982) describes a moderate level of
(FICS; Patterson, 1982) is composed of 29 code inter-observer agreement for most code catego-
categories that include both child behaviors and ries of the FICS, with the categories of Negativism
parental reactions to the child behaviors. These and Self-Stimulation showing the most question-
categories are summarized in Table 8.5. The goal able levels of agreement. One week test-retest
of the FICS was to observe children interacting reliability of the TAB was studied in a sample of
with family members in natural home settings. 27 boys and was found to be quite high (.78). The
However, as described by Patterson (1982), several TAB was also found to discriminate between
restrictions had to be made in the home for the families of children referred for behavior prob-
observational sessions. Specifically, to use the lems and non-referred families and has proven to
FICS, all family members must be present during be sensitive to family-focused treatment for chil-
the prearranged observation times with no guests dren’s conduct problems (Patterson, 1982).
present, and the family is limited to being present Although most of the individual code categories
in two rooms of the house. There can be no cell of the FICS can be coded consistently by two
phone calls out (only brief answers to incoming observers, psychometric information is generally
calls) and no TV. Finally, there is to be no talking limited to the global index of aversive behavior,
to the observers during coding. the TAB.
The FICS was designed to have data coded In summary, the FICS is an example of an
continuously and to provide a sequential account observation system designed to assess a child’s
behavior in the home environment and to assess
Table 8.5 Target behaviors in the Family Interaction parent-child interactional patterns in which the
Coding System behavior is embedded. The FICS is generally
Approval High ratea Physical most useful for younger children (10 and under).
negativea Also, most of the psychometric information
Attention Humiliatea Physical available for the FICS is for the aversive behav-
positive
iors assessed by the TAB. This is not as severe a
Command Ignorea Receive
Command Laugh Self-
limitation as one might think, however, given that
negativea stimulation these aversive behaviors have proven to be quite
Compliance Noncompliancea Talk important to understanding and treating children
Crya Negativisma Teasea with conduct problems (Patterson, 1982).
Disapprovala Normative Touch
Dependencya No response Whinea
Destructivenessa Play Yella evised Edition of the School
R
Reproduced with author’s permission from Patterson, Observation Coding System
G. R. (1982). Coerceive family process. Eugene, OR:
Castalia
a
Denotes aversive behaviors that are included in the Total The Revised Edition of the School Observation
Aversive Behavior (TAB) score Coding System (REDSOCS; Jacobs et al., 2000)
Examples of Observational Systems 167
is an interval coding system designed to assess main REDSOCS codes (i.e., inappropriate behav-
children’s externalizing behavior in the class- ior, off-task behavior) were shown to be sensitive
room. The REDSOCS has been primarily used to to treatment effects in a sample of children ages
assess the classroom behavior of young children 3–6.
(under the age of 6). Behaviors are coded in
10-second intervals using the eight categories
summarized in Table 8.6. tructured Observation of Academic
S
There are three main codes made by the and Play Settings
REDSOCS. First, the number of intervals in
which the child displayed any of the 12 inappro- Milich, Roberts, and colleagues (Milich, 1984;
priate behaviors (see Table 8.6) is coded, as well Milich, Loney, & Landau, 1982; Roberts, Ray, &
as the number of intervals in which no inappro- Roberts, 1984) developed an observational sys-
priate behavior was displayed for the entire tem (Structured Observation of Academic and
10-second interval (i.e., appropriate behavior). Play Settings [SOAPS]) to assess behaviors asso-
Second, noncompliant behavior is coded when ciated with ADHD in a clinic playroom analog
the child does not obey a command made by the setting. In this system a clinic playroom is
teacher within a 5-second period following the designed with age-appropriate toys, 4 tables, and
command. Otherwise, the child is coded as show- a floor divided into 16 equal squares by black
ing compliant behavior. Third, off-task behavior tape. The child is placed in two situations. Free
is coded if at any point during the 10-second play involves the child being placed alone in the
interval, the child does not attend to the material room and allowed to play freely with the toys.
or task assigned by the teacher. Otherwise, the The Restricted Academic Playroom Situation
behavior is coded as on-task for the interval. involves the child being requested (1) to remain
Support for the REDSOCS comes from a seated, (2) to complete a series of academic tasks,
study showing adequate interrater reliability for and (3) not to play with any of the toys.
the codes and the ability of the codes to Each observational situation lasts for 15 min-
differentiate clinic-referred children with behav- utes. A combination of event recording and time
ior problems (ages 3–6 years of age) from non- sampling is used in this observation system.
referred children in the same classrooms (Jacobs Event recording is used to determine the total
et al., 2000). Of particular note, at least two of the number of grids crossed for the entire observa-
tional period. That is, the number of times that a
Table 8.6 Behaviors coded in the Revised Edition of the child moves completely from one square of the
School Observation Coding System (REDSOCS) divided room into another is counted. Event
Comply On-task recording is also used to determine the number of
Appropriate Noncomply Off-task times the child shifts his or her attention from one
Inappropriate No command Not task to another during the entire observational
Whine given applicable
Tantrum period. A 5-second time-sampling procedure is
Destructive used to observe other target behaviors. These
Disruptive include the proportion of 5-second intervals that
Talks out the child is out of his or her seat, fidgeting, noisy,
Cheating
and on task. In addition, a 5-second time sam-
Self- pling is used to determine the number of intervals
stimulation that the child was observed touching forbidden
Demanding toys. Also during the academic task, the number
Yell
Cry of items completed is recorded.
Negativism This observational system is useful clinical
Out of area assessments of ADHD for a number of reasons.
Source: Bagner, Boggs, and Eyberg (2010) First, it is a relatively easy observational system.
168 8 Behavioral Observations
(d) Who—determining who should observe conjunction with other components of the
the target behaviors ASEBA system.
5. In defining target behaviors, one must con- (a) The DOF allows for a direct observation
sider the level at which behaviors will be of a child’s classroom behavior during
defined and then clearly define the behaviors three to six 10-minute observational
to be observed. periods.
6. Observations conducted in a child’s natural (b) The TOF allows for a direct observation
environment have greater ecological validity of a child’s behavior during individual
but allow less control over the observational standardized testing, and it has strong
setting than observations conducted in a lab- normative data.
oratory or analog setting. (c) Both the DOF and TOF have evidence
7. There are three basic ways in which behav- supporting their reliability and for differ-
iors can be recorded in an observational entiating referred from non-problem
system: children.
(a) Event recording—the number of times a 10. The BASC-3-SOS is an observational sys-
behavior occurred is recorded. tem used to assess classroom behavior that
(b) Duration recording—the length of time can be integrated with other assessment
from the initiation to the desistance of a components of the BASC-3 system.
behavior is recorded. (a) It specifies 65 target behaviors and uses
(c) Time sampling—behaviors are recorded a momentary time-sampling procedure
as to whether or not they have occurred for observations.
during preset time intervals. (b) There is no normative information nor is
8. Observations can be conducted by outside there any information on the reliability
observers, people in a child’s environment or validity of this observational system.
(e.g., parents, teachers), or by the child or 11. BATs allow one to observe a child’s response
adolescent himself or herself through the use to anxiety-provoking stimuli.
of electronic diaries. 12. Observational systems have also been devel-
9. The ASEBA-DOF and ASEBA-TOF are two oped to assess peer interactions, parent-child
observational systems that can be used in interactions, teacher-child interactions, and
behaviors associated with ADHD.
Peer-Referenced Assessment
9
The assertion that peer perceptions cannot be One such consideration could be the intrusive-
obtained by other methods comes from the ness of peer ratings. Peer-referenced assessment
meta-analysis conducted by Achenbach, typically involves the use of many peer raters,
Mc-Conaughy, and Howell (1987). These making it more intrusive than many other assess-
authors calculated the average correlations ment techniques. For example, asking a teacher
across studies between peer reports of social, to complete a behavior rating scale adds only one
emotional, or behavioral functioning with the person to the assessment process. In contrast,
reports of teachers and with the child’s self- having a child’s class participate in a sociometric
report. Across 23 studies reviewed, the average exercise involves 15–30 additional people in the
correlation across all psychological domains was assessment process.
.44 between peer and teacher ratings, with the In addition to the sheer number of people that
correlation being somewhat higher for behav- must be involved, the people involved are typi-
ioral (.47) problems than for emotional (.35) cally children who may not appreciate the need
problems. Similarly, across 20 studies in which for discretion and confidentiality. Therefore, spe-
the correlation between peer ratings and the cial precautions to limit the intrusiveness of this
child’s self-report of adjustment was determined, intervention are essential. There are several nec-
the average correlation was .26, again with the essary precautions for the use of any peer-
correlation for behavioral difficulties (.44) being referenced strategy. First, it is important that both
somewhat higher than for emotional difficulties the parent and the child being assessed are clearly
(.31). These data suggest that there are substan- informed and give their consent to the peer-
tial differences between how peers rate children, referenced assessment (Gresham & Little, 1993).
how teachers rate children, and how children Second, peer-referenced techniques should be
describe themselves. Therefore, to understand a designed to ensure that the child’s classmates do
child’s peer network that is heavily influenced by not know that the assessment is focused on any
peer perceptions, these perceptions must be one individual (see Box 9.2 for an example of
assessed directly. instructions provided in a sociometric exercise).
Finally, administration of peer-referenced tech-
niques should be carefully monitored to ensure
Ethics of Peer-Referenced Strategies that answers are not shared, and those involved in
the assessment should be instructed on the impor-
Despite research on the unique and important tance of the confidentiality of their responses
contributions that peer-referenced techniques can after the assessment (McConnell & Odom, 1986).
make to clinical assessments, these techniques
are probably one of the least used assessment
techniques of any reviewed in this book. The fail-
ure to include peer-referenced techniques in Box 9.2 Instructions for Two Peer-
many assessment batteries could be due in part to Referenced Techniques
the paucity of standardized, well-normed, and I.
Asessor-Administered Exercise with
readily available assessment procedures. This Unlimited-Choice Format (Carlso,
exclusion could also be due to the time-consuming Lahey, & Neeper, 1984).
nature of many of the peer-referenced techniques
used in the research literature. However, as we A. After discussing the importance of confi-
discuss later in this chapter, there are several rela- dentiality, sheets are provided to each
tively time-efficient peer-referenced procedures child with the class role and each of 32
that have been used extensively in research, nomination categories. Then read the fol-
which could add to a clinical assessment battery. lowing: “Now we’re ready to begin. We
The low frequency of use then is likely the result (continued)
of other considerations.
174 9 Peer-Referenced Assessment
more of their peers who display a certain correlations between nominations obtained in the
characteristic (e.g., liked, fights, cooperates, shy, full classroom and those completed by experts
leader). Although this procedure is relatively were generally quite high (rs ranging from .55 to
simple and straightforward, there are several .93). Similar correlations were found between the
variations of peer nomination procedures. For full classroom procedure and the nominations
example, there can be a predetermined number of completed by a random subsample (rs ranging
children that can be selected in each category from .49 to .90). Thus, these results by Prinstein
(fixed-choice format). Alternatively, the number suggest that there may be some less cost-intensive
of peers nominated in each category can be left alternatives to obtaining peer nominations, at
entirely to the child providing the nominations least in classrooms with adolescent students.
(unlimited-choice format). An example of The typical level of interpretation of a peer
instructions for both a fixed-choice and an nomination procedure is the number of times a
unlimited-choice format is provided in Box 9.2. child was nominated in a given category. This
Although some authors prefer the unlimited- number is then compared to a normative base for
choice format to avoid forcing a set number of that particular procedure to see if the child was
children into categories, especially negative cat- nominated at a level that is atypical for children his
egories, there is little empirical evidence for any or her age. However, there are some instances
clear advantage of one format over the other where more complex combinations or adjustments
(Gommans & Cillessen, 2015). of peer nominations are desired. For some pur-
A second dimension on which peer nomina- poses, it may be useful to compare the number of
tion procedures can vary is the degree of explicit- nominations obtained by a child in one area with
ness that defines the nominating pool. For the number of nominations that the same child
example, some procedures simply instruct the received in another area. One of the most common
children to consider any child in their class uses for sociometric techniques is to compare the
(Strauss et al., 1988). In contrast, other nomina- number of times a child was nominated as Liked
tion strategies provide children with a roster of Least with the number of times he or she was nom-
names from which to choose (Coie & Kupersmidt, inated as Liked Most by classmates. This approach
1983) or may even provide pictures of classmates allows one to determine the relative balance of the
from which the rater selects nominees for the two nomination categories or to determine the
individual categories, which is made easier by child’s “social preference” score. However, to
computer administration (Verlinden et al., 2014). make such comparisons, the two scores should
The rule of thumb is that the younger the rater, first be converted to standard scores (e.g., Z-scores;
the more explicit the definition of the nominating Coie et al., 1982) to equate for possible differences
pool should be. Also, if the pool of potential in the variance of the raw scores.
nominees is not within a well-defined group (e.g., A second type of conversion is warranted if
only part of a class is participating in the proce- one wishes to compare nominations of one child
dure), then more explicit definitions are required. with the nominations of another child from
Because of the difficulty and level of intru- another nominating pool (e.g., different class). To
siveness involved in collecting peer nominations make this comparison, the number of nomina-
from entire classrooms, Prinstein (2007) com- tions must be adjusted to equate scores for differ-
pared two alternative methods for obtaining peer ing class sizes. For example, 5 nominations of
perceptions. That is, this author compared nomi- Most Cooperative in a class of 12 should be inter-
nations obtained from a full sample of 232 ado- preted differently than 5 nominations in a class of
lescents ages 15–17 years old, with those obtained 30. As an example of this conversion, Strauss
(a) using only a randomly selected subsample of et al. (1988) divided the number of nominations
26 students and (b) using only 2 students in each obtained by a child by the number of children
class to rate other students. These students were participating in the assessment. The quotient was
chosen by their teachers as “social experts.” The
Peer Nominations 177
multiplied by 23 so that the nominations were all combining LM and LL nominations, the nomina-
expressed in terms of a common class size of 23. tions should first be converted to standard scores
to equate for potential differences in the variance
of the two categories. The five groups are based
Sociometrics on two difference scores. The social preference
score is the difference between LM and LL scores
Sociometric techniques focus on a specific, (LM − LL = Social Preference). High social pref-
important aspect of a child’s peer relationships: a erence scores indicate substantially more LM
child’s social status. It answers the question of nominations than LL nominations. The social
whether or not a child is liked and accepted by his impact score is the sum of the LM and LL scores
or her peer group. Sociometrics do not assess (LM + LL = Social Impact). High social impact
specific behaviors of the child. Thus, it does not scores simply determine the number of nomina-
answer the questions of what the child is like or tions a child receives, regardless of whether they
why the child is liked (Asher & Hymel, 1981). are negative or positive. A combination of these
Sociometric exercises have appeared in the scores leads to a child being considered in one of
research literature since the 1930s (see Gresham several social status groups: popular, rejected,
& Little, 1993; Hughes, 1990), and the most neglected, controversial, and average. The
commonly used procedure has changed very lit- method of combining these scores to determine a
tle over this time. An example of this basic tech- child’s social status and two computational for-
nique from Strauss et al. (1988) is provided in mulas that have been used in research are pro-
Box 9.2. vided in Table 9.1.
Sociometric exercises can take the form of Although there have been several variations in
peer ratings, whereby peers rate a child on a the formulas for determining a child’s social sta-
Likert scale as to how well liked or disliked he or tus, the validity of these groupings has been con-
she is (Hamilton, Fuchs, Fuchs, & Roberts, sistently shown in research (see Gresham &
2000). However, the more common method of Little, 1993), including showing good conver-
obtaining sociometrics is through peer nomina- gence with statistical methods for clustering chil-
tions. In this technique, children can be nomi- dren based on peer nominations (Zettergren,
nated by peers as a child who is Liked Most 2007). Further, research has suggested that these
(sometimes defined as most liked to have as a nominations are quite stable over time. Jiang and
best friend or most liked to play with), and/or Cillessen (2005) conducted a meta-analysis of 77
they can be nominated by peers as a child who is studies of over 18,339 participants and reported
Liked Least (or alternatively, Least liked to have that the average level of stability for social status
as a friend or Least liked to play with). Although over periods of less than 3 months ranged from
there is no definitive normative study that speci- r = .70 to r = .80. For periods of over 3 months,
fies exact cutoffs for when nominations are con- the average ratings ranged from .52 to .58.
sidered indicative of problems, in a fixed response Importantly, research has suggested that
format allowing 3 nominations in each category sociometric nominations can be influenced by the
in a class size of approximately 20 students, racial composition of a classroom with social
Liked Most (LM) nominations of less than 2 and preference scores being higher when nomina-
nominations of Liked Least (LL) of greater than tions are obtained from same-race peers
4 are generally considered indicative of problems (Singleton & Asher, 1979). Jackson, Barth,
in peer relations (Dodge, Coie, & Brakke, 1982; Powell, and Lochman (2006) reported on 1268
Green et al., 1981; Strauss et al., 1988). sociometric nominations from children ages
A common way of interpreting sociometric 9–11 years old across 57 classrooms that ranged
nominations is by combining the LM and LL from 3% to 95% African-American. The results
nominations into five distinct social status groups indicated that African-American students’ nomi-
(Hughes, 1990). As mentioned previously, when nations were more sensitive to the racial compo-
178 9 Peer-Referenced Assessment
sition of the classroom than for Caucasian the controversial status group have been less well
students. Specifically, African-American stu- studied. However, one study suggests that chil-
dents’ social preference scores and nominations dren in this social status group tend to exhibit
for fighting and being a leader improved as the aggressive and disruptive behaviors, like the
percentage of African-American students in the rejected children, but also tend to be viewed as
classroom increased. The effects for Caucasian socially skilled and as leaders, like the popular
students were less clear, although there was a children (Coie et al., 1982).
small effect of Caucasian students having lower A distinction that research has increasingly
preference scores if the classroom was less than shown to be important is between whether a child
33% Caucasian. is well-liked by their peers (i.e., accepted) and
Social status has also been associated with whether the child is viewed as “popular”
emotional and behavioral characteristics of the (Prinstein, 2007). First, these ratings have only
child. One of the most consistent findings is that been modestly correlated with each other
rejected children show higher levels of aggres- (Prinstein, 2007; Vaillancourt & Hymal, 2006).
sive and acting-out behavior than non-rejected Second, ratings of greater peer acceptance have
classmates (e.g., Dodge, 1983). However, consistently been related to more positive behav-
neglected status is also associated with problems ioral (e.g., less aggression; more prosocial behav-
in adjustment, most notably with anxiety (Strauss ior) and emotional (e.g., higher self-esteem)
et al., 1988). Several behavioral characteristics functioning (Gresham & Little, 1993; Sandstrom
are associated with popular children, including & Cillessen, 2006). However, children perceived
being more likely to contribute to conversations by their peers as popular often show more physi-
during play, being more likely to engage in paral- cal and relational forms of aggression (McDonald,
lel play, receiving and initiating more positive Putallaz, Grimes, Kupersmidt, & Coie, 2007;
social behavior, and using effective peer-entry Vaillancourt & Hymel, 2006) and are less likely
strategies (Gresham & Little, 1993). Children in to be victimized by others (van den Berg, Lansu,
Peer Nominations 179
decreased, ratings tended to be increas- they suggest that these data clearly indicate
ingly divergent from those obtained at full that assessors using sociometric ratings
participation. Even the ratings at the 75% need to recognize the potential effect of
level were significantly different from the participation rates on the results of these
ratings obtained at full participation. exercises, especially when assessing chil-
Importantly, the authors compared the rat- dren with adjustment problems.
ings of students with learning disabilities
Source: Hamilton, C., Fuchs, D., Fuchs,
and nondisabled students that were under-
L. S., & Roberts, H. (2000). Rates of class-
achieving to average and high-achieving
room participation and the validity of soci-
students. The decline in participation rates
ometry. School Psychology Review, 29,
was especially problematic for the ratings
251–266
of the two groups with learning problems.
The authors noted that the effects of
classroom participation on this peer-rating
procedure may not generalize to the more the Hamilton et al. (2000) study that have poten-
common peer nomination procedure, tially important implications for interpreting
although they make the important point information from sociometric exercises are
that participation effects on nomination summarized.
procedures need to be tested in light of
their results. They also offer an interesting
explanation for why participation rates Aggression
may affect the ratings of less well-adjusted
children to a greater degree. The authors Another common aspect of a child’s functioning
suggest that the better-adjusted children that is assessed by peer nominations is aggres-
may be viewed more similarly across class- sion. The typical format is to have a class nomi-
mates, with a general consensus across stu- nate the children in the class who “Fights most.”
dents on their likeability. In contrast, However, nominations can be obtained to assess
students may have more “polarized” views specific forms of aggression as well. For exam-
of children with problems in adjustment, ple, Blake, Kim, and Lease (2011) developed a
with some classmates liking them, others method for obtaining peer nominations of verbal
disliking them, and still others having a social, nonverbal social, direct verbal, and physi-
neutral view. As a result of this polariza- cal aggression. The questions used in this study
tion, the ratings of these children may be are provided in Table 9.2.
more dependent on which children are par- As with other nomination techniques, the for-
ticipating in the sociometric exercise. In mat can either be in a fixed-choice or unlimited-
actual practice, the problems in the accu- choice format. Peer nominations of aggression
racy in the peer ratings of disturbed chil- have been shown to be correlated with a psychi-
dren at lower levels of participation may be atric diagnosis of conduct disorder and therefore
exacerbated if these children tend to have can be considered an important indicator of the
disturbed friends who may be less likely to impairment associated with this syndrome
volunteer to participate in the exercise. The (Walker et al., 1991). However, one of the most
authors conclude by noting that it is impos- interesting and troubling characteristics of peer-
sible to identify a specific “minimum rate” nominated aggression is its stability. In their
of participation that should be obtained 5-year study, Coie and Dodge (1982) found that
before the results of a peer-rating sociomet- peer nominations of “Starts Fights” showed the
ric exercise are uninterpretable. However, most stability across the 5 years of any of the
peer nomination categories that were obtained,
Peer Nominations 181
Table 9.2 Peer nominations for multiple types of (Kowalski & Limber, 2013). However, it is often
aggression
not observed by anyone but other children within
Type of a social network, such as peers at school (Jones,
aggression Nomination item
Mitchell, & Finkelhor, 2012). Thus, assessment
Overt This person says mean things to people,
calls people names, and teases others in of cyberbullying through peer nominations or
a mean way peer ratings is a potentially important method for
This is somebody who tries to get what determining who might be either perpetrators or
he or she wants by hitting, shoving, victims of such acts.
pushing, or threatening others
In a test of peer nominations of electronic
Verbal This person tries to keep certain people
social from being in their group during aggression, Badaly et al. (2015) reported on a
activities sample of 808 adolescents (ages 14–17 years)
This person tells tell others that he or using nominations on two items:
she will stop liking them unless they do
what he or she t says
• Students that use the Internet or text messages
Nonverbal This person rolls their eyes at someone
social when they are angry with them
to insult or be mean
This person acts bored and sighs often • Students whom other students insult or are
when someone they don’t like is talking mean to using the Internet or text messages
When mad at a classmate, this person
will ignore him/her by looking away or The unlimited nomination procedure was
pretending not to hear what he/she said
used, with participants nominating as many or as
This person will act disgusted,
sometimes with his or her hands on few peers as they wanted. The authors reported
their hips, when he or she doesn’t like that nominations of both electronic aggression
the kid and cyberbullying showed high interrater agree-
Source: Blake et al. (2011) ment and stability over a 1-year period. Further,
peer nominations of electronic aggression were
associated with aggression in the classroom.
exhibiting correlations of .83 between third and Finally, the agreement between peer nominations
fourth grades and .84 across fifth and sixth and self-reports was modest for both aggression
grades. Huesmann et al. (1984) provide even (r = .21, p < .001) and victimization (r = .23,
more dramatic evidence for the stability of peer p < .001). Thus, the findings of this study suggest
nominations of aggression. These authors found that peer nominations of aggression were mea-
that peer nominations of aggression at age 8 suring an important construct that was somewhat
significantly predicted aggression 30 years later. independent what was being measured by using
Therefore, peer nominations of aggression assess self-reports.
an aspect of problematic interpersonal function-
ing that can be highly stable for a child and is
thus an important target for intervention. Hyperactivity
One particularly promising domain for using
peer nominations is to assess electronic forms of Assessment of inattention and motor hyperactiv-
aggression, sometimes called “cyberbullying” ity, behaviors typically associated with ADHD,
(Ciucci, Baroncelli, Franchi, Golmaryami, & has relied primarily on information obtained
Frick, 2014). Electronic aggression involves a from parents and teachers (Loeber et al., 1991).
child attempting to hurt another person using However, Schaughency and Rothlind (1991) pro-
electronic technology, such as sending threats in vide some interesting data to suggest that peer
text messages or e-mails and posting unflattering nominations of inattentive and hyperactive
pictures on social networking sites (Badaly, behavior could aid in the assessment of
Duong, Ross, & Schwartz, 2015). Such bullying ADHD. Specifically, these authors reported that
appears to result in significant harm to its victims peer nominations of “Can’t pay attention,” “Can’t
182 9 Peer-Referenced Assessment
tinuum, a neutral face to anchor the middle, and a group. These dimensions of social status cannot
sad face to anchor the negative end of the con- be adequately assessed by other methods of
tinuum (Asher et al., 1979). assessment, such as teacher ratings or a child’s
One of the most reliable forms of peer assess- self-report. In addition, this crucial aspect of a
ment, especially for very young children child’s social context has been highly related to
(McConnell & Odom, 1986), is the paired emotional and behavioral disturbances.
comparison technique. In this procedure, photo- Therefore, sociometric assessment is a type of
graphs are taken of all the children in the class, peer-referenced assessment that could be an
and photographs of each possible classmate dyad especially important component of many clinical
are paired. The rater is then asked to choose evaluations.
between the two children in each dyad in refer-
ence to some criterion (e.g., fights, liked, shy, Chapter Summary
cooperative, etc.). The salient cues used in mak-
ing the choices between peers make this proce- 1. Peer-referenced assessments provide informa-
dure much more reliable for young children, tion on how a child or adolescent is viewed by
especially of preschool age. However, it is so his or her peers and, thereby, provide impor-
labor-intensive that it is not feasible for use in tant insights into a child’s social milieu.
most clinical assessments. As McConnell and 2. Peer-referenced assessments must be con-
Odom (1986) point out, for a class of 20 children, ducted in light of two important ethical issues:
each child will have to make 171 selections for the importance of minimizing the intrusive-
each criterion. ness of the assessment and the need to mini-
mize potential reactive effects of negative
ratings by peers.
Conclusions 3. Peer nominations are the most commonly
used forms of peer-referenced assessments.
Peer-referenced assessment strategies share the They involve having a child’s or adolescent’s
common characteristic of having a child or ado- peers select one or more children who display
lescent evaluated on important psychological certain characteristics.
dimensions by his or her peers. Due to several 4. Sociometric exercises are a type of peer nomi-
practical and ethical considerations, peer- nation that determines whether a child is
referenced strategies are not commonplace in accepted, rejected, or neglected by his or her
many clinical assessments. This is unfortunate peers.
because, if designed appropriately and with pre- 5. Aggression (both in person and through the
cautions taken to ensure safe administration, use of electronic devices), hyperactivity, and
peer-referenced assessment can provide an depression have also been assessed through
invaluable picture of a child’s social context. This peer nomination procedures. Unfortunately,
picture of a child’s social relationships is essen- the relative utility of peer-referenced assess-
tial for treatment planning, given the importance ments of these psychological dimensions, in
of social relationships for a child’s current and comparison to other assessment modalities,
future psychological adjustment. has not been tested.
One of the most commonly used peer- 6. In addition to peer nominations, perceptions
referenced techniques is the sociometric exercise. of a child’s peers can be obtained by having
This peer nomination technique allows one to children rate each other along certain dimen-
determine whether or not a child is accepted, sions on a Likert-type scale.
rejected, or ignored (neglected) by his or her peer
Projective Techniques
10
Roberts, 1982). Both of these approaches to pro- the judgments of other experienced clinicians as
jective testing are discussed in more detail later a guide to making interpretations.
in this chapter. However, it is important to note The importance of understanding this debate
that both systems share the goal of providing is not to decide which view is right. What is more
very clear and explicit guidelines on how the crucial is for one to recognize the two disparate
tests are to be given, scored, and interpreted. ways of using projective techniques and the
Such standardization is a prerequisite to further unique strengths and weaknesses of both. For
psychometric evaluation. example, using projectives as a psychometric
This method of addressing the criticisms of technique allows one to compare a person’s
projective tests has not met with unanimous score with those from a normative group, or with
approval. Instead, it has been argued that projec- those from some relevant clinic group, or with
tive tests should not be evaluated by traditional some other clinically important criterion (e.g.,
measurement theory and any attempt at stan- response to treatment). However, to use the
dardization will limit the clinical utility of the scores in this way, one must maintain rigorous
technique. For example, Haak (1990) has argued standardization in procedures and be willing to
that: live within the confines of the data that are avail-
The problem with all of these standardization able. A frustrating aspect of clinical assessments
efforts is the amount of destruction they wreak on is realizing the limitations of what our assess-
the essential nature of projectives. All such ments can provide.
approaches result in a huge loss of the rich and On the other hand, using projectives as a clini-
complex information that is obtained by using the
technique in the first place. (p. 149) cal tool allows one greater flexibility in adminis-
tration and interpretation. However, with this
This argument is based on the contention that flexibility, the interpretations that result from the
projectives are part of the older clinical tradition assessment are much more susceptible to influ-
that seeks to describe the individual person in an ences that are idiosyncratic to the assessor.
idiographic approach that is not concerned with Interpretations of the same case material may
how the individual differs from the norm or how vary widely across clinicians. As such, interpre-
his or her scores compare to those of some other tations should be clearly viewed as clinical
reference group (e.g., those with diagnoses of impressions and not be evaluated in the same way
depression). Instead, the goal is simply to under- as empirically derived interpretations. In Box
stand the person’s unique qualities. In this con- 10.2, we have provided a more detailed discus-
ceptualization, “validity” takes on a very different sion of the importance of clearly defining one’s
meaning than the one that is typically used in approach to projective assessment and then rec-
measurement theory. One is not concerned with ognizing the limitations inherent in either
how a score compares to some objective criterion method.
outside the person being tested.
For some psychologists, this clinical approach
might seem unscientific. However, all clinicians Box 10.2 Two Approaches to Projective
rely on intuition at some point in an evaluation to Testing: You Can’t Have It Both Ways
understand the nuances of an individual case.
Our science of human behavior is not at a point The divergent approaches to the interpreta-
where every clinical decision can be guided by tion of projectives can be descriptively
well-established principles, and, given the com- labeled as the psychometric approach and
plexity of psychological functioning, such pure the clinical approach. The problem that
empiricism may never be possible. Therefore, arises in the use of projectives is that many
the clinical view of projectives considers these clinical assessors aren’t aware of the
techniques as a structured way of obtaining these (continued)
intuitions. By using this structure, one can use
188 10 Projective Techniques
Projection of Behavioral Sample? It is evident from this quotation that one can
view responses to projective tests as samples of
Even more basic than the debate over the behaviors from which one would like to general-
method of interpretation is confusion over what ize to behaviors in other situations, outside of the
psychological processes projective techniques testing environment. In fact, this type of inter-
are supposed to measure. The critical nature of pretation underlies Rorschach’s original devel-
this question is obvious from a psychometric opment of the inkblot test and has been the
viewpoint. Validity is the critical property of a guiding principle for Exner’s more recent system
test, and it is often defined as evidence that the of interpretation. Rorschach, and later Exner
test is measuring what it is supposed to measure (see Exner & Martin, 1983; Exner & Weiner,
(Anastasi, 1988). Therefore, if it is unclear what 1994), describes the inkblot tests as a “percep-
a test is supposed to measure, then it will be tual test,” meaning that a person’s perception of
unclear as to what are the most appropriate the inkblot is used as a sample of behavior with
methods of determining its validity. which to generalize to the person’s perception of
One dominant view of projective tests, which other, more clinically relevant, stimuli.
is the view that led to the name projective, is best These two competing views of what is mea-
described by Murray (1943): “There is the ten- sured by projectives have several important
dency for people to interpret an ambiguous implications for the assessment process. As
human situation in conformity with their past already mentioned, how one views the process
experiences and present wants” (p. 1). This will determine what evidence is used to establish
forms the basis of the projective hypothesis. The the test’s validity. For example, if one views the
projective hypothesis rests on the assumption test as a behavioral sample, then one would want
that people, in the absence of clear environmen- evidence that the behaviors obtained from the
tal demands, will project basic aspects of them- test are associated with behaviors outside the
selves in their interpretations of environmental testing situation. Alternatively, if one views pro-
stimuli. Freudian theory, which dominated clini- jectives as tapping unconscious conflicts, then
cal psychology for decades, heavily emphasized the relationship to overt behavior is not expected
unconscious conflict as the basic element of to be one-to-one, because the same conflicts can
human personality. Projection is seen by many as be manifested in different behaviors (Koppitz,
being a window to these unconscious dynamics 1983). In this case, validity would be best estab-
(Rabin, 1986). lished by showing that responses on a projective
However, there is a second view of projec- technique are associated with other indicators of
tives. Rather than seeing them as windows to hid- unconscious conflicts.
den or unconscious motives and drives, many Implicit in this discussion is the important
assessors view projectives as a behavioral sam- point that how one views the psychological pro-
ple. For example, Knoff (1983) writes: cess that is being measured by projective tests
A student completes an incomplete sentence will determine the types of interpretations that
blank with “I hate myself” or “My father beats will be made from a child’s or adolescent’s
me up all the time,” and these hypotheses are responses. A person viewing the results in terms
confirmed through self-injurious behavior or a of projection will make interpretations about
physically abusive father, is this a hidden aspect
of personality? Or how is a student’s response to drives and motivations. It is these types of predic-
a thematic approach which uses real photographs tions that one wishes to make. In contrast, a per-
depicting significant interpersonal situations son viewing the results in terms of a sample of
(i.e., peer group acceptance, attitudes toward behavior will make interpretations about behav-
school-work, reactions to new sibling) different
from an interview question asking how she/he is ioral tendencies that are likely to be manifested in
getting along with peers on the playground? situations outside of the testing situation. For
(p. 448) example, if the Rorschach is used as a sample of
190 10 Projective Techniques
perceptions, one would wish to make predictions selection of a technique that is not well suited for
about how these perceptual tendencies will be their purpose. The rest of this chapter will high-
manifested in other situations. light characteristics of specific projective tech-
The final impact of viewing projective tech- niques. However, these general issues are
niques as either projection or a behavioral sample paramount in understanding and using these
is its influence on the selection of the type of techniques; therefore, these issues are revisited
stimulus used. Specifically, if one is operating throughout this chapter with reference to specific
from the projective hypothesis, one would want techniques.
as ambiguous a situation as possible. For exam-
ple, the Thematic Apperception Test (Murray,
1943) contains a blank card that has no picture on Inkblot Techniques
it, and the person is required to make up a story
about this card. This is an example of a very One of the most commonly used projective tech-
ambiguous situation with few demand character- niques is the inkblot technique. The stimulus is
istics, or very little stimulus pull, that would simply an inkblot, and the child is asked to inter-
guide a person’s response. This stimulus allows pret this ambiguous stimulus in some way. The
for the purest form of projection. best known of the inkblot techniques, the
In contrast, if one wishes to obtain a behav- Rorschach, consists of 10 cards with standard-
ioral sample from the projective technique, high ized inkblots. Although the Rorschach is often
levels of stimulus pull may actually be benefi- considered synonymous with inkblot techniques,
cial. If one knows the demand characteristics a notable alternative is the Holtzman inkblot
that promoted the response, then one would have technique (HIT). This technique was developed
some clue as to what situations one might gener- to overcome psychometric limitations in the
alize (i.e., ones with similar demand characteris- Rorschach by constructing a completely new set
tics). For example, cards from the Roberts of inkblots (Holtzman & Swartz, 2003). The HIT
Apperception Test for Children (McArthur & consists of two parallel forms, each of which con-
Roberts, 1982) were designed to pull for specific tains 45 inkblots.
themes (e.g., peer conflict, school problems, Volumes have been written on the different
marital discord in parents). This property is not interpretive systems for inkblot techniques for
desirable from the pure projection viewpoint children (e.g., Ames et al., 1974; Exner &
because it increases the demand characteristics Weiner, 1994; Holtzman & Swartz, 2003). The
of the stimulus and thereby limits the degree of primary variations among these systems are
projection required. along the dimensions discussed in the introduc-
tion to this chapter: whether they view the ink-
blot test as a projective approach or as a
Summary behavioral sample, and whether they take a clini-
cal or psychometric approach to interpretation.
Our approach to the debate over the use of pro- Rather than giving a superficial overview of the
jective techniques is that assessors should use or different approaches to interpretation, we will
not use projective testing based on a careful con- focus on one of the most commonly used meth-
sideration of critical assessment issues. Assessors ods of interpretation for children: Exner’s
are often unclear about what approach to mea- Comprehensive System (ECS) for Rorschach
surement they are using (i.e., clinical or psycho- interpretation. This system attempts to integrate
metric) and often make inappropriate five major approaches to Rorschach interpreta-
interpretations based on this confusion. Further, tion into a single Rorschach approach (Exner &
assessors are often unclear about what they are Martin, 1983). Even limiting our focus to one
trying to measure with projectives, and again, system of interpretation does not allow us to do
this leads to confusion in interpretations or to the justice to the intricacies of Rorschach interpreta-
Inkblot Techniques 191
tion using the ECS; accordingly, the reader is might this be?”. The only unacceptable response
referred to Exner and Weiner (1994) for a more is, “It’s an inkblot.” If the child provides this
in-depth discussion of this system of interpreta- answer, then he or she is encouraged to see it as
tion for assessing children and adolescents. something it’s not. All 10 cards are administered
in this way, and the subject’s responses during
this free association phase are recorded verbatim
he Exner Comprehensive System
T for later scoring.
for Rorschach Interpretation If the child responds to the first card with only
one answer, the examiner hands the card back to
Process Measured The ECS treats the the child and says “Some people see more than
Rorschach as a perceptual-cognitive task. When one thing.” Such instructions are given only for
viewed in this way: this card, and, as a result, the number of responses
that are given across the 10 cards can vary widely.
The Rorschach becomes a task to which people Given the potential influence that the number of
respond by exercising their perceptual-cognitive responses can have on the interpretations made
abilities and preferences. To articulate their from the test, alternative administration proce-
answers, they must select parts of these variegated dures have been proposed that (a) prompts for at
stimulus fields to which they wish to attend, use
some mixture of the features of the stimulus and least two responses to each card and (b) prevents
their own needs to guide formulations, and iden- more than four responses being given to any card
tify objects that will give substance to their (Reese, Viglione, & Giromini, 2014). It is not
impressions. In short, they decide how to scan the clear, however, how this method of administra-
stimulus, how to translate the stimulus input, and
what to report. (Exner & Weiner, 1982, p. 3) tion called the Rorschach Performance
Assessment System (R-PAS) should influence
Weiner (1986) outlines four basic factors that the interpretation of scores.
influence a child’s response to the inkblot. First, After this free association phase, the exam-
the nature of the stimulus itself may lead the iner enters the inquiry phase. The assessor read-
child to classify a blot in a certain way. Although ministers each card and reads to the child his or
the inkblots were designed to minimize stimulus her initial responses. The child is instructed to
pull, there are clearly some common or typical show the examiner which part of the blot led to
answers that are based on the specific features of the response and what made him or her think it
the blots. Second, responses may be influenced looked that way. The child is informed that the
by concerns about making a particular impres- assessor would like to see it “just the way you
sion which could lead to some censoring of did,” and several standardized prompts are pro-
responses in a socially desirable manner. Third, vided (e.g., “What is in the blot that makes it
responses are influenced by personality traits look like that to you?” Exner & Wiener, 1994).
that predispose a person to perceive the blots in The child’s responses during this inquiry phase
idiosyncratic ways. Fourth, responses are partly are also coded verbatim to be used in later
a function of situational psychological states scoring.
that affect a person’s perceptual experience. The heart of the ECS is the extensive and
Each of these factors provides an important con- detailed scoring procedure of a child’s test proto-
text for interpreting a child’s response to the col (i.e., verbatim responses to Rorschach cards).
Rorschach inkblots. This system includes approximately 90 possible
scores. There are seven major categories of codes,
Administration and Scoring In contrast to the which are described in Table 10.1: location,
complexity of Rorschach interpretation, adminis- determinants, form quality, organizational activ-
tration of the task is relatively simple. The child ity, popularity, content, and special scores. The
being assessed is handed with individual cards ECS utilizes a structural summary, which shows
with the inkblot stimuli and merely asked, “What all of the possible scores, plus various summary
192 10 Projective Techniques
Table 10.1 Summary for scores used in Exner’s Comprehensive System to Rorschach interpretation
Categories Description Example of scores
Location Part of the blot used by respondent W = whole blot
D = common area
Dd = uncommon area
S = white space
Determinant Features of the blot that contributed to F = form
the formation of the response C = color
T = texture/shading
M = human movement
Form quality Measures the perceptual accuracy of + = superior-overelaborated
the response (i.e., does the area of the 0 = ordinary-common
blot really conform to the child’s U = unusual-rare but easy to see
perception) − = minus-distorted, arbitrary, and unrealistic
Content Places into categories the various H = whole human
persons, places, and things that form An = anatomy
the child’s response Bl = blood
Fi = fire
Fd = food
Hh = household items
Popular Codes the number of times the child P = number of popular responses given in the entire
gave a high-frequency (very common) protocol
response to a blot
Organizational Provides an estimate of the efficiency Z score = higher scores indicate greater organizational
activity of a child’s organization of the effort
stimulus field
Special scores Denotes unusual verbal material in a INCOM = incongruous combination merges details or
child’s response images in unrealistically way
MOR = morbid-response includes references to death or
clear dysphoric feeling
AG = aggressive movement response includes action
that is clearly aggressive
scores that are ratios, percentages, and other The extensive normative database for children
derivations of the individual scores that provide available with the ECS aids in such interpreta-
important information for interpretation. tions. This normative base is one of the major
reasons for the popularity of the ECS for use with
Norming The best normative data on the Exner children. However, it is important to note that the
system were obtained in a large (n = 1870) adequacy and consistency of the normative scores
nationwide sample of children between the ages across different samples of children and adoles-
of 5 and 16 (Exner & Weiner, 1994). At each age cents have been questioned (Hunsley & DiGiulio,
in this 12-year range, there were at least 105 chil- 2001; Meyer, Erdberg, & Shaffer, 2007).
dren. There was also fairly equal gender repre-
sentation at each age, and the inclusion of Reliability Because of its explicit and standard-
minority children was at a proportion that ized administration and scoring procedure, it is
approximated national demographics. From this not surprising that the ECS has proven to be more
normative sample, Exner and Weiner (1994) doc- reliable than many other inkblot interpretive sys-
umented several age-related trends in scores. tems, showing high interrater and high test-retest
These trends are summarized in Table 10.2. Users reliability (Hiller et al., 1999). For example, in a
of the ECS with children should be aware of sample of 25 8-year-old children, 1-week test-
these developmental changes in the Rorschach retest coefficients for individual and summary
responses and interpret scores within a normative scores ranged from r = 0.49 to r = 0.95, with a
perspective. mean coefficient of r = 0.84 (Exner & Weiner,
Inkblot Techniques 193
Table 10.2 Age trends in the normative data for the Table 10.2 (continued)
Exner Comprehensive System for Rorschach
Score Age trend
interpretation
Popular There is a steady increase in the
Score Age trend responses number of popular responses with age,
Length of Younger children tend to give fewer with adolescents giving approximately
record responses than older children. one third more popular responses than
Protocols of less than 17 are not children under age 8
uncommon before age 15, and records Special Many of the special scores that
of more than 25 are unusual prior to scores document unusual verbalizations and
age 13 cognitive slippage are more common
Location Younger children (less than 11) give in young children than in adolescents
more responses that include the whole and adults
blot rather than a specific area. More
children, and especially very young
children, give at least one response that 1994). In fact, the only coefficient to drop below
uses an infrequently identified area of r = 0.70 was the coefficient for inanimate move-
the blot ment (r = 0.49).
Develop- Younger children frequently give many
mental vague responses in which diffuse
quality impressions of the blot or blot area are Validity and Interpretations While Rorschach
given without clearly articulating scores are reliable over short-time intervals, the
specific outlines or structural features. stability of the scores over longer periods is lower
Such vague responses account for
than the stability of adult scores. Specifically,
one-third of the responses of children
ages 5, 6, and 7 Exner, Thomas, and Mason (1985) tested 57 chil-
Movement Younger children give few human dren at 2-year intervals from age 8 to age 16. In
determinants movement responses. It is not unusual general, most scores showed only moderate con-
for the responses of 5-, 6-, and sistency over each 2-year interval until the inter-
7-year-olds to have few or none such
determinants, whereas it is uncommon
val between the ages of 14 and 16. Some notable
for this to occur after age 11 exceptions were the fairly stable coefficients for
Chromatic It is not uncommon for children to give the use of good form, the use of popular responses,
color color responses that are not created the number of active movement responses, and
determinants based on the form features of the blot.
the use of shading features for making depth and
The presence of such pure color
responses is often interpreted as dimension responses.
indicative of poor affective regulation.
About 70% of 5 year-olds, 35% of 8 The modest stability of Rorschach scores is
year-olds, 23% of 12 year-olds, and not unique to this type of assessment, but is char-
8% of 16 year-olds give at least one acteristic of most assessment techniques in chil-
pure color response dren (e.g., McConaughy, Stanger, & Achenbach,
Form Answers that include the impression of
1992). In fact, this is probably a positive attribute
dimension depth, distance, or dimensionality
increase with age such that, by age 8, of the scores because it suggests that the scores
they appear at least once in over half of capture the rapid developmental changes experi-
all subjects’ records enced by children and adolescents. However,
Reflection In contrast to responses of adults, there is a tendency to equate Rorschach responses
responses images reported as a reflection or
mirror image because of the symmetry with personality assessment and to equate per-
of the blot are quite common in child sonality with stable dispositions. These findings
protocols. Such responses appear in on the low stability of Rorschach scores clearly
about half of the protocols of children argue against making strong dispositional state-
under the age of 8. Although the
incidence of such responses declines ments based on a child’s Rorschach protocol.
over time, they still are found in about One common use of the ECS has been to
25% of the protocols of 15-year-olds assess depression in children. Exner (1983) ini-
(continued) tially developed a depression index (DEPI) based
194 10 Projective Techniques
on six scores from a child’s protocol. responses that condense blot details or images
Unfortunately, the DEPI, based primarily on into a single incongruous percept in which the
research in adults, showed very poor agreement parts or attributes do not belong together: “a per-
with other measures of depression in children, son with the head of a chicken” (Weiner, 1986,
which led Exner to revise the DEPI (Exner, 1990) p. 217). Second, perceptual inaccuracies are
in an effort to increase its correspondence with suggested when the child has a protocol with
other measures of depression. Tests of the revised many responses that do not correspond closely to
DEPI index have also failed to find consistent the form structure of the blot (i.e., poor form
associations with other measures of depression quality) or protocols with few common or popu-
(Archer & Krishamurthy, 1997; Ball et al., 1991; lar responses. Third, interpersonal inadequacies
Carter & Dacy, 1996). These findings could be a that are often associated with schizophrenia can
function of inadequate methods of assessing be assessed in responses involving human move-
childhood depression in general, which results in ment. Movement responses with poor form qual-
the failure to have an appropriate standard with ity are considered indicative of inaccurate or
which to judge the Rorschach. Alternatively, unrealistic interpretations of interpersonal situa-
Weiner (1986) has argued that the Rorschach: tions (Exner & Weiner, 1994). Fourth, the irregu-
is a measure of personality processes, not diagnos- lar content of a protocol, such as one with very
tic categories … it can only help to identify forms violent (e.g., two boys stabbing each other in the
of psychopathology only to the extent that they chest) or very bizarre (e.g., flowers squirting poi-
identify personality characteristics associated with sonous gas) content can be considered suggestive
the types of disorder. (p. 155)
of disturbed ideation that is often associated with
However, these findings suggest that users of schizophrenia.
the DEPI from the ECS should not expect the Exner and Weiner (1982) reported data on 20
scores to be highly related to other indexes of children (ages 9–16) reliably diagnosed with
depression. schizophrenia and 23 nonschizophrenic chil-
An even more extreme caution is in order for dren. These data indicated that the use of these
the suicide constellation for children, a set of indexes produced a high correct classification
scores based on an index used to assess for sui- rate (90.7%). These positive findings must be
cidal tendencies in adults (Exner, 1978). The sui- interpreted with three cautions, however. First,
cide constellation was developed by selecting the indicators of perceptual disturbances for the
eight best predictors of suicide from the ECS in a Rorschach have not always shown high correla-
small sample (n = 39) of children who had tions with other behavioral indicators of thought
attempted or committed suicide within fewer disorders (Smith, Baity, Knowles, & Hilsenroth,
than 60 days after the Rorschach was taken 2001). Second, while these indexes appear to be
(Exner & Weiner, 1994). Unfortunately, the pre- correlated with a diagnosis of schizophrenia, it
dictive validity of these indexes has not been rep- is unclear how much utility these indexes pos-
licated in other samples, making the interpretation sess over the actual behavioral symptoms of the
of this index questionable at present (Allen & disorder (Gittelman-Klein, 1986). In other
Hollifield, 2003). words, there is no evidence to suggest that chil-
Another common use of the Rorschach is in dren who show elevations on these indexes, but
the detection of cognitive and perceptual irregu- who do not show overt behavioral manifesta-
larities that could be associated with schizophre- tions of the disorder, are at risk for developing
nia (Weiner, 1986). Exner and Weiner (1994) schizophrenia. Second, Gallucci (1989) studied
outline four sets of Rorschach scores that can aid a sample of 72 intellectually gifted children and
in this detection. First, disordered and illogical found elevated rates on these indexes, compared
thought processes are the focus of several special to age norms, but no other signs of maladjust-
scores in the Exner system. For example, the ment in the children. Intellectually superior chil-
incongruous combination score identifies dren may process the Rorschach stimuli in
Thematic (Storytelling) Techniques 195
nonconventional ways, but these differences Thematic tests have been a popular type of
should not be considered indicative of a psy- projective technique with children because the
chotic process. Similarly, Holaday (2000) storytelling format is usually nonthreatening and
reported that children and adolescents diagnosed fun for them. However, it does require a signifi-
with post-traumatic stress disorder also reported cant level of verbal ability in the child.
significantly higher scores on the Rorschach There are many different thematic techniques
indicators of schizophrenia. Thus, indicators of that have been used with children and adoles-
schizophrenia on the Rorschach appear to have a cents, which vary in the types of pictures that are
high rate of false positives for thought disorders used to promote children’s stories (see Kroon,
(i.e., many children who score high do not have Goudena, & Rispens, 1998 for a review). One of
other indicators of psychosis). the most commonly used techniques for children
and adolescents is the thematic apperception
Evaluation These examples are only a small technique (TAT; Murray, 1943). The TAT con-
sample of the common uses of the ECS in clinical sists of 31 cards with black-and-white pictures
assessments of children and adolescents. of primarily adult figures involved in some rela-
However, these examples illustrate two issues tively ambiguous action or interaction. Because
that are important for using the Rorschach in the the TAT was designed primarily for use with
assessment of children and adolescents. In gen- adults and contained adult pictures, the
eral, Rorschach responses do not typically corre- Children’s Apperception Test (CAT; Bellak &
spond closely to behaviorally based diagnoses; Bellak, 1949) was developed especially for chil-
therefore, use of the Rorschach for diagnostic dren. The original CAT contains 10 cards with
purposes is not recommended. Second, many of pictures depicting animal figures, although a
the Rorschach scores and indexes were devel- later CAT-H was developed with human figures.
oped and validated on adults. Unfortunately, the The pictures on both the CAT and the CAT-H
extension to children has not met with great suc- were designed to elicit typical childhood con-
cess, as evident by studies on the children’s flicts that are predicted from psychodynamic
depression index and suicide constellation of theory (e.g., sibling rivalry, oedipal urges, toilet-
children. Therefore, users should be wary about ing concerns). Another thematic test heavily
using adult-oriented systems for interpreting the influenced by psychodynamic theory is the
Rorschach responses of children. Blacky picture test (Blum, 1950). The Blacky
picture test consists of 11 cartoons whose cen-
tral figure is a dog named Blacky. Like the CAT,
Thematic (Storytelling) Techniques the pictures were designed to depict psychosex-
ual conflicts that were believed to be common in
The second type of projective techniques for chil- children.
dren is the storytelling or thematic approaches. In Two apperception tests, the School
these techniques a child is shown a moderately Apperception Method (Solomon & Starr, 1968)
ambiguous picture or photograph and asked to and the Michigan Picture Test-Revised (Hutt,
tell a story about it. For example, the instructions 1980), were designed more specifically for use
to the Roberts Apperception Test for Children in educational settings. Another apperception
(McArthur & Roberts, 1982) are: test designed specifically for use with children is
I have a number of pictures I am going to show you the Roberts Apperception Test for Children
one at a time. I want you to make up a story about (RATC; McArthur & Roberts, 1982). The RATC
each picture. Please tell me what is happening in is quite explicit in the themes assessed by the
the picture, what led up to this scene, and how the stimulus pictures. Unlike many other thematic
story ends. Tell me what the people are talking
about and feeling. Use your imagination and techniques, the themes the pictures were
remember that there are no right or wrong answers designed to assess are not specific to psychody-
for the picture. (p. 7) namic theory. Also, the RATC is one of the few
196 10 Projective Techniques
thematic techniques that includes an explicit scores, but most assessors do not use any sys-
scoring system. The RATC is reviewed in greater tematic scoring system. Given the lack of con-
detail later in this chapter. sistency in administration and scoring, it is not
Several thematic approaches have specific surprising that evidence for the reliability and
sets of pictorial stimuli for specific groups of validity of thematic techniques is limited.
children. This is based on research showing that Thematic techniques are often interpreted
children provide greater verbalization on the- within an idiographic or clinical tradition in
matic apperception techniques when the stimu- which clinical impressions of an individual
lus material more closely matches their ethnicity, child are obtained through an analysis of the
gender, and age (Constantino & Malgady, 1983). child’s stories.
For example, the TAT and RATC contain some Clinical interpretation of a child’s story is
pictures that are gender-specific, and the RATC typically based on two broad aspects of a child’s
contains supplementary pictures depicting response. The first step is a process interpreta-
African-American children (McArthur & tion. In this part of the interpretation, one notes
Roberts, 1982). Of particular note, the tell-me- such characteristics of the stories as how elabo-
a-
story technique (TEMAS: Constantino, rate the stories were, whether the stories were
Malgady, & Rogler, 1988) is a Thematic coherent and tied to the stimulus card, and
Apperception Test that was specifically designed whether there were any specific cards for which
to be a culturally sensitive test for inner-city the child had difficulty formulating a story. This
children and adolescents. The TEMAS involves type of interpretation can be used to determine
23 brightly colored cards depicting inner-city how invested the child or adolescent was in the
themes involving peer and family interactions. assessment process, whether there were any
There are 11 sex-specific cards and 2 parallel potential disturbances or idiosyncrasies in
sets for minority (depicting Hispanic and thought processes, and whether there were any
African-American characters) and nonminority specific types of stimuli that elicited defensive
children. Also, unique to the TEMAS are sepa- reactions from the child.
rate norms for Caucasian, African-American, The second part of the interpretive process is
Puerto Rican, and other Hispanic children a content analysis. Children’s stories are typi-
across three age groups (5–7, 8–10, 11–13). cally analyzed for (1) the characteristics of the
However, these norms are based on a rather lim- hero or main character (e.g., motives, needs,
ited sample (n = 642) of children from public emotions, self-image), (2) forces that affect the
schools in New York City (see Flanagan & hero in his or her environment (e.g., rejection by
DiGiuseppe, 1999). peers, punitiveness from parents, frightening
forces, support by parent, affection from sib-
ling), (3) the coping or problem-solving strate-
eneral Interpretation of Thematic
G gies used by the hero (e.g., aggression,
Techniques compromise, nurturance), and (4) the outcomes
of the story (e.g., positive or negative, outcomes
One important issue in the use of thematic tech- brought about by the hero or someone in his
niques is the lack of standardized administra- environment; outcomes are realistic). The con-
tion or scoring procedures for most systems. Of tent analysis should determine whether there are
the 12 Thematic Apperception Tests used for any consistent themes in a child’s story, espe-
children and adolescents reviewed by Kroon cially themes that transcend the stimulus pull of
et al. (1998), only 5 had standardized and objec- a card. For example, an aggressive story pro-
tive methods of scoring children’s responses. vided for a card that shows two children fighting
For example, assessors administering the TAT is less informative than a story with an aggres-
often select certain cards to administer. Further, sive theme based on a picture of two people sit-
there are many different systems for obtaining ting next to each other in a park.
Thematic (Storytelling) Techniques 197
Roberts Apperception Technique Table 10.3 Depictions in the stimulus cards from the
Roberts Apperception Test for children
for Children
Card
number Description Common themes
The Roberts Apperception Techniques for
1 (B Both parents Elicits themes of
Children (RATC; McArthur & Roberts, 1982) is & G) discussing family confrontation
one example of a thematic technique that was something with and stories in which
explicitly designed for use with children and is child parents are giving
one of the few storytelling procedures with an advice or punishing a
child
explicit and standardized scoring system. This
2 (B Mother hugging Elicits themes of
instrument illustrates some major components in & G) child maternal support and
the interpretive process of thematic techniques. dependency needs in
relation to a material
figure
Content The RATC is intended for use with
3 (B Child working on Elicits themes related
children and adolescents ages 6–15. There is a & G) homework to child’s attitude to
standard set of 27 stimulus cards depicting com- school, teachers, tests,
mon situations, conflicts, and stresses in chil- and homework
dren’s lives (McArthur & Roberts, 1982). Eleven 4 One child standing Elicits themes with
over another child aggression, accidents,
cards have parallel male and female versions,
in prone position and illnesses
and there is a supplementary set of stimulus 5 (B Parents are shown Elicits themes related
cards featuring African-American children. A & G) in an embrace with to a child’s attitude
description of the RATC cards and the themes child looking on toward parental
the cards were designed to elicit are provided in displays of affection
6 (B Two white children Elicits themes related
Table 10.3.
& G) are shown to peer interactions and
interacting with a racial attitudes
Administration and Scoring The administra- black child
tion procedures of the RATC are quite simple 7 (B Child sitting up in Elicits themes of
(the instructions given to the child were provided & G) bed awake anxiety and bad dreams
8 Both parents Elicits themes related
earlier in this chapter). The RATC provides
speaking to male to family discussions,
explicit instructions for scoring the stories pro- and female child such as around
vided by a child. Each story is scored on 16 cod- discipline or planning a
ing categories. There are eight adaptive family activity
categories, five clinical categories, and three cat- 9 Child standing Elicits themes related
with clenched fists to peer aggression
egories labeled as “indicators”. A description of over a child sitting
these categories is provided in Table 10.4. As evi- on the ground
dent from this box, the RATC coding categories 10 (B Mother holding Elicits themes of
are quite similar to traditional content areas used & G) baby with child sibling rivalry and
looking on attitudes toward the
to interpret other thematic techniques. Scores
birth of a new sibling
used in interpretations from the RATC are the 11 Child cowering Elicits themes of fear
total number of times a given code was present with hands in front and anxiety
across all stories. This allows one to determine of face
consistent themes (high scores within a category) 12 (B Adult male glaring Elicits themes of
across stories. & G) at a distressed parental conflict and
adult female with parental depression
child looking on
Norming One objective of the authors of the 13 (B Child preparing to Elicits themes of anger
RATC was to develop a standardized scoring & G) throw chair onto and aggressive feelings
system so that normative data could be gener- the ground
ated and used by other users of the system (continued)
198 10 Projective Techniques
Table 10.4 Profile scales and indicators from the Roberts Apperception Test for children
Description of
scale Purpose Scoring criteria
Reliance on Designed to measure the adaptive capacity Main character (1) seeks assistance for handling
others (A) to use help to overcome problem problem or completing tasks, (2) asks permission, or
(3) asks for approval or material objects
Support other Reflects tendency to support others Main character (1) gives object or does something
(A) requested or (2) gives emotional support or
encouragement
Support child Measures self-sufficiency, maturity, Main character (1) shows appropriate self-
(A) assertiveness, and positive affect confidence, assertiveness, self-reliance, perseverance,
or delay of gratification or (2) experiences positive
emotions
Limit setting Measures the extent that parent places Story describes some appropriate disciplinary action
(A) reasonable and appropriate limits on child or constructive discussion following child
wrongdoing
Problem Measures child’s ability to articulate Character in story states a problem or obstacle or
identification problem situations experiences contradictory feelings
(A)
Resolution 1 Measures child’s tendency to seek easy or Story involves a situation in which a problem is
(A) unrealistic solutions to problems solved without any clear mediating process or by
some unrealistic or imaginary process
Resolution 2 Measures a child’s tendency to use Story involves constructive resolution of internal
(A) constructive resolutions to problems feelings, external problem, or conflicted interpersonal
relationship; solution is limited to present situation
without an explanation of the process involved in
working through the problem
Resolution 3 Measures a child’s tendency to use Same as Resolution 2, except the story explains how
(A) constructive resolutions to problems the character worked through problem
Anxiety (C) Measures a child’s tendency to interpret Story involves (1) character showing apprehension,
situations as dangerous and fearful self-doubt, or guilt or (2) themes of illness, death, or
accidents
Aggression Measures a child’s aggressive impulses Story involves angry feelings, physical or verbal
(C) attack, or destruction of objects
Depression Measures a child’s tendency to depression Story involves dysphoric feelings, giving up, or
(C) vegetative symptoms of depression
Rejection (C) Measures issues that a child might have Story involves physical separation, rejection, dislike
concerning separation and rejection of another person, needs unmet by others, or racial
discrimination
Unresolved Measures a tendency toward having an Story involves an emotional reaction without a
(C) external locus of control or an inability to resolution
control one’s emotions
Atypical Measures distortions in a child’s thought Story involves a distortion of stimulus card, is an
response processes or unusual ideation illogical story, involves homicidal or suicidal
ideation/action, involves death, or involves child
abuse
Maladaptive Measures poor problem-solving abilities Story involves characters that behave inappropriately
outcome (I) or the presence of a pessimistic or to solve a problem or when the story ends with the
hopeless cognitive style main character dying
Refusal (I) Measures a lack of investment in the task Child refuses to respond or stops in the middle of a
or extreme defensiveness to certain stimuli story and refuses to go on
Source: McArthur, D. S., & Roberts, G. E. (1982). Roberts apperception test for children. Los Angeles, CA: Western
Psychological Services
Note: Descriptions of scoring criteria are not full criteria and should not be used in place of the explicit criteria provided
in the RATC manual. Manual also provides concrete examples of each criteria
(A) adaptive scale, (C) clinical scale, (I) indicator
200 10 Projective Techniques
caution in interpreting the RATC stems from the surveyed a random sample (n = 100) of members
poor normative base from which norm- of the Society for Personality Assessment, and
referenced scores provided in the RATC are they obtained a 60% response rate to their sur-
derived (Bell & Nagle, 1999). These scores vey. On questions related to why and how they
should he regarded as suspect until further infor- used SCTs in their clinical practice, the only
mation becomes available from a larger and response endorsed by the majority of respon-
more representative sample of children and dents (67%) was as “part of a more comprehen-
adolescents. sive assessment battery.” A substantial minority
was split in endorsing “to determine personality
structure” (i.e., as a projective test) and in endors-
Sentence Completion Techniques ing “as a structured interview” (i.e., as a behav-
ioral sample) to describe their use of SCTs, with
Another type of projective technique that is fre- 30% and 25%, respectively, endorsing these
quently used in the clinical assessment of chil- uses. Interestingly, 28% endorsed the use “to
dren is the sentence completion technique (SCT). obtain quotable quotes” as a justification for
In fact, SCTs may be the one projective technique inclusion of SCTs in their assessment battery,
that is used more frequently in the assessment of suggesting that the information provided by the
children and adolescents than in the assessment SCT is often used to obtain examples to illustrate
of adults (Piotrowski, 2018). The sentence com- findings from other assessment procedures (e.g.,
pletion method involves providing the child, clinical diagnoses).
either orally or in writing, a number of incom-
plete sentence stems such as, “My family is …”
or “I am most ashamed of …” As is evident from Features of SCTs
these examples, the stimulus employed in sen-
tence completion techniques is much less subjec- Content Despite a common format, there are
tive than the other projective methods reviewed numerous SCTs available that vary on their con-
in this chapter. That is, the sentence stems have a tent, length, complexity, and purpose. The Rotter
high degree of stimulus pull in prompting certain Incomplete Sentence Blank (Rotter & Rafferty,
types of answers. As a result, many have debated 1950) is one of the oldest and most common of
whether sentence completion methods should the SCTs. It was originally developed for use
even be considered projective, given their more with adults and consists of 40 items designed to
objective nature (Hart, 1986). elicit information on psychosexual conflicts. It is
Although SCTs clearly require a different available in three forms, and the authors provide
level of inference than other projectives, they a quantitative scoring procedure that can be used
probably are closer to other projectives in design to determine the degree of conflict present in
and interpretation than to self-report rating each response. Another commonly used SCT is
scales. However, a decision as to whether or not the Hart Sentence Completion Test for Children
to use an SCT in a clinical assessment goes back (HSCT). The 40-item HSCT was designed spe-
to our initial discussion of projective techniques cifically for use with children, and the content
in general. If one wishes to interpret projectives was designed to elicit children’s perceptions of
as a behavioral sample, then the objective nature family, peers, school, and self (Hart, 1986).
of the SCT and the lower level of inference There are numerous other SCT procedures that
required is a distinct advantage. In contrast, if are beyond the scope of this chapter to discuss in
one wishes to enhance projection, then the detail (see Haak, 2003; Holaday et al., 2000 for
strong stimulus pull of the SCT is less reviews).
desirable.
To illustrate the diversity in how SCTs are Administration Administration of SCTs is
used, Holaday, Smith, and Sherry (2000) straightforward and typically includes instruc-
Sentence Completion Techniques 201
tions that inform children that they are to com- Interpretation As with most projective tech-
plete the sentences in whatever manner they niques, there is great variability in how SCTs are
choose and that there are no right or wrong scored and interpreted. Many SCTs do not have
answers. explicit scoring or interpretive guidelines, and,
There are three important dimensions on even for those that do, many users do not follow
which the administration procedures of SCTs dif- them in practice. For example, of the 60 respon-
fer (Hart, 1986). First, SCTs and users of SCTs dents from the survey of users of SCTs con-
can differ on whether the sentence stems are read ducted by Holaday et al. (2000), only 17% of
aloud to the child or adolescent or whether the those respondents who said they use SCTs in the
child being assessed is asked to read the ques- assessment of children stated that they score the
tions and respond privately. The choice of which test according to a manual or according to the
administration format to use is partly a function authors’ instruction, and 27% of those respon-
of the child’s reading level and age, with asses- dents who said they use SCTs with adolescents
sors tending to read sentence stems to children reported doing so. In fact, 25% of respondents
more often than to adolescents (Holaday et al., did not even know the name of the SCT that they
2000). However, the verbal interchange that used in practice, and 13% reported that they
results from the reading of questions to a child write their own sentence stems to address their
can also provide an assessor with additional client’s needs.
information (e.g., a child’s affective response to a However, interpretation of SCTs typically
sentence stem, child’s apparent motivation relies on an analysis of the manifest content of a
toward the task) on which to evaluate his or her child’s responses. As was the case with the the-
responses. matic techniques, an assessor would analyze a
Second, some users of SCTs request that the child’s response for consistent themes that might
child answer as quickly as possible by saying the provide clues to the child’s emotional adjustment
first thing that comes to his or her mind in an or his or her perceptions of certain persons or
effort to elicit spontaneous and unguarded situations. For example, positive responses to
responses. In contrast, other users attempt to pro- stems designed to assess perceptions of parents
mote deliberation by telling the subjects that they (My father is the best; What I like best about my
can complete the sentences in any way they like father is he is nice) are thought to be an indica-
and that the purpose of the test is to better under- tion of a positive father-child relationship. In
stand their real feelings. The first use of the SCT support of this type of interpretation, Kugler
is typically preferred if the goal of administration et al. (2013) showed that, in a sample of 189 chil-
is projection. The second administration proce- dren ages 6–17, trauma-related responses on an
dure is typically preferred if the goal is to obtain SCT was correlated with post-traumatic symp-
a behavioral sample. toms. This is an example of the low level of
Third, SCTs differ in whether or not they inference that is often applied to the interpreta-
include an inquiry process. In the inquiry phase, tion of SCTs.
children are asked to explain their responses in Some assessors also analyze the process of a
more depth. This questioning helps the assessor child’s responses, such as are they complex, are
determine why a child may have completed the they perseverative, are they expressive and imagi-
sentence in a particular way. This information is native, and are they coherent and related to the
especially useful for responses that are unusual, sentence stem (Haak, 2003). This type of analysis
ambiguous, or diagnostically important (Hart, can provide the assessor with insight as to how
1986). Because of the important clinical informa- invested a child was in the task and some possible
tion obtained by this inquiry, it is often an inte- clues about a child’s thought processes. Box 10.3
gral part of the administration of SCTs for many provides an overview of the most common
assessors (Haak, 2003). approaches to interpreting responses on SCTs.
202 10 Projective Techniques
Box 10.3 Research Note: Interpretive & Rafferty, 1950) and is based on psycho-
Approaches to Sentence Completion dynamic theory. Each response is analyzed
Techniques according to the degree of intrapsychic
conflict evident in the response. It is a
The author of the Hart Sentence Completion quantitative scoring system in which the
Test for Children (Hart, 1986) provided an severity of the conflict is rated as negative,
interesting summary of various interpretive neutral, or positive.
approaches to sentence completion tech-
niques (SCTs) in the clinical assessment of Strategy 4
children and adolescents.
The fourth approach involves compar-
Strategy 1 ing responses on an SCT to some prede-
termined criteria. This approach attempts
The most common interpretive approach to limit the unreliability inherent in the
to SCT is to review each item’s content and other strategies by minimizing the reli-
obtain clinical impressions about a child’s ance on the assessor’s theoretical orienta-
personality dynamics. The assessor tion. An example of this approach is the
searches for patterns, clues, and thought Hart Sentence Completion Test for
processes and generates hypotheses consis- Children (Hart, 1972). In a standardiza-
tent with the assessor’s view of human tion sample, a large pool of responses was
behavior. This approach often leads to dif- obtained for each sentence stem. The
ferent interpretations of the same set of responses were placed into positive, nega-
responses by different clinicians. What is tive, and neutral categories by expert
viewed as important or diagnostic will judges. In each rating category, represen-
depend on the assessor’s theoretical tative responses were identified for each
orientation. sentence stem to aid assessors in making
their determination of the valence of a
Strategy 2 child’s response.
Source: Hart, D. H. (1986). The sen-
The next approach places sentence tence completion techniques. In H. M.
stems into clusters with similar item con- Knoff (Ed.), The assessment of child and
tent that are judged to elicit similar psycho- adolescent personality (pp. 245–272).
logical information. The assessor New York, NY: Guilford
determines if there are important patterns
of responses within a cluster of items.
However, like the first strategy, the inter- Evaluation As is evident from the discussion
pretations are heavily dependent on the of SCTs to this point, most systems do not have
assessor’s orientation. What items deter- explicit and standardized administration, scor-
mine a meaningful cluster and what consti- ing, and interpretive procedures. These decisions
tutes an important pattern of responses are often left to the judgment of the assessor,
within a cluster are based on an assessor’s who can be guided by the advice of the authors
theoretical orientation. of the SCT or other experienced clinicians (e.g.,
Haak, 2003). As a result, most SCTs can be con-
Strategy 3 sidered to fall in the clinical tradition of projec-
tives, with most techniques failing to have
The third approach is exemplified by the well-established psychometric properties
Rotter Incomplete Sentence Blank (Rotter (Anastasi, 1988). Of particular concern is the
Drawing Techniques 203
this person. Koppitz notes that a child’s drawing technique, a child is asked to “Draw a picture of
may (1) be a reflection of a child’s wish, fantasy, everyone in your family, including you, doing
or ideal; (2) be an expression of real attitudes or something” (Burns & Kaufman, 1970, p. 5).
conflicts; or (3) be a mixture of both. To help These instructions emphasize the family engag-
clarify this issue, many assessors either note a ing in some activity, hence the term kinetic. As
child’s spontaneous verbalizations about a figure was the case for the HTP technique, there is an
or ask the child to tell a story about the figure. As inquiry phase in which a child is asked to describe
noted above, the assessor may follow up with an and explain his or her drawing (Cummings,
inquiry phase in which he or she asks specific 1986). The first part of the inquiry typically
questions about the figure such as “Who is he/ involves the child explaining who each figure is
she?” or “Whom were you thinking about while (e.g., name, relationship to the child, age). The
you were drawing?” or “What is he/she thinking child is then asked to describe all the figures,
about? or How does he/she feel?”. what they are doing in the picture, how they are
feeling, and what they are thinking about. After
these initial descriptive questions, the child is
House-Tree-Person asked to tell a story about the drawing, saying
what happened immediately before the actions
A second projective drawing technique is the depicted in the drawing took place and what hap-
house-tree-person (HTP) technique (Cummings, pens next. Finally, the child is asked to describe
1986). In this technique, the child is asked to anything that he or she would change about the
draw a house, a tree, and a person. The order is picture if he or she could.
always the same, and the drawings are done on The popularity of the KFD lies in its ability to
separate sheets of paper. After the drawing, chil- assess a child’s perceptions of his or her family in
dren are asked a series of questions to give them a fun and nonthreatening way. Burns and
an opportunity to describe and interpret the Kaufman (1970) outline a three-part interpretive
objects that were drawn (Cummings, 1986; process that is heavily dependent, not just on the
Koppitz, 1983). According to one of the origina- drawing, but on the inquiry phase that follows.
tors of the HTP technique, the three figures give The first part of the interpretive process is the
insight into different facets of a child’s function- analysis of the actions portrayed in the drawing.
ing (Hammer, 1958). The house is thought to It refers not only to the movements between peo-
elicit feelings associated with the children’s home ple but the energy (e.g., avoidance, conflict, nur-
situation and familial relationship. In contrast, the turance) and emotion (e.g., love, anxiety, anger)
tree is thought to elicit deeper and unconscious captured in the picture. The next part of the inter-
feelings about themselves and their relationships pretive process deals with the style of the family
with their environment. Unlike the self-portrait, drawing. Style refers to the patterns of interac-
the tree is thought to have less pull for conscious tions among significant family members and
self-descriptions and therefore to involve a greater often reflects a child’s defense system (e.g.,
level of projection. And, finally, the drawing of a denial, isolation). The final stage of the interpre-
person is thought to reflect more of a conscious or tation is the symbolic interpretation, which is
semiconscious view of the child’s self, the child’s analogous to the content interpretations of other
ideal self, or a significant other. projective drawing techniques.
has been hotly debated. In a review of the psy- The greatest value associated with projective draw-
ings does not lie in the graphic symbols repre-
chometric properties of drawing techniques, sented on the paper. Rather, the value of the
Cummings (1986) found that the lack of explicit technique may be in the practitioner’s opportunity
scoring and interpretive guidelines for projec- to observe the examinee’s behavior while drawing.
tive drawings has caused most systems to have Drawings provide a nonthreatening beginning
point which should lead to an in-depth exploration
poor reliability. Even for those systems in which of attitudes, feelings, and beliefs via the synthesis
high reliability estimates have been obtained, of direct interviews, third-party interviews, obser-
correlations between drawings and other mea- vations, and test data. (pp. 238–239)
sures of a child’s adjustment have not been con-
sistently shown (Joiner, Schmidt, & Barnett,
1996). Most studies have found that clinicians Conclusions
are unable to distinguish clinically identified
children from non-clinical controls using pro- In this chapter, we have outlined some of the
jective drawings. Even one of the most com- major issues in the debate of when and how to
monly used interpretations from drawings, the use projective testing in the clinical assessment
size of the drawing to interpret the child’s emo- of children and adolescents. As with most
tional state, has met with limited empirical sup- assessment techniques, the problem with pro-
port (Strange, Van Papendrecht, Crawford, jective testing lies not in the techniques them-
Candel, & Hayne, 2010). selves but in the inappropriate purposes for
Of great concern is the use of projective which they are often employed. This issue is
drawings to detect child sexual abuse. exacerbated in the use of projective techniques
Summaries of this research have not been able because of widespread disagreement over the
to find indicators from drawings that have con- basic nature of these techniques. There is con-
sistently and reliability differentiated abused siderable debate over what psychological pro-
from non-abused children across multiple sam- cesses they are designed to measure, and there
ples (Garb, Wood, & Nezworski, 2000). That is, is lack of agreement over what method of inter-
while some graphic indicators (e.g., gentialia, pretation (e.g., clinical or psychometric) is
omission of body parts) have differentiated most appropriate for a given technique. The
abused from non-abused children in individual most important goal of this chapter was to pro-
studies, the results rarely replicate across sam- vide the reader with a clear discussion of these
ples (Allen & Tussey, 2012). issues so that projectives can be used appropri-
This inability to demonstrate the validity of ately, with the assessor clearly recognizing the
projective drawings has led some authors to sug- limitations of whichever interpretive approach
gest that the use of drawings in clinical assess- is used.
ments of children is unethical (Martin, 1983). We have also summarized some of the major
This strong stance has sparked a lively debate methods of projective testing that are used with
(Knoff, 1983). It is clear that content analyses of children. We have discussed inkblot techniques,
drawings have rarely been shown to predict storytelling techniques, sentence completion
overt behavior, yet many users still try to make techniques, and projective drawings. Space lim-
behavioral predictions (e.g., aggression, anxi- itations prevent an exhaustive review of specific
ety, history of sexual abuse) from drawing tech- techniques and interpretive systems. However,
niques. A quote from Cummings’ (1986) review we have attempted to provide selected examples
of projective drawing research seems to summa- of each type of projective method as a basis for
rize a sensible way to view assessment with pro- developing greater expertise in the use and inter-
jective drawings and possibly a good way to pretation of these techniques through further
view projective testing in general: didactic and clinical training.
Conclusions 207
It should be noted that, at this time, many of the The following is an example of the stem/follow-up
question format used by many diagnostic interview
long-standing or widely used structured diagnos- schedules. This question was taken from the questions
tic interviews for children and adolescents have assessing ADHD from the DSM-5 version of the
not been updated to reflect DSM-5 criteria. Our KSAD-PL (Kaufman et al., 2016). The below example
discussion below presents the most recent updates is for child self-report.
available at the time of this writing, describes Is it hard for you to remember what your parents
and teachers say?
some of the main issues that will need to be con- Do your parents or teachers complain that you don’t
sidered in revising prior versions of structured listen to them when they talk to you?
diagnostic interviews for compatibility with Do you “tune people out?”
DSM-5 (Leffler, Riebel, & Hughes, 2014), and Do you get into trouble for not listening?
provides a description of a few diagnostic inter- Scoring
views that align with DSM-IV for further exem- 0 = no information
plars of issues pertaining to the use of these tools. 1 = not present
2 = subthreshold: Occasionally doesn’t listen.
Structured diagnostic interviews consist of a
Problem has only minimal effect on functioning
set of questions that the assessor asks the infor- 3 = threshold: Often (4–7 days/week) doesn’t listen.
mant (e.g., parent or child). There are explicit Problem has significant effect on functioning
guidelines on how responses are to be scored.
Interview questions generally start with a stem
question (e.g., Have you been involved in many that the interviews are continuously updated to
physical fights?). If the stem is answered affirma- keep up with changing diagnostic criteria or
tively, then follow-up questions are asked to modified for some specific application (e.g., out-
determine other relevant parameters such as fre- come measure for treatment of childhood depres-
quency (e.g., How many fights have you been in sion, screening device for an epidemiological
the past year?), severity (e.g., Have you ever used study of child disorders). Therefore, many of the
a weapon in a fight?), onset (e.g., When was the interviews have multiple versions and are con-
first time you got in a fight?), and impairment tinually being revised.
(e.g., Has fighting caused problems for you at Most interviews organize questions by diag-
school, home, or with kids your age?). An exam- nosis. Which DSM disorders are assessed is gen-
ple of the question format from the Schedule for erally quite similar across the different interview
Affective Disorders and Schizophrenia for schedules. All of the interviews can be used to
School-Age Children-Present and Lifetime assess for the disruptive behavior disorders,
Version (KSADS-PL) DSM-5 interview affective disorders, and anxiety disorders in chil-
(Kaufman et al., 2016) is provided in Table 11.1. dren and adolescents. Each of these interviews
Note from Table 11.1 that the KDSADS-PL is also allows for at least a brief screening for
semi-structured interview, and as a result, there is schizophrenia. The majority of structured inter-
some subjectivity to scoring on the part of the views also allow for an assessment of substance
interviewer, and the interviewer is advised to also use, elimination, and eating disorders. Tic disor-
use observational data in arriving at a score. ders are covered by the CAPA and K-SADS. The
CAPA is unique in its detailed assessment of
sleep disorder symptoms.
ommonalities Across Interview
C To promote multi-informant assessments,
Schedules most interviews contain parallel forms to ask
identical questions of both the child and a parent.
In Table 11.2, we summarize the basic character- This format allows for easy, direct comparison of
istics of some of the most commonly used diag- the child’s symptoms as reported by each infor-
nostic interviews for children. It should be noted mant. Most current structured interviews allow
Overview 211
administration via computer. The computer for- substantial variation in the amount of “leeway”
mat was designed to enhance the reliability of the given the assessor across the various interviews.
interviews by increasing the ease of administra- For example, the K-SADS is one of the least
tion and data collection. Similar to a pencil-and- structures of the interviews. Typically, the inter-
paper version, the computer format was designed viewer is encouraged to begin by building rap-
to be administered by an examiner. The examiner port in what would necessarily be an unstructured
reads items from the computer screen and enters format. As the clinician moves toward the inter-
the client’s responses. The computer quickly view questions, the conversation becomes more
scores and stores responses, selects the appropri- structured and is focused on sample questions
ate follow-up questions, and skips out of diag- such as “Do you feel sad?” “Do you cry?” The
nostic sections when the respondent fails to meet interview then asks other questions revolving
a certain threshold of severity. Also, the computer- around such symptoms (Edelbrock & Bohnert,
administered interviews have programs that 2009).
quickly score the interview and provide various In contrast, the DISC was designed to have a
summary scores (e.g., symptom indexes, number high degree of structure in administration. The
of diagnostic thresholds met) that aid in the inter- manual for its administration includes the follow-
pretation of the results. ing instructions:
Most of the interviews were designed to assess The DISC symptom questions are designed to be
children and adolescents between the ages of 8 read exactly as written. There is very limited scope
and 17. However, many interviews were designed for independent questioning. DO NOT deviate
with parent-report formats that enable assess- from the prescribed question sequence. DO NOT
make up your own questions because you think
ment of younger children. Self-report forms of you have a better way of getting at the same infor-
structured interviews also are generally oriented mation, or because you think the question is poorly
toward the same age range. The length of time worded. (Fisher, Wicks, Shaffer, Piacentini, &
that it takes to administer a diagnostic interview Lapkin, 1992, p. 31)
is heavily dependent on the child being assessed.
Because of the stem/follow-up question format, The trade-off between leeway and structure is
children with more symptoms will require more obvious. Less structure allows the assessor to tai-
interview time because of the need to ask more lor the interview according to the needs of the
follow-up questions. However, as is evident from individual client. However, these interviews gen-
Table 11.2, the average time to administer the erally require a greater degree of experience to
interviews does not vary much across the differ- administer and often have lower levels of reliabil-
ent schedules and lasts typically from 60 to ity (see Kaplan & Saccuzzo, 2018).
90 min. Another major variation among the structured
interviews for children is the time frame used to
assess symptoms and diagnoses. All of the inter-
ajor Sources of Variation Across
M views discussed herein assess whether problems
Interview Schedules are currently evident. This is called a present
episode frame of reference. Most interviews con-
From the previous discussion, it is clear that the sider present episodes to be within the previous
various interview schedules probably have more 6 months, although in some instances the time
similarities than differences. However, one of frame may be as short as within the last 2 weeks
the major differences across schedules is the (e.g., CAPA) or as long as within the last year
degree of structure inherent in the interview for- (e.g., DISC-IV for conduct disorder). Of note,
mat. All of the interviews provide some degree the CAPA restricts the assessment of symptoms
of structure and give guidelines for standardized to the previous 3 months due to concerns with
administration and scoring. However, there is the reliability of memory in children and adults
Evaluation of Diagnostic Interviews 213
over longer time intervals. Similarly, the efficient behavior rating scales? Some of the
Dominic-R does not obtain any temporal infor- more important advantages of structured diag-
mation, such as onset and duration of symptoms, nostic interviews follow.
given concerns as to the validity of such infor-
mation in young children. Nevertheless, the 1. Structured interviews are useful in obtaining
major source of variation is whether or not inter- important parameters of a child’s behavior
view schedules are limited to present episodes. A that are not typically assessed by most behav-
number of interviews restrict the focus of assess- ior rating scales. Specifically, most interview
ment to the present episode time frame (e.g., schedules provide questions that elicit infor-
CAPA and ChIPS). However, an increasing mation on the duration of a child’s behavioral
number of interviews provide for the assessment difficulties and the age at which the problems
of both present and lifetime diagnoses. For began to emerge. This temporal information
example, the DISC-IV provides a more recently allows one to take a developmental perspec-
incorporated whole-life module assessing for tive in understanding a case.
whether or not a child has exhibited symptoms of 2. Interviews also allow one to determine the
diagnoses since age 5 but prior to the current temporal sequencing among behaviors. For
year (Shaffer et al., 2000). Similarly, there are example, it is important to determine whether
lifetime formats for both the ISCA and the periods of sadness occurred contiguously with
K-SADS. The DICA is unique in its exclusive other issues and impairments associated with
focus on lifetime diagnoses. depression (e.g., thoughts of death and dying,
A third source of variation within the inter- changes in sleep or appetite) to start to deter-
view schedules is the answer format. For many mine whether or not they meet criteria for a
structured interviews, the responses are coded major depressive episode (Klein, Dougherty,
into a categorical format (yes, no). This categori- & Olino, 2005).
cal format is consistent with the DSM orienta- 3. Most structured interviews assess the level of
tion in which symptoms are considered either impairment associated with behaviors being
present or absent. In contrast, the CAPA and reported. Specifically, most interviews have
K-SADS have answer formats that can be placed questions that elicit information on the degree
on a Likert-type scale that allows one to rate the to which a child’s difficulties are affecting his
severity of a symptom. Although this format or her functioning in major life arenas (e.g., at
makes it more difficult to translate responses home, at school, and with peers).
into DSM diagnoses, it does not create an artifi- 4. Diagnostic interviews also enhance the corre-
cial dichotomy between the presence and spondence between assessment techniques
absence of a symptom and allows symptom and diagnostic criteria. As mentioned previ-
scores to reflect gradations in severity. ously, the most common structured interviews
were specifically designed and revised to cor-
respond to the changing DSM system.
Evaluation of Diagnostic Interviews Therefore, the usefulness of the interview is,
in part, dependent on the usefulness of the
Advantages diagnostic definitions that are being assessed.
There are some diagnostic categories that
Structured interviews share with behavior rating have been well-validated and others for which
scales the goal of obtaining a detailed description the evidence supporting their utility is limited.
of a child’s emotions and behaviors from multi- However, this connection between assessment
ple informants. The logical question is: What and diagnosis can be advantageous for several
advantages do the time-consuming structured reasons: First, it promotes revisions of the
interviews offer in comparison to the more time- interviews to correspond with advances in our
214 11 Structured Diagnostic Interviews
knowledge of the basic characteristics of child 1. The time-consuming nature of the interviews
and adolescent psychological disorders. coupled with some question as to whether
Second, it allows one to make a diagnosis they provide incremental validity in the
based on strict adherence to diagnostic crite- assessment of some disorders such as ADHD
ria. Either due to theoretical, empirical, or (Pelham et al., 2005) suggests that structured
practical reasons (e.g., insurance reimburse- interviews may not be practical or useful in
ment), many clinicians attempt to make DSM many situations.
diagnoses as a result of their assessments. Too 2. In addition, structured interviews depend on
often, diagnoses are made based on informa- DSM criteria, which is a strength for assess-
tion (e.g., rating scales, projective tests) that ing well-validated syndromes but a weakness
do not directly assess the diagnostic criteria or for assessing disorders with a weak empirical
through techniques that are unsystematic in basis.
their assessment of symptomatology and asso- 3. Also, diagnostic interviews are subject to the
ciated features (e.g., unstructured interviews; same potential reporter biases that were dis-
Klein et al., 2005). As a result, the meaning of cussed in previous chapters on self-report
the diagnosis is ambiguous. Structured diag- inventories and parent and teacher behavior
nostic interviews do not have this problem. rating scales.
They tend to result in more reliable diagnoses 4. In addition, making differential diagnoses
(Silverman & Ollendick, 2005) and can pro- with the assistance of structured interviews
vide important information that may alter still often does directly translate to plans for
diagnostic impressions and treatment recom- intervention, given that many diagnoses are
mendations (Kashner et al., 2003). They may not directly tied to treatment (Mash &
also help rule-in or rule-out potential diagno- Hunsley, 2005).
ses that were under consideration, but were 5. There is great difficulty in making norm-
not clear, from other assessment tools. referenced interpretations from interviews.
5. Diagnostic interviews are also helpful in
Clinically significant levels of symptoms are
training clinical assessors. As assessors are often based on DSM criteria rather than based
developing their competence in interviewing, on a comparison with a representative norma-
it is often helpful to have an explicit format tive sample. Therefore, the appropriateness of
from which to conduct the interview. It gives clinical elevations for a given age depends on
the assessor a good way to learn the basic the appropriateness of the diagnostic criteria
characteristics of childhood emotional and for that age. For example, age-typical mani-
behavioral disorders. After being trained in festations of anxiety (e.g., distress due to sep-
administration procedures and after conduct- aration from caregivers) should be
ing several interviews with actual clients, differentiated from the same manifestations of
assessors often begin to internalize the diag- anxiety that a child typically overcomes by a
nostic criteria for the most common disorders certain age (Pine et al., 2011).
of childhood. This knowledge can then be 6. Another weakness of most diagnostic inter-
applied in situations in which a structured views is the failure to provide a format for
interview is not possible. obtaining information from a child’s teacher.
As a result, information from teachers must be
obtained by some other method, thereby mak-
Disadvantages ing it difficult to determine whether discrep-
ancies between a teacher’s report and the
Diagnostic interviews also have a number of reports of others are due to real differences in
weaknesses of which the clinician should be a child’s classroom behavior or due to differ-
aware (the first two were mentioned earlier). ences in the assessment format.
Recommendations for Use of Structured Interviews 215
Recommendations for Use
of Structured Interviews that she was having difficulty paying atten-
tion and was daydreaming, interrupting
Based on the strengths and weaknesses of struc- others, and making careless mistakes in her
tured interviews, several recommendations can work. Her parents requested a comprehen-
be made on their appropriate use in the clinical sive evaluation to determine the severity
assessment of children and adolescents. First, and possible cause of these difficulties and
like any assessment technique, the diagnostic to get recommendations for possible inter-
interview should never be used alone in a clinical ventions to aid in his school adjustment.
evaluation. It should be one part of a comprehen- Chloe’s background, developmental,
sive assessment battery. For example, informa- and medical history were unremarkable.
tion from diagnostic interviews should be During testing, Chloe had great difficulty
supplemented by assessment techniques that pro- concentrating and was easily distracted.
vide better norm-referenced scores (e.g., behav- She was also very fidgety and restless.
ior rating scales) and by assessment techniques Intellectually, Chloe had much better ver-
that provide information on a child’s classroom bal reasoning abilities than nonverbal
functioning (e.g., behavioral observations in the perceptual-organizational abilities.
school). Consistent with her verbal abilities, Chloe
In addition, the diagnoses derived from the scored in the above average range on mea-
diagnostic interviews should be viewed within sures of reading and math achievement.
the context of the overall assessment. Such a Chloe’s emotional and behavioral function-
diagnosis can be viewed similarly to the way an ing were assessed through the use of struc-
elevation on a behavior rating scale is interpreted. tured interviews conducted with Chloe, her
Specifically, it is one piece of information that parents, and her teachers as well as through
needs to be integrated with other sources of infor- rating scales completed by his parents and
mation to develop a good case formulation. teacher. The structured interview was the
Stated simply, diagnoses based solely on diag- parent version of the KSAD-PL adminis-
nostic interviews should not be considered final tered to Chloe’s mother, and the child ver-
clinical diagnoses. Such final diagnoses should sion administered to Chloe.
be based on an assessor’s integration of multiple The following is an excerpt from the
sources of information (see Chap. 15 for a report of Chloe’s evaluation that illustrates
detailed discussion of integrating multiple how information from the KSADS-PL was
aspects of an assessment). A case example in integrated with other assessment informa-
which a diagnostic interview was used as part of tion in a concise manner.
a comprehensive assessment battery is provided The only problematic areas that emerged
in Box 11.1. from this assessment of Chloe’s emotional
and behavioral functioning were significant
problems of inattention, disorganization,
impulsivity, and overactivity that seem to be
Box 11.1 Case Example: The KSADS-PL in the causing her significant problems in the
Evaluation of a 9-Year-Old Girl with classroom. On the KSADS-PL, Chloe’s
Attention-Deficit Hyperactivity Disorder mother reported that such difficulties were
noticeable by the time she was age 6. Chloe
herself endorsed many symptoms of inat-
Chloe is a 9-year, 3 month-old girl who tention, but not hyperactivity/impulsivity,
was referred for a comprehensive psycho- on the KSADS-PL. In an unstructured
logical evaluation by her parents upon the interview, Chloe’s teacher described her as
recommendation of her teachers. Her being very restless and fidgety, being easily
teachers had reported to Chloe’s parents (continued)
216 11 Structured Diagnostic Interviews
history, family psychiatric history, the child’s perfectly accurate, and integrate it with other
educational status, the child’s peer relationships, information to derive the best evidence-based
and the child’s interests and hobbies. conclusions about the child’s functioning.
There is not an introductory script or set of The KSADS-PL has a long tradition of use
instructions for the interviewer to read, but there and reasonable adaptations for use in clinical
are some guidelines that may be conveyed to par- practice and research. Prior versions of the
ents or children, particularly for more lengthy KSADS-PL have demonstrated good reliability
structured interviews when multiple domains and convergent validity across domains. It is also
will be covered. These points include: unique among structured interviews for children
in that it has been updated for DSM-5. However,
1. There are no right or wrong answers. The best at the time of this writing, there is no independent
answer is the one that tells the most about the research on the psychometric properties of the
child. most recent version.
2. The informant should try to answer “yes” or
“no” to those questions that are oriented
toward “yes” or “no” responses. Conclusions
3. Many of the questions on the form will be left
out depending on responses. Structured diagnostic interviews have become an
4. Some questions may be asked more than once. important part of many clinical assessments of
5. It is possible to take breaks, if needed. children and adolescents. Like behavior rating
scales, diagnostic interviews provide a reliable
In the KSADS-PL, some notes are provided to means of assessing a child’s emotional and
guide interpretation or follow-up. For example, behavioral functioning. In this chapter, we have
in the depression screening, the following is attempted to highlight the advantages and disad-
stated: vantages of using diagnostic interviews.
Sometimes the child will initially give a negative Diagnostic interviews enhance clinical assess-
answer at the start of the interview but will become ments by providing a format for determining how
obviously sad as the interview goes on. Then these long a child’s problems have been occurring, for
questions should be repeated eliciting the present determining the temporal sequencing of behav-
mood and using it as an example to determine its
frequency. Similarly, if the mother’s report is that iors, and for estimating the degree of impairment
the child is sad most of the time, and the child associated with a child’s emotional or behavioral
denies it, the child should be confronted with the problems. These important parameters of a
mother’s opinion and then asked why he thinks his child’s emotional and behavioral functioning are
mother believes he feels sad so often.
often not assessed by other assessment modali-
This guidance is provided to acknowledge the ties. In addition, diagnostic interviews are typi-
possibility of underreporting, but we caution cli- cally tied to the most recent revisions of the DSM,
nicians to proceed carefully with this type of fol- which closely links assessment with this system
low-up. First, as discussed elsewhere in this of classification.
volume, some degree of informant discrepancy is On the negative side, diagnostic interviews are
to be expected. In addition, it may be preferable often time intensive, and they typically do not
to protect parent or child confidentiality rather provide any norm-referenced information on a
than confronting the other party with discrepant child’s functioning above that which is accorded
information. Further, directly pointing out a by DSM criteria. In addition, diagnostic inter-
child’s mood, particularly if the subjective views typically do not include a format for
impressions are based only on subtle cues, could obtaining information from a child’s teacher, and
cause the child undue anxiety. As with any assess- their reliability in obtaining self-report informa-
ment tool, the clinician’s task is to take the infor- tion for young children (i.e., below age 9) is
mation provided, recognizing that it may not be somewhat questionable, although pictorial or
Conclusions 219
Gerald Patterson (Patterson 1982) and Sargent, Liebman, and Silver (1985)
his colleagues at the Oregon Social describe family characteristics that pro-
Learning Center developed a social learn- vide a context in which the psychological
ing model for the development of aggres- features of anorexia nervosa fit and are
sion. In this model, family interactions adaptive. Families of a person with
provide a training ground for a child to anorexia have been found to often have
learn coercive methods of controlling inter- parents who are overinvolved in their
actions with others. Through analyses of child’s life. This overinvolvement prevents
Why Is Assessing Family Context Important? 223
Two other facts emerge from research on child- Box 12.2 Family History and Differential
hood psychopathology that point to the impor- Diagnosis: Case Study of a 7-Year-Old Girl
tance of assessing a child’s family context in with Social Anxiety
clinical assessments. First, childhood emotional
and behavioral problems tend to be rather amor- Harper is a 7-year, 2 month-old girl who
phous, lacking clear boundaries, more so than is was in the middle of the first grade when
the case in adult psychopathology (see her teacher recommended that she be tested
Lilienfeld, 2003). Stated another way, children at an outpatient mental health clinic. Her
with problems often have multiple types of teacher was concerned that she might have
problems, and it is often difficult to know what Attention-Deficit Hyperactivity Disorder.
is primary and what might be secondary. Second, Harper seemed bright and capable of learn-
there seems to be a parent-child link to many ing and, in fact, performed quite well in
types of psychopathology, with parents and chil- one-on-one situations with the teacher or
dren showing similar patterns of adjustment teacher’s aide. However, in the general
(McMahon & Dev Peters, 2002). Taking these classroom setting, Harper rarely finished
two facts together, assessment of the adjustment her work. She was often noted to be staring
of parents in whom the type of problem may be off into space, and she had to be constantly
more clearly defined, may provide clues to the (continued)
primary problem of the child.
224 12 Assessing Family Context
were chosen for review were parent report or Parenting Styles and Practices
child report measures of family functioning.
This focus was for several reasons. First, these There is broad consensus that parenting behav-
methods typically are the most time-efficient iors exert a significant influence on child devel-
method for collecting information on the child’s opment. There is less consensus regarding the
family. Second, these measures are often stan- specific aspects of parenting that are most crucial
dardized and easily obtainable. Third, these rat- to child adjustment. However, Darling and
ing scales tend to have the best normative data Steinberg (1993) provide a good context for con-
that allow for interpretations of scores based on ceptualizing parenting and its effect on child and
some comparison group. Thus, these rating adolescent adjustment. Darling and Steinberg
scales tend to be the most useful in many clinical (1993) divide parenting into two main compo-
assessments. nents: parenting styles and parenting practices.
However, such assessment methods are not These authors define parenting styles as “a con-
without limitations. Marsbach and Prinz (2006) stellation of attitudes toward the child that are
reviewed eight measures that use parent report communicated to the child and create an emo-
of their own parenting behaviors. Their evalua- tional climate in which the parents’ behaviors are
tion of these measures suggests that most dem- expressed” (p. 493). These authors use
onstrated acceptable internal consistency (0.70 Baumrind’s (1971) classic typology to exemplify
and above). Further, most measures showed parenting style. Baumrind divides parenting
moderate concordance between parent and child styles into three types. The authoritarian style is
ratings of parenting (0.23–0.37) with somewhat characterized by a rule-adherence orientation that
higher concordance in the report between two emphasizes compliance and deemphasizes auton-
parents and between parent report and observa- omy and emotional support. The permissive style
tions of parenting behavior. However, these is a child-centered style in which child autonomy
authors noted that parents are often asked to is of primary importance, and rules and demands
make estimates of high-frequency behaviors are minimal. The authoritative style is character-
(e.g., yelling) over long periods of time (e.g., ized by emotional support and respect for appro-
6 months) which may make accurate reports dif- priate autonomy in the child but in the context of
ficult. Further, they noted that many of the ques- clearly defined and consistently enforced rules. It
tions deal with sensitive issues that may not be is this last parenting style, authoritative, that
socially desirable and may be considered intru- research has consistently linked to more healthy
sive by parents. As a result, such questions could child adjustment (Hoeve, Dubas, Gerris, van der
result in biases in their responses. Finally, these Laan, & Smeenk, 2011).
authors also noted that many scales often include In contrast to parenting style, parenting prac-
vague quantifiers (e.g., frequently, sometimes, tices are defined as the techniques used by the
never) that may also influence the accuracy of parent to socialize their child and enforce rules.
parents’ responses. For example, a specific discipline practice (e.g.,
Thus, although we have chosen to focus degree of corporal punishment), use of positive
largely on rating scales for the assessment of parenting strategies (e.g., praise and reward for
family functioning for the reasons noted above, it appropriate behavior), consistency in parenting,
is important to recognize the limitations in the and appropriate supervision and monitoring of a
information obtained by this assessment format. child’s behavior by a parent are all examples of
As noted by McMahon and Frick (2007), infor- parenting practices that have been linked to child
mation obtained by these measures should be adjustment (Frick, 1994).
interpreted with other assessment information, The unique contribution of Darling and
such as interviews and behavioral observations, Steinberg’s model of parenting is not only its
whenever possible. explicit distinction between parenting style and
230 12 Assessing Family Context
Table 12.1 Subscales of the Family Environment Scale Norms The FES manual reports information on
(FES) a large normative sample that included 1125
Description of item families from all regions of the country, single-
Dimension Subscale content parent and multigenerational families, families
Relationship Cohesion Commitment, help,
drawn from ethnic minority groups, and families
and support
provided by family of all age groups (Moos & Moos, 1986). It is
members unclear how representative the sample is on each
Expressiveness Extent to which of these variables. However, the authors note
family members are that 294 families from the normative sample
encouraged to
express feelings
were drawn randomly from specified census
Conflict Amount of anger, tracts in the San Francisco area, and the means
aggression, and and standard deviations of FES scales did not
conflict among differ between this group and the rest of the nor-
family members mative sample.
Personal Independence Extent to which
growth self-sufficiency, Although the normative group includes fami-
assertiveness, and lies of all age groups, it is notable that the major-
independence are ity of the normative sample was based on the
encouraged in the
reports of adults, with much less data available
family
Achievement Extent to which on the reports of children and adolescents. The
orientation activities of family reliance on parent report is important because the
members are authors found small but systematic differences
achievement- between the scales completed by parents and
oriented and
competitive adolescents (Moos & Moos, 1986). Specifically,
Intellectual- Degree of interest in adolescents perceived less emphasis on cohesion,
cultural political, social, and expressiveness, independence, and intellectual/
cultural activities religious orientation and more emphasis on con-
orientation
flict and achievement than did their parents.
Active- Emphasis placed on
recreational participation in
social and Reliability and Validity The ten subscales of
recreational-oriented the FES generally have been shown to have
activities acceptable levels of reliability in many samples.
Moral-religious Importance placed
The manual reports internal consistency esti-
emphasis on ethical and
religious issues mates in a large community sample (n = 1067)
System Organization Importance placed ranging from alpha = 0.61 to 0.78 (Moos &
maintenance on having a clear Moos, 1986). Two-month test-retest reliability
family structure and in a smaller community sample (n = 47) ranged
well-defined roles
from r = 0.68 to 0.86. One note of caution for
Control Degree to which
rules and procedure the reliability of the FES is the finding that the
for family are reliability of the scales may be lower in adoles-
explicit cent samples (Boyd, Gullone, Needleman, &
Source: Moos and Moos (1986) Burt, 1997).
Moos and Moos (1986) provide a good sum-
influence is evident from the emphasis on family mary of over 100 research articles using the FES,
structure and organization and the focus on the which attests to its correlation with other mea-
transactional patterns between members of the sures of family functioning, its ability to differen-
family in the FES item content. Also, as evident tiate distressed from nondistressed families, and
from the content, the FES is best thought of as a its sensitivity to treatment effects. Many more
measure of parenting style or family climate, studies have been published since this review
rather than of specific parenting practices. (Hamilton & Carr, 2016).
Parenting Styles and Practices 233
As noted above, the FES has been widely used Most of the research using the APQ has used
with several different clinical populations of chil- the questionnaire formats. Items on this format
dren and adolescents. For example, using the are rated on a 5-point Likert-type frequency scale
FES, Suveg et al. (2005) noted that mothers of and ask the informant how frequently each of the
children with an anxiety disorder showed less various parenting practices typically occurs in
emotional expressiveness than non-clinic-the home. On the telephone interview format,
referred children (ages 8–12). As another exam- four interviews are conducted with parents and
ple, Pressman et al. (2006) found that families of children with at least 3 days between each inter-
children and adolescents with Attention-Deficit view. The informant is asked to report the fre-
Hyperactivity disorder reported higher rates of quency with which each parenting practice has
family conflict on the FES. Finally, Lucia and occurred over the previous 3 days, and responses
Breslau (2006) reported that higher levels of fam- for each item are averaged across the four
ily cohesion, as measured by the FES when chil- interviews.
dren were age 6, were associated with fewer
emotional and attentional problems in the chil- Content The content of the APQ was devel-
dren at age 11. oped to assess the five dimensions of parenting
Probably the biggest threat to the validity of that have been most consistently related to
the FES is the failure to validate the scale struc- behavior problems in youth: Involvement,
ture through factor analyses (Hamilton & Carr, Positive Parenting, Poor Monitoring/
2016). The scales were designed primarily based Supervision, Inconsistent Discipline, and
on content and face validity. Unfortunately, fac- Corporal Punishment (Shelton et al., 1996). It
tor analyses have generally isolated anywhere also includes several other items assessing “other
from two (e.g., Fowler, 1982) to six (Sanford, discipline practices,” such as use of time out or
Bingham, & Zucker, 1999) or seven (Robertson taking away privileges. The items on the APQ
& Hyde, 1982) factors on the FES. No study has and its subscales are provided in Table 12.2. The
provided convincing evidence supporting the items used on the 9-item short version are also
10-scale structure that is the basis for most inter- designated in this table (Elgar et al., 2007).
pretations from the FES. Studies using the APQ often have used scores
from the individual scales (e.g., Frick, Christian,
& Wootton, 1999), or they have used composites
Alabama Parenting Questionnaire of these scales (e.g., Frick, Kimonis, Dandreaux,
(APQ) & Farrell, 2003). There are two common ways of
forming composite scores. The first way is to
The APQ (Shelton, Frick, & Wootton, 1996) is a standardize the scores (e.g., create Z-scores) and
measure of parenting behavior that was devel- then combine the Involvement and Positive
oped for use with parents of elementary school- Parenting scales into a Positive Parenting
aged children and adolescents (6–17 years old). Composite and the Poor Monitoring/Supervision,
However, it has been used in samples as young as Inconsistent Discipline, and Corporal
age 3 and 4 (Clerkin, Marks, Policaro, & Punishment scales into a Negative Parenting
Halperin, 2007; Dadds, Maujean, & Fraser, 2003) Composite. The second method for forming
with some modification of its content. It consists composites from the APQ is to create a single
of 42 items that are presented in both global dysfunctional parenting composite by standard-
report (i.e., questionnaire) and telephone inter- izing all five scales, inversely scoring the
view formats, and there are separate versions of positive parenting scales (Z-score ∗ −1), and
each format for parents and children. Also, Elgar, summing all five dimensions.
Waschbusch, Dadds, and Sivaldason (2007) have
developed a 9-item short version of the scale for Norms One of the main limitations in the APQ
use as a brief screener. is the lack of norm-referenced scores that can be
234 12 Assessing Family Context
Table 12.2 Subscales of the Alabama Parenting 12. You feel that getting your child to obey you is
Questionnaire more trouble than it’s worth
Involvement 22. You let your child out of a punishment early
1. You have a friendly talk with your child (like lift restrictions earlier than you originally
4. You volunteer to help with special activities that said)a
your child is involved in (such as sports, boy/girl 25. Your child is not punished when he/she has done
scouts, church youth groups) something wrong
7. You play games or do other fun things with your 31. The punishment you give your child depends on
child your mood
9. You ask your child about his/her day in school Corporal punishment
11. You help your child with his/her homework 33. You spank your child with your hand when he/
14. You ask your child what his/her plans are for the she has done something wrong
coming day 35. You slap your child when he/she has done
15. You drive your child to a special activity something wrong
20. You talk to your child about his/her friends 38. You hit your child with a belt, switch, or other
object when he/she has done something wrong
23. Your child helps plan family activities
Other discipline practices
26. You attend PTA meetings, parent/teacher
conferences, or other meetings at your child’s 34. You ignore your child when he/she is
school misbehaving
Positive parenting 36. You take away privileges or money from your
child as a punishment
2. You let your child know when he/she is doing a
good job with somethinga 37. You send your child to his/her room as a
punishment
5. You reward or give something extra to your child
for obeying you or behaving well 39. You yell or scream at your child when he/she
has done something wrong
13. You compliment your child when he/she does
something wella 40. You calmly explain to your child why his/her
behavior was wrong when he/she misbehaves
16. You praise your child if he/she behaves wella
41. You use time out (make him/her sit or stand in
18. You hug or kiss your child when he/she has done
corner) as a punishment
something well
42. You give your child extra chores as a
27. You tell your child that you like it when he/she
punishment
helps around the house
Poor monitoring/supervision
a
Items that are included in the 9-item short screening
6. Your child fails to leave a note or to let you know version of the APQ (Elgar et al., 2007)
where he/she is goinga
10. Your child stays out in the evening past the time used to interpret the scales. However, there are
he/she is supposed to be homea now several studies that provide scores from
17. Your child is out with friends you do not knowa fairly large samples of non-referred children,
19. Your child goes out without a set time to be
from which cut scores can be developed. One
home
21. Your child is out after dark without an adult with
study was conducted with 1402 children ages 4–9
him/her (Elgar et al., 2007) in Australia. A second study
24. You get so busy that you forget where your child was conducted with 1219 German school chil-
is and what he/she is doing dren ages 10–12 (Essau, Sasagawa, & Frick,
28. You don’t check that your child comes home 2006). A third study was conducted with 674
from school when he/she is supposed to
children ages 9–12 in Canada and the United
29. You don’t tell your child where you are going
30. Your child comes home from school more than
States (Maguin, Nochajski, DeWit, & Safyer,
an hour past the time you expect him/her 2016) A fourth study provided data on 790 chil-
32. Your child is at home without adult supervision dren ages 6–10 in the United States (Randolph &
Inconsistent discipline Radey, 2011). Finally, in one of the only studies
3. You threatened to punish your child and then do to provide normative data on an adolescent sam-
not actually punish him/hera ple, Zlomke, Bauman, and Lamport (2015) pro-
8. Your child talks you out of being punished after
vided data on 358 adolescents ages 11–18 in the
he/she has done something wronga
United States.
Parenting Styles and Practices 235
Reliability and Validity As noted previously, Discipline being most strongly associated in
most of the published research using the APQ to young children (ages 6–8), Corporal Punishment
date has utilized the global report formats of the being most strongly associated in older children
scale, with the exception of Shelton et al. (1996). (ages 9–12), and Involvement and Poor
Across studies, the reliability and stability of Monitoring/Supervision being most strongly
APQ scores have generally been acceptable with related in adolescents (ages 13–17). Parental
several notable exceptions. First, the internal involvement and parental use of positive rein-
consistency of the short 3-item Corporal forcement, as measured by the APQ, have also
Punishment scale has often been quite low on all been significantly associated with levels of CU
formats. Second, the internal consistency of the traits in an ethnically diverse sample of kinder-
Poor Monitoring/Supervision scale has been low garten students (Clark & Frick, 2018) and in a
in the interview format (all alphas below 0.50, sample of at-risk school age youth (Pardini,
Shelton et al., 1996). Third, Frick, Lilienfeld, Lochman, & Powell, 2007)
Ellis, Loney, and Silverthorn (1999) showed poor Although the most common use of the APQ
reliability of the child report formats in very has been to assess parenting in families of youth
young children (before age 9) (see also Shelton with conduct problems, it has been used to assess
et al., 1996). family correlates to anxiety disorders (Pfiffner &
Several studies have provided factorial sup- McBurnett, 2006) and to assess parenting in fam-
port for the five dimensions around which the ilies of depressed parents (Cummings, Keller, &
scale was developed (Elgar et al., 2007; Essau Davies, 2005). Finally, several studies have used
et al., 2006; Zlomke, Lamport, Bauman, Garland, the APQ scales to test changes in parenting
& Talbot, 2014). Also, parent ratings on the APQ behaviors following treatment (e.g., Babinski,
are significantly associated with observations of Waxmonsky, & Pelham, 2014; Crandal et al.,
parenting behavior in 4- to 8-year-old boys 2015; Haack, Villodas, McBurnett, Hinshaw, &
(Hawes & Dadds, 2006) and children ages 9–15 Pfiffner, 2017).
(Parent et al., 2014). However, the most common
use of the APQ has been to study parenting in
families of children with conduct problems. An Parenting Scale (PS)
association between APQ scales and conduct
problems has been reported in community (Crum, The PS (Arnold, O’Leary, Wolff, & Acker,
Waschbusch, Bagner, & Coxe, 2015; Dadds 1993) is another commonly used measure of
et al., 2003), clinic-referred (Frick et al., 1999; parenting. It is a 30-item parent report scale that
Hawes & Dadds, 2006; Muratori et al., 2015), focuses specifically on parents’ attitudes and
and inpatient samples (Blader, 2004), as well as beliefs about discipline. The items are all rated
in families with deaf children (Brubaker & on a 7-point scale in which the parent is asked
Szakowski, 2000) and families with substance- to estimate the probability with which they
abusing parents (Stanger, Dumenci, Kamon, & would use a particular discipline strategy (e.g.,
Burstein, 2004). Also, these studies have docu- when my child misbehaves, I spank, slap, grab,
mented this relationship in samples as young as or hit my child). It was originally developed for
age 4 (Osa, Granero, Penelo, Domenech, & use with young children (ages 18–48 months;
Ezpeleta, 2014; Hawes & Dadds, 2006) and as Arnold et al., 1993) and has primarily been used
old as age 17 (Dandreaux & Frick, 2009; Frick to assess parenting in preschool children.
et al., 1999). However, there is evidence for its utility in sam-
Importantly, Frick et al. (1999) reported ples of children as old as 12 years of age (Lorber
some differences in which dimensions of par- & Slep, 2015; Prinzie, Onghena, & Hellinckx,
enting were most strongly associated with con- 2007; Steele, Nesbitt-Daly, Daniel, & Forehand,
duct problems at different ages, with Inconsistent 2005).
236 12 Assessing Family Context
approximately 15 minutes to complete. However, parent. The last two scales, Problems with
a brief 24-item version of the scale has been Family and Problems with Others, assess the
developed and has proven to be highly correlated level of family conflict in the extended family
with the full version (Ondersma, Chaffin, and the level of conflict with persons outside the
Mullins, & LeBreton, 2005). family or community agencies.
Content The CAPI contains three validity Norms Normative information is available in
scales: Lie, Random Response, and Inconsistency. the CAPI manual (Milner, 1986) from a sample
The Lie scale was designed to detect tendencies of 836 parents, child care workers, and parent
to distort responses in a socially desirable man- aides from Florida, California, North Carolina,
ner. Both the Random Response and Hawaii, Oklahoma, Illinois, New York, and West
Inconsistency scales were designed to detect Germany. It is unclear how this normative sam-
haphazard or random responses to items without ple was selected, and the representativeness of
regard to item content. To test the usefulness of this sample in terms of parental education, socio-
these validity scales, Milner and Crouch (1997) economic status, and ethnicity is also unclear.
had two groups of parents, 106 community vol- This information is crucial because there is evi-
unteer parents and 80 parents attending a pro- dence that family functioning can vary as a func-
gram for parents at risk for abuse, complete the tion of these variables (Maccoby & Martin,
CAPI in several different ways: answering hon- 1983). As a result, lack of accessible information
estly, answering in a way to make themselves on the normative sample hinders the ability to
“look good,” answering in a way to make them- make norm-referenced interpretations from the
selves “look bad,” and answering inconsistently. CAPI scales.
These differing instructions did affect how par-
ents answered the CAPI questions, suggesting Reliability and Validity The CAPI manual pro-
that parents can intentionally distort their rat- vides evidence that the composite Abuse scale
ings. With the exception of detecting the faking- and the Distress and Rigidity scales exhibit
bad condition (58% correct identification), the acceptable internal consistency and temporal sta-
CAPI validity indexes were good at detecting bility. The reliability of the four other individual
most of the other response conditions, ranging scales tends to be more inconsistent across sam-
from 82% to 100% correct identification across ples. In terms of validity, there is evidence that
both samples of parents. CAPI scores are associated with documented
There are six primary scales of the CAPI that risk factors for child abuse (Budd, Heilman, &
are combined into a composite Abuse scale. The Kane, 2000; Grietens, De Haene, & Uyteeborek,
items were developed from an extensive review 2007; Haskett, Scott, & Fann, 1995). Also there
of research on child abuse and neglect. The is evidence that the composite Abuse scale can
Distress scale assesses parental anger, frustra- successfully discriminate between proven abus-
tion, impulse control, anxiety, and depression. ers and control subjects (Walker & Davies,
The Rigidity scale assesses parents’ flexibility 2010), and this differentiation extends across
and realism in their expectations of children’s cultural groups (Milner & Crouch, 2012).
behavior. It includes such items as “A child Finally, the Abuse scale has proven to be sensi-
should never disobey,” “A child should always be tive to the effects of intervention with high-risk
neat,” and “A child should never talk back.” The parents (Walker & Davies, 2010). Therefore, it
Unhappiness scale assesses a parent’s degree of appears that the CAPI provides a reliable method
personal fulfillment as an individual, as a parent, of assessing dysfunctional elements of a child’s
as a marital/sex partner, and as a friend. Problems family environment, including several aspects of
with Child and Self is a scale with items tapping parental stress that are associated with child
parents’ perceptions of their child’s behavior and abuse. In addition, the composite Abuse scale
their perceptions of their own self-concept as a does seem to be an index of risk for abuse,
Parenting Stress 239
although it is important to recognize that many the scale and now makes the norm-referenced
parents who score high on the CAPI have no scores more useful for interpretations in clinical
documented evidence of abuse in the home (i.e., assessments.
false positive rates; Haz & Ramirez, 1998).
Reliability and Validity The manual of the
PSI-4 provides convincing evidence for the
arenting Stress Index: Fourth
P internal consistency and temporal stability of the
Edition (PSI) three composite scores: Total Stress, Parent
Domain, and Child Domain. The reliability coef-
The PSI-4 (Abidin, 2014) is unique in its focus ficients for the individual scales, however, are
specifically on stressors related to parenting. It much more variable and typically exhibit rela-
was primarily designed to assess the family con- tively low reliability estimates. The manual pro-
text of preschool children between the ages of 1 vides one of the best summaries of the extensive
and 4, although it has been used in older samples use of the PSI-4 and its predecessors in research
of preadolescent children. Completion of the PSI on the family context of preschoolers (see also
requires at least a fifth-grade education. It Abidin, Flens, & Austin, 1995 and Johnson,
contains 120 items and generally takes
2015 for updated reviews). In general, the vari-
20–30 minutes to complete. A short form of the ous versions of the PSI have been correlated with
PSI-4 has been developed with 36 items and has other measures of family functioning
proven to be highly correlated with the full ver- (Pinderhughes, Dodge, Bates, Petit, & Zelli,
sion (Abidin, 2012). 2000), including correlations with observations
of parenting behavior (Bigras, LeFreniere, &
Content The items of the full PSI are divided Dumas, 1996) and maternal depression (Barroso,
into two main categories: Child Domain and Hungerford, Garcia, Graziano, & Bagner, 2016).
Parent Domain. The Child Domain consists of Also, the PSI has differentiated families who are
items that assess qualities of a child that make it experiencing major stressors from nonstressed
difficult for parents to fulfill their parental role. families (Holden & Banez, 1996; Whiteside-
The Parent Domain assesses sources of stress and Mansell et al., 2007) and has proven sensitive to
disability related to parental functioning. treatment effects (Nixon, Sweeney, Erickson, &
Table 12.4 provides a summary of the scales that Touyz, 2003).
constitute the Child and Parent domains. The Also, factor analyses generally support the
PSI-4 also allows for the computation of a com- broad Parent and Child Domains for grouping
posite score that provides an overall indicator of the various PSI subscales, although some studies
the amount of stress in the parent-child system. have provided support for a third Parent-Child
Also, the PSI-4 has an optional Life Stress scale Interaction Domain, which includes the sub-
that includes 19 items rated as “yes” or “no” for scales of Child Acceptability, Child Reinforces
the parent to rate if she or he has experienced any Parent, and Parent Attachment to Child
of a list of life stressors, such as divorce, death of (Hutcheson & Black, 1996; Solis & Abidin,
parent, start of new job, etc. 1991). Further, Bigras et al. (1996) reported that,
in a sample of 218 mothers of preschoolers, the
Norms The normative sample for the PSI-4 con- Parent and Child Domains predicted different
sisted of 534 mothers and 522 fathers of children parental, familial, and child outcomes obtained
ages 1–12 years of age. The sample was recruited from other sources. Specifically, the Parent
from various regions of the United States and was Domain was more strongly and independently
recruited to include a diversity of families accord- associated with measures of marital adjustment
ing to race, ethnicity, and educational level. This and maternal depression, whereas the Child
normative sample represents a major improve- Domain was more strongly and independently
ment from the one used in previous editions of associated with child difficulties reported by the
240 12 Assessing Family Context
mother and children problems observed during well-being (e.g., conduct problems, school
parent-child interactions. These results are achievement, social adjustment, self-concept).
important in suggesting that the two domains are These studies suggested that the relationship
valid in assessing somewhat independent dimen- between divorce and psychological difficulties
sions of family functioning. in children was greatest within 2 years immedi-
ately following a divorce, although some effects
can be more long lasting (Sands, Thompson, &
Marital Conflict Gaysina, 2017).
The meta-analysis also provided intriguing
There is a long history of research showing a data in support of the theory that it is the conflict
link between divorce and child behavior prob- that occurs between parents before and during
lems. Amato and Keith (1991) conducted a the separation that has the most detrimental
meta-analysis of 92 published studies of the impact on a child’s adjustment (see also Jamison
impact of divorce on a child’s psychological & Freeman, 2015). Whereas children of divorced
well-being. The combined samples from the 92 families tended to have poorer adjustment than
studies involved over 13,000 children. This children in low-conflict, intact families, chil-
meta-analysis revealed that divorce consistently dren in intact, high-conflict homes tended to
had a negative impact on several types of child have the poorest adjustment of all three groups.
Parental Adjustment 241
Also consistent with this perspective, several ask for overall estimates of the amount of verbal
studies found that less conflict and better divorce and physical hostility between spouses that is
cooperation between parents predicted better witnessed by the child.
postdivorce adjustment for children. There is little normative data on the
The implications of these findings are impor- OPS. However, 2-week test-retest reliability in
tant to clinical assessments of children and ado- a sample of 14 families was found to be quite
lescents. They suggest that it is not simply high (r = 0.92; Porter & O’Leary, 1980). These
enough to determine the marital status of a authors also reported that the OPS was corre-
child’s parents for understanding the potential lated with several types of maladjustment in
impact of the parents’ marital relationship on the children. Highlighting the importance of focus-
child. A more important focus of assessment is ing specifically on overt conflict, these authors
the overt conflict between parents that is wit- found that the OPS was more consistently asso-
nessed by the child. ciated with child adjustment difficulties than
There are several marital inventories that are was a measure of general marital satisfaction
frequently used in research and clinical practice (i.e., the MAT). This association between scores
and often included in the clinical assessment of on the OPS has been replicated in other studies
children and adolescents (McMahon & Frick, (Davies, Martin, & Cummings, 2018; Forehand,
2007). The Marital Adjustment Test (MAT; Long, & Hedrick, 1987; Mann & MacKenzie,
Locke & Wallace, 1959) and the Dyadic 1996). Thus, it appears that the OPS captures
Adjustment Scale (DAS; Spanier, 1976) are two the critical component of marital discord in
self-report instruments that have been shown to terms of its detrimental effect on child adjust-
produce reliable scores and differentiate persons ment. In addition, the OPS has proven to be
in distressed and nondistressed marriages. sensitive to treatment, showing changes follow-
However, these inventories tend to focus on gen- ing a preventive intervention to improve parent-
eral marital satisfaction rather than on overt con- ing in families with a history of parental
flict per se. The O’Leary-Porter Scale (OPS; depression (Sullivan, Parent, Forehand, &
Porter & O’Leary, 1980) is a brief rating scale Compas, 2018).
that focuses on overt marital conflict and, even
more specifically, on marital conflict that is wit-
nessed by the child or adolescent. As a result, the Parental Adjustment
OPS is uniquely suited for use in clinical assess-
ments of children and adolescents. In the introduction to this chapter, we discussed
several ways in which assessing parental psychi-
atric adjustment is critical to the clinical assess-
O’Leary-Porter Scale (OPS) ment of children. We discussed two areas of
research, one on parental depression and another
The OPS (Porter & O’Leary, 1980) was devel- on parental anxiety, that suggest that information
oped specifically for studying the association obtained from a parent must be interpreted in
between marital adjustment and child behavior light of the parent’s level of emotional distress.
problems. The OPS is a 20-item self-report We also discussed the importance of obtaining a
inventory within which are embedded nine family psychiatric history for making differential
items that assess the degree of marital conflict diagnoses and treatment recommendations. In
witnessed by the child. A parent rates on a this section, we provide a brief overview of basic
5-point frequency scale (Never to Very Often) research showing the link between parent and
how often the child witnesses arguments child adjustment difficulties that can aid the
between himself or herself and the spouse over clinical assessor in structuring assessments
money, discipline, wife’s role in family, and per- and making appropriate interpretations from the
sonal habits of the spouse. Two questions also assessment information.
242 12 Assessing Family Context
One type of parental adjustment that has a well- A comprehensive review of the literature found
documented link to child development is paren- that, like parental depression, parental alcohol-
tal depression and, in particular, maternal ism is associated with a number of child adjust-
depression. A meta-analysis of 193 studies ment problems. West and Prinz (1987) reported
involving 80,851 mother-child dyads indicated studies finding an association between parental
that maternal depression was associated with alcohol abuse and the following problems in their
internalizing, externalizing, and general psycho- children: hyperactivity, conduct problems, delin-
pathology, as well as negative affect/behavior quency, substance abuse, intellectual impairment,
and lower levels of positive affect/behavior. somatic problems, anxiety, depression, and social
Weighted correlations ranged from 0.23 between deficits. Like depression, some of the lack of
maternal depression and internalizing problems specificity in its effect on child adjustment may
in the child to −0.10 between maternal depres- be due to a failure to define subgroups within par-
sion and positive affect in the child (Goodman ents who abuse substances (Frick, 1993).
et al., 2011). Therefore, it seems that parental Alternatively, West and Prinz reviewed several
depression is a non-specific risk factor for prob- studies suggesting that the effects of having a
lems in children. That is, it is not specifically substance-abusing parent on a child’s adjustment
related to the development of a single type of may be mediated through the impact on the home
child behavior problem. environment and the impact on parent and child
There are two possible exceptions to this interactions. Consistent with this view, parental
non-specific relationship, both related to sub- substance use has been linked to a host of prob-
types within affective disorders. First, Weissman lematic parent practices, including higher rates of
et al. (1988) found some preliminary evidence abuse (Ondersma, 2016; Walsh, MacMillan, &
that early-onset recurrent depression (depres- Jamieson, 2003).
sion that has its initial onset before adulthood)
might have a more specific link with childhood
depression. Second, family histories of bipolar Parental Antisocial Behavior
disorders in parents predict a risk for subsequent
bipolar disorder in adolescents with a childhood- The intergenerational link to antisocial behav-
onset of depressive symptoms (Geller et al., ior is a consistent finding in research and one
1993). that has long intrigued social scientists and
Goodman and Gotlib (1999) reviewed the lit- policy-makers alike (see Frick & Loney, 2002,
erature on the risk for problems in adjustment in for a review). Early studies tended to focus on
children of depressed mothers. They outline four the intergenerational link to criminality. This
possible mechanisms to explain this risk: (a) an research found that the link was independent of
inherited predisposition transmitted from parent socioeconomic status, neighborhood, and intel-
to child; (b) failure of the child to develop appro- ligence (Glueck & Glueck, 1968). More recent
priate emotional regulation strategies; (c) expo- studies have focused on psychiatric definitions
sure to negative maternal moods, thoughts, and of antisocial disorders. As in studies of crimi-
behaviors; and (d) exposure to a high rate of nality, children diagnosed with antisocial disor-
stressors associated with mothers’ depression ders are significantly more likely to have
(e.g., higher rates of marital conflict). These parents with antisocial disorders than are chil-
links clearly illustrate the need to assess the dren without conduct problems (Frick et al.,
parental depression to understand several 1992; Monuteaux, Faraone, Gross, &
potential causal factors that could help to Biederman, 2007).
explain a child’s or an adolescent’s problems in An important methodological point in the
adjustment. more recent family history studies was the fact
Parental Adjustment 243
that each used clinic control groups and found problems associated with ADHD before the age
that histories of antisocial disorders in parents of 18 and then completed a similar family history
were specific to conduct problems in children. questionnaire for all first-degree relatives.
That is, children with conduct problems not only Children with ADHD were more likely to have
had higher rates of parental antisocial disorder mothers, fathers, and other biological relatives
(APD) than normal controls, but they also had who also exhibited ADHD as children than were
higher rates of parental APD than clinic-referred other clinic-referred children. In fact, approxi-
children with other problems in adjustment (Frick mately 75% of the 103 children with ADHD had
et al., 1992). Therefore, unlike parental depres- one biological relative with a significant history
sion and substance abuse, parent antisocial of ADHD (27% of mothers and 44% of fathers),
behavior appears to have a more specific relation- and 46% had two biological relatives with a sig-
ship to a particular child problem (i.e., conduct nificant history of ADHD.
problems). As with many of the other forms of parental
Frick and Loney (2002) reviewed data sup- psychopathology, one way that parental ADHD
porting several potential mechanisms to explain may have a negative influence on a child’s adjust-
this link including an inherited disposition ment is through its effects on a parent’s ability to
passed from parent to child, parental modeling parent effectively. For example, Park, Hudec, and
of antisocial behaviors, and disruptions in the Johnson (2017) conducted a meta-analysis of the
family caused by the parent’s antisocial behav- association between parental ADHD and parent-
ior. In support of at least some inherited predis- ing behaviors. They reported on 29 studies
position, Tapscott et al. (1996) showed that a involving 4981 participants with children ranging
paternal history of antisocial personality disor- in age from 7 months to 15 years old. They
der was associated with a higher rate of Conduct reported that parental ADHD symptoms were
Disorder in their biological offspring, even if the associated with harsh parenting (weighted over-
father had no contact with the child since the first all r = 0.17, p < 0.001) and lax parenting
year of life. (weighted overall r = 0.18, p < 0.001).
Importantly, these effects were consistent for
mother and fathers and even controlling for other
Parental ADHD forms of parental psychopathology, such as
depression and antisocial behavior.
There is evidence that parents and other biologi- Again, all of this research suggests that an
cal relatives of children with ADHD show more assessment of parents’ current and childhood his-
attentional problems and a higher rate of ADHD tories of ADHD behaviors could aid in assess-
(Musser et al., 2014; Rowland et al., 2018). ment of ADHD in children.
However, these studies may have underestimated
the link between parent and child ADHD by
studying the parents’ current adjustment. Given Parental Anxiety
that 30–50% of children with ADHD may not be
diagnosed with this disorder as adults (Barkley, Another type of problem that appears to have a
Fischer, Smallish, & Fletcher, 2002), it may be familial link is anxiety. Last et al. (1987) reported
that many of the parents of children with ADHD that in their sample of children with an anxiety
exhibited ADHD as a child but are not currently disorder (n = 58), 83% of the children had a
showing symptoms. mother with a lifetime history of anxiety disor-
To test this possibility, Frick et al. (1991) ders. Furthermore, 57% had a mother experienc-
studied the childhood histories of parents of
ing significant levels of anxiety concurrently with
clinic-referred children. A child’s biological par- the child. Both of these proportions were signifi-
ent reported on whether or not he or she had cantly greater than what was found in parents of
244 12 Assessing Family Context
such as when one wants to focus on some spe- (a) Familial influences often play major
cific domain of parental adjustment. For exam- causal roles in a child’s or adolescent’s
ple, the Beck Depression Inventory (BDI; Beck, psychological difficulties.
Steer, & Garbin, 1988) and the State-Trait (b) A family psychiatric history can be
Anxiety Inventory (Spielberger, Gorsuch, & instrumental in making differential
Lushene, 1970) are brief screening measures for diagnoses.
depression and anxiety, respectively, that are (c) Understanding a child’s or adolescent’s
often used to assess parental adjustment in clinic- family context can help to interpret
referred children. These are just a few of a host information provided by members of the
of domain-specific rating scales that can be used family.
to assess a specific area of adjustment in a child’s (d) Understanding a child’s family context
parent or other relatives (see McMahon & Frick, can help to determine the most important
2007 for others). targets for intervention.
2. Many behavior rating scales reviewed in pre-
vious chapters, such as the BASC-3, have
Conclusions scales that assess various aspects of a child’s
family context.
In this chapter we discussed several reasons why 3. Research suggests that two critical dimen-
the assessment of the family environment is criti- sions of family functioning related to child
cal to most clinical assessments of children and adjustment are parenting style, which is the
adolescents. Family factors often play a critical emotional climate provided by the parents,
causal role in child maladjustment, and familial and parenting practices, which are tech-
factors can aid in making differential diagnoses niques used by parents to socialize their
and treatment recommendations, two important children and enforce rules.
goals of many clinical assessments. Further, fac- 4. The Family Environment Scale-Second
tors within the family are often important for Edition (FES) is a commonly used measure
interpreting information provided by members of of parenting style and the emotional
the family on a child’s or adolescent’s adjust- climate of the family. The FES is divided
ment. There are a large number of dimensions of into subscales from three domains:
family functioning that can influence child Relationships, Personal Growth, and
adjustment. We focused on four dimensions that System Maintenance.
we think are especially important to assess in 5. The Alabama Parenting Questionnaire
most clinical assessments of children and adoles- (APQ) and the Parenting Scale (PS) are mea-
cents: parenting styles and practices, parenting sures that focus more specifically on parent-
stress, marital conflict, and parental adjustment. ing behaviors, such as parents’ discipline
For each of these dimensions, we reviewed the strategies.
research linking them to child adjustment and 6. The Dyadic Parent-Child Interaction Coding
then provided several methods for assessing them System (DPICS) is an observational coding
in clinical assessments. system designed to assess parent and child
behaviors in two standardized parent-child
interaction tasks.
Chapter Summary 7. The amount of stress experienced by parents
is also important to child adjustment. Scales
1. There is no context more important to under- can assess general stressors or stressors spe-
standing a child’s emotional and behavioral cific to parenting.
functioning than understanding the child’s 8. The Child Abuse Potential Inventory-
family context because: Second Edition (CAPI) was developed
246 12 Assessing Family Context
include age of onset, course/prognosis, etiology, clinician may investigate more carefully the diag-
previous treatment history, and other relevant nosis of ADHD with its associated early age of
background information. onset. (See also the case study of Jeff, Box 13.1).
after the clinical interview indicates that the ogy to assess for comorbidity, as well as issues
behavior problems have started to show up at that could have actually precipitated current dif-
school, as well as at home. ficulties (e.g., a history of depression that now
3. A child is initially not given a diagnosis of manifests partly as difficulty concentrating). In
ADHD when the rating scales show no prob- addition, attempting to determine etiology for a
lems at home and only problems at school. specific client helps clarify both case conceptual-
However, the clinical interview indicated that izations to answer the referral question and the
the child had problems at home, but they were recommended interventions targeting the pre-
reduced dramatically when his parents started sumed etiologies. After all, these activities are the
very structured routines around homework primary purpose of child assessment, and history
and chores. Further, the child was asked to taking is valuable to these ends.
leave two sport teams (i.e., soccer and base- Unfortunately, for most childhood problems, the
ball) because of misbehavior due to hyperac- range of potential single and multiple etiologies is
tivity. Thus, a diagnosis of ADHD was given, often extensive. Moreover, multiple etiologies may
after the clinical interview revealed that the be interacting to produce symptoms of a problem
impairment was not confined to school. such as depression. It would not be unusual, for
example, for a child with depressive symptoms to
be affected by parental depression, poverty, and the
Etiology death of a friend—all of which may require simul-
taneous and/or coordinated intervention.
Etiology refers to the likely presumed cause of a Commonly used history-taking forms may not
child’s difficulties. The assessment of etiology is fully address the various etiologies associated
crucial in that it has important implications for with a problem or set of problems. Clinicians
treatment. The discovery that a child’s learning often have to use their knowledge of child devel-
difficulties in mathematics did not begin until the opment, psychopathology, and other areas to go
ninth grade may very well rule out a developmen- beyond the standard questions included on a his-
tal disorder such as a learning disability, depend- tory form (i.e., branching) to rule out important
ing on other aspects of the client’s history. If the high-frequency etiologies.
clinician then discovers that the child was for the This process of branching from a history form
first time placed in the advanced section of a is difficult for even the savviest clinician. For the
mathematics class, then curriculum placement majority of examiners, the most realistic option
becomes a potential etiological agent. A simple will be to assess history over the course of two or
change in classes may be an effective interven- more assessment sessions. A second session
tion for this suspected mathematics learning dis- could be as simple as a telephone call that allows
ability. Such a parsimonious intervention may the clinician to rule out an etiology that has been
not have been tried, however, without important hypothesized based on previous history, assess-
historical information that led the clinician to a ment data, or other information. Needless to say,
potential etiology for the problems in math. the etiology of many childhood disorders is com-
Research on problems as varied as intellectual plex. Moreover, determining the nature of the
disabilities (Toth & King, 2010) and suicidality presenting problem, its potential etiology, and
(Edlavitch & Byrns, 2014) points to the impor- plans for treatment is probabilistic rather than
tance of understanding the etiology of a problem definitive (Mash & Barkley, 2014). Thus, clini-
for a client rather than merely documenting test cians may be well-served by conducting assess-
scores or symptoms. Importantly, etiological fac- ments, including follow-ups from clinical
tors may then provide a road map for treatment interviews, over a couple of sessions to enhance
options. Information about etiology is perhaps the accuracy of their case conceptualizations.
best gained from a sound approach to taking a However, it is acknowledged that many practitio-
client’s developmental history. ners are working in settings or under circum-
It is apparent that part of this assessment stances in which assessments are expected to be
should include a history of other symptomatol- completed very quickly. Therefore, it is important
Family Psychiatric History 251
outset of the assessment. The influence of context Most history-taking forms, however, require
can also be protective, and thus, these variables considerable English language fluency on the
may also serve as strengths around which inter- part of parents. Technical topics, such as the type
ventions can be built. of special education program that their child
attends or a previous medical condition, are dif-
ficult for even the most sophisticated parents.
Content Regardless, parents may be able to complete part
of the history on forms and thereby save the
There is considerable overlapping content among examiner interviewing time. Assessment systems
history measures although varying degrees of such as the BASC (Reynolds & Kamphaus, 2015)
emphasis exist. Some of the most common broad discussed in previous chapters on rating scales
areas of inquiry include (McConaughy, 2013): also include forms for history taking (e.g.,
BASC-3 Structured Developmental History;
Complaints/symptoms SDH). An example of a history form is shown as
Developmental/medical history Table 13.1. Of course, history forms will vary
Family history greatly in their depth and breadth. The important
Social functioning issue is to cover essential history information as
Academic functioning part of the assessment and to ideally do so in an
Family relations efficient manner. Forms such as the one shown in
Interests/strengths Table 13.1 allow the clinician to gain such infor-
View of the problems mation but also require the clinician to follow up
based on the parent’s responses.
Morrison and Flegel (2016) offer additional Many clinicians, however, prefer an interview
suggestions for specific areas to assess (e.g., pre- format because it allows them to use a branching
cipitating stressors, developmental milestones, procedure to pursue topics that are of particular rel-
family cultural background, risk of self-harm, evance to issues of onset, course, impairment, eti-
abuse history, mood, insight, etc.) within these ology, and previous treatment history among other
domains. Of particular importance is that a clear factors without a potentially cumbersome form
description of the concerns/symptoms (rather that includes some irrelevant information for a par-
than labels such as “hyperactive” or “depressed”) ticular child. A sample scenario is a case where a
is preferred, as the manifestation of such prob- 9-year-old child is referred due to decreases in aca-
lems likely will be idiosyncratic to the child. In demic achievement. The examiner may discover
addition, a history that is of interest to a certain that the child’s performance in school was exem-
specialty (e.g., neuropsychology) may have a dif- plary until the current academic year, when the
fering emphasis. child missed 35 days of school due to an automo-
bile accident. The examiner, in this case, will likely
branch to ask questions about the child’s length of
Format absence from school and the extent of documented
head injury. New and detailed information about
History taking is typically conducted via two for- these problems may profoundly influence interven-
mats—the interview and a written form. While tion planning. Special education may be placed in
history-taking research has focused on the inter- abeyance in favor of remedial coursework and
view, there are often occasions where a written monitoring to see if the child recovers his premor-
form may be used, particularly with the parents bid level of functioning with little intervention, as
of children. Parents are already accustomed to would often be predicted.
completing such forms while waiting in physi- Hence, the interview or partial interview for-
cians’ offices. Increasingly, parents are also mat allows the clinician the flexibility necessary
accustomed to completing rating scales in psy- for thorough exploration of variables that may
chologists’ offices, as well. impact intervention rather than have the parent
256 13 Clinical Interviews
Psychology Clinic
Child and Adolescent History Form
Mailing address:
School: Grade:
Teacher’s name:
Person filling out this form (circle one) Mother Father Stepmother Stepfather Grandparent
Other:
Relationship to child
Parents’ occupation(s)
Table 13.1 (continued)
Format 257
My child:
Engages in behavior that could be dangerous to self or others Yes No
Describe:
Has unusual habits (e.g., thumb sucking, pulling hair, etc.) Yes No
Describe:
Is likable Yes No
Academic Information
Table 13.1 (continued)
258 13 Clinical Interviews
Table 13.1 (continued)
During pregnancy did the child’s mother drink alcoholic beverages? Yes No
During pregnancy, did the child’s mother use drugs? Yes No
If yes, what type?
Has your child had any hospitalizations for medical reasons? Yes No
If yes, describe and give the ages at which they occurred:
Format 259
Family History
Alcoholism Yes No
If yes, relationship to child
Schizophrenia Yes No
Relationship to child
Depression Yes No
Relationship to child
Anxiety Yes No
Relationship to child
Table 13.1 (continued)
260 13 Clinical Interviews
Table 13.1 (continued)
Psychological History
Has your child been on medication for emotional or behavioral problems? Yes No
If yes, when and what kind?
Has your child ever been arrested or had contact with the police? Yes No
If yes, please describe:
Has your child ever been hospitalized for emotional or behavioral problems? Yes No
If yes, please describe:
Other Information
Please provide any other information that may be useful to us in getting to know and in helping your child:
Conclusions 261
Box 13.4: Observing Nonverbal Cues During Adapted from Green (1992), p. 455
History Taking
early identification is so severe that it has been children’s social, behavioral, and emotional
described as a public health crisis (Campaign for problems from worsening to the point of having
Mental Health Reform, 2005). an ED classification. For example, students clas-
The personal and societal costs of this failure sified as ED experience comparatively poor edu-
to identify, treat, or intervene to alleviate cational outcomes including underachievement
problems early are substantial and, in some cases, in both reading and mathematics (Bradley,
tragic. In a 2013 children’s mental health surveil-Doolittle, & Bartolotta, 2008), higher rates of
lance report by the US Department of Health and suspensions and expulsions (Wagner, Kutash,
Human Services, Centers for Disease Control Duchnowski, & Epstein, 2005), absenteeism
and Prevention (CDC) concluded that: (Lane, Carter, Pierson, & Glaeser, 2006), school
dropout, and comparatively poor participation
• One in five children suffers from a mental rates in postsecondary educational opportunities
health disorder in any given year. (Wagner et al., 2005). Researchers have sug-
• Suicide was the second leading cause of death gested that schools and communities are gener-
of children between 12 and 17 years of age in ally ill-prepared for effectively and systematically
2010. identifying children with these emotional and
• The prevalence rate of mental health disorders behavioral problems (Roeser, 2001). While
increases with the age of the child. screening for vision, hearing, and academic
achievement is common, screening for behav-
The cost to American society of these disor- ioral and emotional problems rarely occurs—
ders is estimated at $247 billion per year (CDC, fewer than 2% of school districts in the US screen
2013). Research studies from the past few for such problems (Jamieson & Romer, 2005).
decades have documented a strong link between This is in spite of research that suggests early
a child’s behavioral and emotional problems and identification of these problems is particularly
his or her academic performance (Gutman, important for maintaining an optimal school
Sameroff, & Cole, 2003; Jimerson, Egeland, & environment.
Teo, 1999; McEvoy & Welker, 2000). Anxiety Early identification of behavioral and emo-
and depression have been shown to impede aca- tional problems also has benefits beyond the
demic achievement even after accounting for the school by improving the community through a
effects of general intelligence (Rapport, Denney, reduction in overall healthcare burden and costs
Chung, & Hustace, 2001). Also, some of these (Aos, Lieb, Mayfield, Miller, & Pennucci, 2004).
studies have underscored the fact that such prob- While widely available, comprehensive behavior
lems often begin in early childhood. In one study rating scales are not well suited to large-scale
that included classrooms from the federally testing due to the time and effort that would be
funded Head Start program, children who were required to assess hundreds or thousands of chil-
later diagnosed or classified with behavioral or dren within a given school district (Flanagan,
emotional problems were generally under- Bierman, & Kam, 2003). Similarly, multi-gate
identified for these problems by staff members. systems that include combinations of teacher
These children had the greatest risk for exhibiting rankings, ratings of students, and classroom
poor academic readiness (Fantuzzo, Bulotsky, observations have not had widespread adoption
McDermott, Mosca, & Lutz, 2003). According to by schools due to the amount of teacher time
the National Center for Education Statistics, such tasks demand. The lack of practical screen-
about 5% of the student population is served ing tools with associated evidence of reliability
under the special education classification of emo- and validity has been a significant barrier to the
tional disturbance (ED) (U.S. Department of establishment of behavioral and emotional
Education, Institute of Education Sciences, screening programs in schools, hospitals, and
National Center for Education Statistics, 2019). pediatric practices (DiStefano & Kamphaus,
Additional research suggests the need to prevent 2007; Kamphaus et al., 2000).
Why Is Screening Important? 265
On the other hand, the practice of childhood symptoms suffered from significant impairments
screening for behavioral and emotional problems both inside and outside of school (Bertha &
is long-standing, dating to an effort to screen Balázs, 2013). The results were similar for an
children launched by the California Department analogous review of 18 studies of children with
of Education (Cowen et al., 1973). Nevertheless, subsyndromal problems associated with ADHD
the practice never took hold. A lack of training of (Balázs & Keresztény, 2014).
educators and other school staff is likely one of While the size of the population of students
the root causes of the absence in uptake of the with BER is unknown, various researchers have
practice. A report by the National Academy of estimated that about 15% of students might pos-
Sciences, Institute of Medicine Panel on mental sess subsyndromal problems in the social, behav-
health disorder prevention concluded (O’Connell, ioral, and emotional domain (Schanding Jr. &
Boat, & Warner, 2009): Nowell, 2013; Storment & Reinke, 2013). Harris
Neither the core curriculum for a bachelor’s degree and colleagues (2012) used the term behavior
nor the process for obtaining a teaching certificate challenges to refer to children with subsyndro-
anticipate that teachers will be prepared to recog- mal problems in their school-based work. For the
nize risk factors or detect early evidence of MEB purposes of their research, this group was opera-
[mental, emotional, or behavioral] disorders in
their pupils. Coursework for education degree stu- tionally defined as students who “demonstrated
dents includes descriptions of mental disorders initial or soft signs of antisocial behavior” as
(along with physical disorders and retardation), but indicated by teacher ratings of behaviors, such as
it does not systematically include how to identify, stealing, lying, cheating, sneaking, behavior
intervene, or refer children at risk for MEB disor-
ders. (p. 366–367) problems, peer rejection, negative attitude, and
aggressive behavior. Kamphaus (2012) proposed
In addition, there is a large group of children and an alternative label for describing youth with
adolescents who do not possess either a mental subsyndromal social, behavioral, and emotional
health disorder or ED, but nevertheless have mild problems: behavioral and emotional risk (abbre-
social, behavioral, and emotional problems that viated as BER).
put them at risk for developing a mental health Unfortunately, research has clearly docu-
disorder or ED at a later point in time. Individuals mented that most youth with BER go unrecog-
with mild problems, or “symptoms” as they are nized and miss secondary prevention
called in the mental health literature, have been opportunities at their schools. Forness et al.
referred to as subsyndromal (Cantwell, 1996). (1998), for example, conducted a longitudinal
They, for example, might possess early symp- study of 3323 students first assessed at age 4 or
toms of a mental health disorder, but not enough 5 in kindergarten. The researchers created two
symptoms to be classified as having depression, rigorous research-based definitions and classifi-
Attention-Deficit Hyperactivity Disorder cation criteria for special education classification
(ADHD), etc. (Okasha, 2009). Borrowing health- as ED. They found that, by the end of second
care metaphors, these might be children who are grade, 17.2% of the students met one set of crite-
“prehypertensive” or “prediabetic” but for behav- ria for classification, and 6.7% met the other defi-
ioral or emotional disorders. nition. By contrast, the schools only identified
Their risk for later special education place- 0.9% of the students as eligible for special educa-
ment (Koot & Verhulst, 1992) or developing a tion under the category of ED. Even worse, if stu-
mental health disorder (Hinshaw et al., 2012) is dents were classified as special education eligible,
well-documented. Research has also shown that they were identified as belonging in a category
youth with these subsyndromal problems often other than ED. This finding has been replicated in
have similar problems in daily living as those other samples (Scott, Anderson, Mancil, & Alter,
who meet diagnostic criteria. In the case of ado- 2009).
lescent depression, for example, a review of 27 In addition to under-identification, the stan-
studies found that children with subsyndromal dard of practice of relying on teacher and parent
266 14 Behavioral and Emotional Risk Screening
referral results in disproportionate identification could be drawn from popular omnibus rating
of children of color. Research has produced scales, such as the Behavior Assessment System
strong evidence that harsh disciplinary practices, of Children-Third Edition (BASC-3; Reynolds &
such as school suspensions and expulsions, can Kamphaus, 2015) and Achenbach System of
lead to disproportionate referral of children of Empirically Based Assessment (ASEBA;
color, particularly boys, for behavior problems Achenbach & Rescorla, 2001) to obtain wide
(Gregory, Skiba, & Noguera, 2010). Even milder content coverage. On the other hand, suicidal ide-
forms of discipline, office discipline referrals ation, for example, would not be a good choice
(ODRs), are clearly characterized by unequal for a screening measure because this symptom is
referral of African-American boys (Bradshaw, a key symptom of a problem that has progressed
Mitchell, O’Brennan, & Leaf, 2010). By provid- to the stage of more serious mental health prob-
ing a more accurate and standardized identifica- lems, such as a depressive disorder. Items assess-
tion of youth with BER than the “standard of ing unhappiness would be a better choice if one is
practice” of teacher or parent referral (Dowdy, interested in assessing BER.
Doane, Eklund, & Dever, 2013; Eklund et al., Prior research indicates that there are three
2009; Hill et al., 2004), universal screening has primary characteristics of the subpopulation of
the potential to decrease the disproportionate students with BER that need to be considered by
placement of boys and students of color in spe- the measures used to identify them. First, risk is
cial education (Dowdy, Doane, Eklund, & Dever, dimensional in that it reflects deviant standing on
2013; Raines, Dever, Kamphaus, & Roach, important dimensions of child social, behavioral,
2012). and emotional adjustment (Okasha, 2009). The
presence of BER is typically established based
on formal US national norm-referenced assess-
efining Behavioral and Emotional
D ment results for an individual child, where risk
Risk status is defined as a T-score at or above 60 or
about the 84th percentile rank if scores from the
For the purposes of screening, we have adopted scale are approximately normally distributed
the National Academies definition of BER, which (Kamphaus & Reynolds, 2007). Second, the con-
defines the construct as early symptoms of disor- tent of BER measures includes items/problems/
ders that may eventually result in special educa- symptoms representing empirically supported
tion placement or mental health treatment dimensions of child adjustment: (a) internalizing
(O’Connell et al., 2009). The National Academies problems (e.g., sadness, worry, headaches, lone-
report gives the BER definition as the following: liness), (b) externalizing problems (e.g., high
For prevention, one of the goals of screening activity level, inattention, aggression), and (c)
should be to identify communities, groups, or indi- adaptive skills (e.g., prosocial skills, positive atti-
viduals exposed to risks or experiencing early tude toward school, regulates emotions)
symptoms that increase the potential that they will (Kamphaus, 2012). Adaptive skills are important
have negative emotional or behavioral outcomes
and take action prior to there being a diagnosable to include because a student’s standing on these
disorder. (p. 223) has been shown to differentiate risk groups
(Eriksson et al., 2013). Third, deviant standing on
Adoption of this definition then provides a con- these three dimensions is associated with a
tent blueprint for creation of BER measures in greater likelihood of developing mental health
that they should include early marker symptoms disorders and/or referral and placement in special
of a mental health disorder and cover those pre- education at a later point in development (Balázs
liminary symptoms associated with a variety of & Keresztény, 2014 ; Bertha & Balázs, 2013;
child disorders so that they have broad predictive Koot & Verhulst, 1992).
validity for poor schooling or mental health out- Several studies have demonstrated the impact
comes. Thus, the content of screening measures of these types of BER. Specifically:
Screening Measures 267
• Of students identified by student self-report some of the more commonly used screening
screening, about a third would have not been measures is provided below.
identified by the standard of practice using Choosing which screening instrument to uti-
teacher and parent referral methods that is cur- lize is a decision that will impact subsequent deci-
rently in common usage (Scott et al., 2009). sions. When evaluating a screening instrument for
• The implementation of screening alone leads potential adoption and use, the measure should be
to more referrals for prevention services evaluated on the basis of its norming sample and
(Husky, Sheridan, McGuire, & Offson, 2011) derived scores, reliability, and validity evidence to
and follow-up on referrals (Husky, Kanter, support score inferences, as discussed in Chap. 2
McGuire, & Offson, 2012). for measures used in clinical assessments more
• Screening may, in fact, improve delivery of generally. For the specific purpose of a screening
evidence-based preventive interventions measure, however, the most important evidence
(Husky et al., 2011). for validity should be the ability of scores to detect
students who are at risk for later mental health
In an investigation by Husky and colleagues problems (i.e., its predictive validity).
(2011), students were randomized to two condi- Below we summarize some of the more com-
tions, referral via universal screening (n = 365) monly used measures designed specifically for
and referral via the standard of practice mental health screening. As with other chapters in
(n = 291). Screened students were significantly the text, we cannot provide an exhaustive evalua-
more likely than control students to be referred tion of all of the various measures that have been
for mental health services. In addition, students developed for this purpose. Instead, we limit the
and families were willing to access needed ser- discussion to a few exemplars of screening mea-
vices by acting on the referrals made as indi- sures and to a summary of a few of the strengths
cated by their 95.5% participation in and weaknesses of each screener, to illustrate the
school-based services subsequent to referral. considerations that are important when choosing
However, Lubman and colleagues (2008) sug- one for use. However, we refer the interested
gested that for screening training to be effective, reader to Glover and Albers (2007), Severson,
it has to be cast as “a core service issue” and not Walker, Hope-Doolittle, Kratochwill, and
in its typical role as ancillary to intervention. If Gresham (2007), and Feeney-Kettler, Kratochwill,
it is a core issue, then preservice and in-service Kaiser, Henneter, and Kettler (2010), for more
training for teachers and other school personnel extended reviews of screening measures.
must include it as such.
he Behavior and Emotional
T
Screening Measures Screening System (BESS; Reynolds &
Kamphaus , 2015)
Screening measures have proliferated in the last
decade or so, offering school districts many more The BESS consists of brief screening measures
options for universal screening. Some studies in (30 items or less) that can be completed by teach-
the past have used omnibus rating scales, such as ers, parents, and/or students to identify BER in
the Child Behavior Checklist (Achenbach & youth ranging from preschool through high
Rescorla, 2001), as screening measures. However, school (Kamphaus & Reynolds, 2007). The
new measures that were specifically designed for BESS manual includes considerable evidence to
universal screening programs in schools have a support score inferences including high reliabili-
fraction of the items of omnibus behavior rating ties with median internal consistency coefficients
scales, typically 30 or fewer items, and forms that for the teacher, parent, and student forms of .96,
can be completed by parents, teachers, or chil- .94, and .92, respectively. Several factor analytic
dren themselves in 5–10 minutes. A review of studies noted that the BESS measures the major
268 14 Behavioral and Emotional Risk Screening
dimensions of child BER (Dever, Mays, Dowdy, use DBR to document specific observations
& Kamphaus, 2012; Dowdy, Chin, Twyford, & about students that are captured through a sys-
Dever, 2011; Dowdy, Dever, DiStefano, & Chin, tematic and semi-structured observation tool in
2011), and measurement invariance was sup- the moment of the behavior’s occurrence, reduc-
ported by a study of children with limited English ing the likelihood of error due to memory bias
proficiency by Dowdy, Dever, DiStefano, and (Christ et al., 2009). Also, specific DBR systems
Chin (2011). A 4-year longitudinal study of the exist, such as the DBR-Single Item Scales (DBR-
predictive validity of the BESS reported that only SIS; Christ et al., 2009), which is a DBR scale
the teacher and student scores were significantly where single items provide information about a
predictive of GPA 4 years later, whereas the par- student’s behavior, as opposed to relying on com-
ent scores were not predictive (Kamphaus & posites of multiple items, which makes the pro-
Reynolds, 2007). A more recent study, however, cess quicker for teachers or other staff who can
found the predictive validity of the BESS Student report on fewer items to document behavior.
Form to decrease over time when internalizing Christ et al. (2009) provides an example of DBR-
problems were used as the outcome variable of SIS where teachers can give a rating on a scale
interest (Dowdy et al., 2016). The Student Form regarding overall disruption of a student (if there
predicted internalizing problems for high school is an operational definition for this), or a teacher
students 1-year post initial screening, but not at can rate several dimensions of disruption (e.g.,
3 years. talking out, aggression) that combined can pro-
vide a score of disruption.
A study of 1108 students in first, fourth, and
irect Behavior Rating (DBR;
D seventh grades across the Midwest, Northeastern,
Chafouleas, Riley-Tillman, & Christ, and Southeastern United States sought to deter-
2009) mine the effectiveness of DBR-SIS as a screening
measure for behavior concerns (Kilgus, Riley-
In the Direct Behavior Rating (DBR), a rater Tillman, Chafouleas, Christ, & Welsh, 2014).
records behavior in specific contexts using an Almost half of students were White (48%), and
observation system (e.g., on a scale from 1 to 10, researchers determined that gender and race/eth-
rating academic engagement; recording behavior nicity reflected typical distribution of school sites
on a daily point card) of standardized behavior at each geographic location. This study con-
ratings. The DBR uses direct observation of cluded that the DBR-SIS could serve as an effec-
behavior (i.e., the observer records the behavior tive behavior screener using multiple or single
in real time), uses operationally defined behav- measures of ratings, given its concurrent correla-
iors that an observer would be able to see in the tions with other screening measures, such as the
environment of the observation (e.g., verbal dis- BESS, and the diagnostic accuracy of the mea-
ruptions in a class period, disengagement), and is sure across grade levels. However, this study did
structured through a series of observer ratings report that, for fourth-grade students specifically,
(Christ, Riley-Tillman, & Chafouleas, 2009). disruptive behavior and respectful behavior
DBR also provides information about the context showed less diagnostic accuracy.
of the behavior across time, which can support
clinicians in better understanding the behavior
occurrences, which can inform intervention. Data ediatric Symptom Checklist (PSC;
P
collection is brief, making the tool more usable Jellinek et al., 1988)
and efficient for teachers or other staff to
administer. The Pediatric Symptom Checklist (PSC) is
DBR can be utilized to track behaviors in chil- unique in that it was specifically designed for use
dren across K–12 with different ratings used by pediatricians to screen for emotional and
depending on what is determined to be the focus behavioral problems. The items are brief state-
of the observations. For example, teachers may ments describing a behavior (e.g., “complains of
Screening Measures 269
aches/pains,” “has trouble sleeping”), and parents ranged from 62% to 69% and specificity from
of the child rate how often that behavior occurs 92% to 98% for predicting mental health prob-
from 0 (never) to 2 (often). There are four scores lems, which was not quite as high as previous
that make up the PSC: Total Score, Internalizing studies found. Overall, the PSC does provide a
Subscale, Attention Subscale, and Externalizing screening instrument that is valid and reliable
Subscale, each receiving a rating of high or lower for use in pediatric clinical settings to assist
risk. This survey can be used as a tool for facili- pediatricians and other health providers in rec-
tating communication between parents and pedi- ognizing psychological issues in children and
atricians regarding children’s difficulties with adolescents.
emotions or behaviors to better assess what is
occurring and to determine whether more inten-
sive examination is needed, potentially by a men- ocial, Academic, and Emotional
S
tal health professional. The PSC may be given to Behavior Risk Screener (SAEBRS;
parents before, during, or after an appointment Kilgus & von der Embse, 2015)
with a pediatrician or mental health clinician.
The PSC is also available in multiple languages The Social, Academic, and Emotional Behavior
and in youth self-report (35- and 17-item ver- Risk Screener (SAEBRS) is a screening measure
sions), a shorter parent form (17 items), and a for K–12 students for social emotional and
picture-based format. behavior problems. There are currently two forms
The PSC has been studied in large and ethni- of this measure for different raters, including
cally diverse samples of participants (e.g., Bernal teachers (SAEBRS-Teacher Rating Scale [TRS])
et al, 2000). A recent study in Florida of 6590 and students (SAEBRS-Student Report Scale
children and young adults aged 6–22 years old [SRS]). The SAEBRS-TRS is completed by
who received Medicaid was conducted between classroom teachers who report on 19 items across
1998 and 2007; children and young adults were 3 domains, social behavior, academic behavior,
53% male, 38% White, and 37% African- and emotional behavior, reporting how often a
American, with nearly half (49%) in Medicaid student engages in each behavior from 0 (never)
because of a documented disability (Boothroyd to 3 (almost always). For example, teachers are
& Armstrong, 2010). In this study, internal con- asked how often students cooperate with peers,
sistency, stability of scores over time, and dis- have impulsive behaviors, or show sadness,
criminant validity and construct validity were among others. Teachers may complete the screen-
documented for the PSC. Boothroyd and ing measure for each student (or classroom) up to
Armstrong (2010) caution on the use of the same five times per year. There is also a form in devel-
cutoff score for males and females, given that in opment for parents (SAEBRS-Parent Rating
their study they found that males tended to have Scale [PRS]) (Taylor, Allen, Kilgus, von der
higher raw and T-scores for most PSC scores. Embse, & Garbacz, 2016). Group screening is
Parents with reading limitations or who speak also available for schools and teachers to track
a different language from their healthcare pro- students in classrooms, which provides critical
vider may benefit from the pictorial PSC, where information for informing whether students need
pictures are provided along with descriptions of more support regarding social emotional and
behaviors (e.g., three images of child show child behavioral skills. Reports are generated that
alone, near another child, or playing with another highlight specific areas (academic, social, and
child, and parents circle or point to the image emotional) in which an individual student may be
that shows how often their child engages with experiencing difficulties, while the group screen-
others). Many early studies of the pictorial PSC ing report provides information about specific
were with Mexican or Mexican-American fami- students in a format where teachers can compare
lies. In a study of 411 Mexican-American dyads across their students.
in the United States, Leiner and colleagues Kilgus and colleagues (2016) examined the
(2007) found that sensitivity of the measure psychometric properties of the SAEBRS-TRS in
270 14 Behavioral and Emotional Risk Screening
2 studies totaling 2398 elementary and middle time), and there is a form for 4- to 16-year-old
school students from schools in the Southwestern children completed by teachers and/or parents, a
and Southeastern United States, of which about self-report form for 11- to 16-year-old children,
half (55%) were Caucasian and half (52%) were and a form for preschool teachers and parents. A
male. While the internal consistency, concurrent longer form exists to target questions relevant to
validity, and diagnostic accuracy were docu- duration, frequency, and level of impairment for
mented, there were mixed findings regarding cut behaviors, and forms exist in over 30 languages.
scores for the Emotional Behavior subscale (e.g., There are five subscales within the SDQ:
items that asked about students being sad or wor- Emotional Symptoms, Conduct Problems,
rying) but not the Social Behavior and Academic Hyperactivity/Inattention, Peer Relationship
Behavior subscales. Specifically, in one study Problems, and Prosocial Behavior; the first four
the cut scores for the Emotional Behavior sub- scales are combined to form the Total Difficulties
scale were not in the sufficient range. A more Score.
recent study by Kilgus and colleagues (2018) Internal consistency estimates for total and
examined the SAEBRS-TRS across 68 teachers subscale scores produced by the SDQ have been
of a combined 1242 elementary school students below the standard of practice in some studies,
in kindergarten through fifth grade, where about with internal consistency estimates as low as .63
half were White (55%) and male (51%). Internal and generally not much higher than .83 (Goodman
consistency was documented in this study; how- & Scott, 1999; Hysing, Eigen, Gillberg, Lie, &
ever, the cut score again for the Emotional Lundervold, 2007). Ruchkin, Jones, Vermeiren,
Behavior subscale was not sufficient. A compari- and Schwab-Stone (2008) found subscale inter-
son between the SAEBRS-TRS and BESS was nal consistency estimates to be frequently below
done in this study, and it was found that for more 0.60 for an urban sample, leading them to the
than 95% of students not identified via the conclusion that the reliability was insufficient,
SAEBRS-TRS as at risk on the Emotional and the issue should be investigated further, per-
Behavior subscale were also identified by the haps ultimately resulting in test modifications.
BESS as not being at risk, meaning there were The internal factor structure of the SDQ has
few false positives. However, about half to two- been studied extensively cross-culturally with
thirds of students identified on the SAEBRS- mixed results (Van Roy, Veenstra, & Clench-Aas,
TRS as at risk regarding emotional behaviors 2008). One study concluded that the SDQ factor
were at risk according to the BESS, which means structure was so problematic that it should be
there may were a number of false negatives. The used with caution and in conjunction with other
SAEBRS-TRS also predicts academic behavior scales (Mellor & Stokes, 2007). In particular, fac-
from the composite score and each subscale (i.e., tor analyses conducted on large US samples have
academic, social, emotional), as demonstrated in yielded poor fit for a five-factor model in favor of
a study of 1058 elementary school students (52% a three-factor model consisting of externalizing
female, 87% White) (Kilgus, Bowman, Christ, & problems, internalizing problems, and positive
Taylor, 2017). construal (Dickey & Blumberg, 2004; Percy,
McCrystal, & Higgins, 2008; Ruchkin et al.,
2008). Similar support for a three-factor model
he Strengths and Difficulties
T was also found in Flemish (Leeuwen,
Questionnaire (SDQ; Goodman, 1997) Meerschaert, Bosmans, De Medts, & Braet,
2006), Norwegian (Ronning, Handegaard,
The Strengths and Difficulties Questionnaire Sourander, & Morch, 2004), and Russian sam-
(SDQ) is a brief, 25-item behavioral screening ples (Ruchkin, Koposov, & Schwab-Stone,
(www.sdqinfo.com, http://www.sdqinfo.com). 2007).
Respondents rate items on a 3-point scale rang- With regard to predictive validity, a study of
ing from 0 (not at all) to 2 (very much or all the 7912 students by Goodman and Goodman (2009)
Screening Measures 271
the PPV and NPV of a screener (Meehl & Rosen, Lochman and the Conduct Problems Prevention
1955). As Hill and colleagues (2004) explained, Research Group, 1995) that have demonstrated
the accuracy rate of a measure will be signifi- that including another informant added little
cantly affected when reported without reference variance to the screening process beyond that
to PPV, NPV, and base rates. For example, “a test provided by the first informant. Jones and col-
with sensitivity of .80 and specificity of .95 has a leagues (2002) concluded, similarly, that the
PPV of about 74% if the base rate is 15%, but the effect of combining parent and teacher ratings
PPV is reduced to 46% if the base rate is 5%” was equal to or minimally better than that of the
(p. 810). When considering emotional and behav- teacher-only rating.
ioral problems in general, prior research suggests Of note, these results which focus on parent
that an annual base rate would be around 20% and teacher ratings largely focused on the screen-
(Campaign for Mental Health Reform, 2005; ing for behavior problems in young children. One
Friedman, Katz-Leavy, Manderscheid, & exhaustive review concluded that self-reports are
Sondheimer, 1996; Hill et al., 2004). However, most useful for detecting internalizing problems
the base rate will be lower when considering a among adolescents (Smith, 2007). Practically
single disorder. speaking, it is impossible to proactively identify
DiStefano and Morgan (2011) used item students’ internalizing concerns without very
response theory (IRT) methods to examine cut obvious behavioral manifestations (e.g.,
scores for the BESS by assessing their accuracy extremely anxious or overtly morose body lan-
when set using various normative cutoffs: at guage) or without directly asking students them-
about the 84th percentile (or T ≈ 60) and the selves about their emotional experiences and how
98th percentile (or T ≈ 70). They found that they evaluate their own functioning (Furlong,
these two cut scores showed reasonable rates of Dowdy, Carnazzo, Bovery, & Kim, 2014).
identification. Thus, while there often is the need Additionally, self-report screeners help meet the
to adjust cut scores based on local needs, it is practical needs of high schools; anecdotally,
somewhat reassuring to know that simple per- teachers have expressed concerns that they do not
centile ranks are still employable for the identifi- see individual adolescents for long enough peri-
cation of BER. ods in a given day to feel confident of their rat-
ings. We suggest that, as have others (Levitt et al.,
2007), adolescents themselves may be in the best
Selecting Informants position to report on their own BER as it relates
to important future academic outcomes and per-
In Chap. 15, we provide a more extended discus- sonal adjustment.
sion of the research guiding the selection of In most cases the practicalities of the setting
informants for clinical assessments of children will often determine the selection of an informant
and adolescents, and we provide methods for for screening purposes. However, a summary of
combining information across informants when the research on the validity of various informants
making clinical diagnoses. While combining for screening suggests the following recommen-
information across informants is also an oft dations for choice of informants:
advised practice for emotional and behavioral
screening (Jensen et al., 1999; Power et al., • Teachers and parents at preschool/daycare
1998; Verhulst, Dekker, & van der Ende, 1997), settings
the advantage, or incremental validity, of com- • Parents in pediatrician offices and other
bining ratings from multiple raters is not clearly healthcare settings
supported according to reviews by Johnston and • Teachers in elementary school and, perhaps,
Murray (2003) and McFall (2005). Support for middle school
this conclusion may be found in studies • Students in middle and high schools and col-
(Biederman, Keenan, & Faraone, 1990; leges/universities
Implementation of Universal Screening 275
a screening system. It is still unknown (a) whether times. Additionally, schools can get a sense of the
multiple screening gates are, in fact, superior to potential problems that the new student may have
single stage screenings and (b) what are the opti- at an early stage, so early interventions can be
mal number of gates to be utilized. Multiple fac- planned accordingly. Research also shows that
tors, only one of which is diagnostic accuracy, children may be particularly vulnerable to emo-
should be considered, including the feasibility tional or behavioral problems when they move to
and cost-effectiveness of multiple-gating proce- a new school (Csorba et al., 2001; Vander Stoep
dures. In fact, Lane et al. (2009) found the seven- et al., 2005), which gives sufficient reason to
item Student Risk Rating Scale (SRSS; screen for problems at this time.
Drummond, 1994) to perform comparatively to
the three-stage SSBD for the identification of
externalizing problems. Therefore, brevity may BER Surveillance
not always denote inferiority. However, a study
investigating a two-stage screening process using Universal screening is not only useful for assess-
the BESS and the BASC-2 demonstrated that ing BER in individual children. The practice may
adding a comprehensive behavior rating scale as also be used to assess the BER of groups of indi-
second gate significantly improved identification viduals, such as classrooms, schools, school dis-
accuracy over a single gate procedure (Stiffler & trict regions, or entire school districts.
Dever, 2015). Furthermore, these questions Surveillance has often been used to track the
require further research because many screening spread of diseases, but the term is not used as
studies have only estimated the effect of using often in the school screening literature, although
multiple gates rather than actually implementing the concept is equally applicable. Citing the
a real-world multiple gate screening procedure National Academies (O’Connell et al., 2009)
(Stiffler & Dever, 2015). report, the goals of universal screening for sur-
veillance are delineated as follows:
The goals and design of these initiatives should be
Timing targeted to relatively narrow and specific purposes,
for example, (1) improving school success for
Another important issue when implementing a struggling students, (2) preventing bullying and
student harassment, (3) improving teacher and
mental health screening is when the screener peer relationships, (4) increasing school safety and
should be administered. For instance, in the con- security, or (5) learning to regulate and control
text of screening in schools, a common time point behavior. (p. 230)
may be as one enters a new school, such as before
entrance to kindergarten or as one transitions to All five of these goals could be informed by uni-
middle school or high school. Vander Stoep and versal screening for BER. Kamphaus, Dever,
colleagues (2005), for example, targeted students Raines, DiStefano, and Dowdy (2011) conducted
during the transition from elementary to middle a study of BER in the interest of meeting goal 4
school in the Developmental Pathways Screening cited above: increasing school safety and security.
Program (DPSP), with the justification that this The authors reviewed the literature on school
was a critical transitional stage. Preschool screen- screening and found that surveillance is being
ers are also available, allowing practitioners to conducted in schools by the CDC and some state
identify problems early in their developmental departments of education or mental health. The
trajectories and to determine readiness for begin- measures used for this purpose assess a variety of
ning kindergarten (DiStefano & Kamphaus, risk behaviors (e.g., using tobacco and alcohol)
2007; Kamphaus & Reynolds, 2007). Time points but not BER. This conclusion was supported by a
involving transitions are also chosen for practi- study by Dowdy, Furlong, and Sharkey (2013)
cality reasons, since often other paperwork or who conducted a study of the California Healthy
screeners are commonly administered at these Kids Survey (CHKS). They found that supple-
Implementation of Universal Screening 277
menting the CHKS with a self-report measure of guidelines can be interpreted in various ways;
subsyndromal BER, the Student BESS in this therefore, it is important to evaluate the differ-
case, significantly increased the prediction of ences between active and passive consent for
cigarette, marijuana, and alcohol use, binge mental health screening.
drinking, physical fighting, threatening or injury Active consent requires that the written con-
by a weapon, skipping school, and serious con- sent form be returned with a signature from a par-
sideration of suicide. Brown and Grumet (2009) ent approving the screening. Passive consent
also provide an example of a school-based sur- requires that parents be provided with an expla-
veillance study. They conducted clinical inter- nation of the screening program in writing with
views to survey suicide risk in 13 middle and the option of declining for their child to partici-
high schools in the Washington, D.C. area. The pate by selecting this option and returning the
Brown and Grumet (2009) study documented passive consent form to their child’s school; if the
that 45% of adolescents screened positive for consent form is not returned, it is assumed that
“previous suicide attempt or ideation, symptoms the parent has given consent. There are benefits
of depression or anxiety, and/or other emotional and drawbacks for each of these methods. The
problems.” Findings such as these may be used to explicit parental consent received through active
guide prevention practices, including the deploy- consent provides solid assurance of parent
ment of mental health personnel in this case. approval. However, while this method appears
legally sound, this sometimes lengthy and com-
plicated process of distributing and collecting
Permissions signed consent forms can introduce an element of
bias in the sample. This bias may undermine the
Screening occurs in US schools routinely and effectiveness of a universal screening approach,
without controversy. Speech, hearing, health, and whereby all individuals have equal opportunity
academic screening are just some examples. for follow-up and resulting interventions. Active
Academic achievement screening occurs most consent processes yield varying return rates from
frequently, allowing schools to make numerous 34% to 67% (Eaton et al., 2004; Esbensen et al.,
decisions about how to help a child achieve. The 1996). The decreased sample size can be disad-
process of obtaining parental consent and student vantageous, as students who would benefit from
assent for screening remains controversial and the screener and/or services associated with it
conflicting, particularly with regard to screening may miss out due to the active consent process.
for mental health disorders. Informed consent While it is important that parents are informed of
may be either active or passive (Eaton, Lowry, and have control over what their children are sub-
Brener, Grunbaum, & Kann, 2004) and is com- jected to, consent forms are frequently not
monly decided by the respective school district returned for a variety of reasons, and the children
and how they interpret the Protection of Pupil might be adversely affected. Chartier et al. (2008)
Rights Amendment (PPRA) (Chartier et al., found that after switching the school-based
2008). The PPRA “seeks to ensure that schools depression screening process from passive
and contractors obtain written parental consent consent to active consent, the participants
before minor students are required to participate decreased from 85% to 66%. The decrease in par-
in any U.S. Department of Education funded sur- ticipation was not equivalent across subgroups of
vey, analysis, or evaluation that reveals informa- students, and the percentage of students posi-
tion concerning [a variety of sensitive behaviors, tively screened for depression was significantly
such as political affiliation and sexual behavior reduced. This illustrates how more stringent con-
and attitudes]” (p. 157). One area included is sent procedures can potentially reduce the effec-
“mental and psychological problems potentially tiveness of the screener, in which at-risk
embarrassing to the student and his/her family” populations may be less likely to be identified
(U.S. Department of Education, 2006). These and served.
278 14 Behavioral and Emotional Risk Screening
(b) Is completed by parents of children ages (f) Has strong evidence to support the pre-
3–17. dictive validity of its scores
(c) Leads to four scales, including Total 10. The Student Risk Screening Scale (SRSS):
Score, Internalizing Subscale, Attention (a) Is a brief 7-item rating scale completed
Subscale, and Externalizing Subscale. by teachers
(d) Has 35- and 17-item versions, as well as (b) Content focuses only on behavior
a picture-based format. problems
(e) The pictorial PSC can be helpful for par- (c) Can be used for grades K through 6
ents with limited reading ability or who 11. Systematic Screening for Behavior Disorders
speak a different language. (SSBD):
8. Social, Academic, and Emotional Behavior (a) Is a universal screening for children in
Risk Screener (SAEBRS): preschool through the ninth grade.
(a) Is a screening measure of grades K (b) Uses a multiple gating method, where
through 12. students are first ranked by teachers in
(b) Has forms to be completed by teachers terms of their level of internalizing and
and students. externalizing behaviors. Students who
(c) Items assess social behavior, academic are ranked as the top three in each group
behavior, and emotional behavior. then receive more intensive assessment.
9. The Strengths and Difficulties Questionnaire 12. When implementing a screener, it is impor-
(SDQ): tant to consider the rate of true and false
(a) Is a brief 24-item screening positives and true and false negatives when
(b) Has versions for parents, teachers, and choosing a cut score to designate BER.
self-report 13. One way to optimize a screener’s ability to
(c) Has been translated into over 30 detect most children in need of treatment in a
languages cost-effective manner is to use a multiple
(d) Includes five subscales, including emo- gating procedure.
tional symptoms, conduct problems, 14. Universal screening can not only be useful
hyperactivity/inattention, peer relation- for assessing BER in individual children but
ship problems, and prosocial behavior may also be used to assess the prevalence of
(e) Has received only modest support for BER across classrooms, schools, and school
the factor validity of the five-subscale districts.
structure
Integrating and Interpreting
Assessment Information 15
overall mean level of agreement between parent being reported by a child’s parent. However, fur-
and child ratings of mental health problems was ther assessment of the home context may indicate
.41 with a range of .17 (Jamaica) to .58 Denmark. that the parents do not put many demands for sus-
One explanation for this low rate of agreement tained attention, organization, or sitting still on
between informants is that different people see a the child in the home. As a result, the behaviors
child in different settings. That is, the low corre- are not problematic in this setting.
lations between informants may reflect real dif- Another important aspect of the lack of agree-
ferences in a child’s behavior across different ment between informants is determining the level
settings. The meta-analysis conducted by of analysis at which agreement is being measured.
Achenbach et al. (1987) provides some support Individual behaviors tend to show less consistency
for this possibility. Across the 119 studies that across informants than broader dimensions or
they reviewed, the average correlation in ratings composites of multiple behaviors. A good example
between informants who see a child in different of this phenomenon comes from a study by
settings (e.g., parent/teacher) was .28. In con- Biederman, Keenan, and Faraone (1990). In a
trast, the average correlation between informants clinic sample of children and adolescents, the aver-
who see a child in the same setting (e.g., pairs of age correlation between parent and teacher reports
parents or pairs of teachers) was .60. This sub- of individual symptoms of Attention-Deficit
stantial increase in agreement between infor- Hyperactivity Disorder (ADHD) on a structured
mants who see a child in the same setting provides interview was quite low (r = .21). However, these
evidence that the lack of informant agreement authors found that, if one looked at the diagnostic
reflects, at least in part, real situational variabil- level, the level of agreement between parents and
ity in children’s behavior, rather than idiosyn- teachers was quite high. If a parent reported that a
cratic judgements made by different informants. child had ADHD on the structured interview, there
Konold, Walthall, and Pianta (2004) also pro- was a 90% probability that the child’s teacher
vide evidence to support the contention that much would report the presence of the disorder. The
of the cross-informant disagreement is due to issue of level of analysis was also evident in the
situationally specific behaviors in children across findings reported by Rescorla et al. (2013), who
settings. Specifically, these authors obtained par- compared the level of adolescent and parent agree-
ent and teacher ratings of young children’s ment on mental health problems across 25 differ-
behavior. They reported that the general form and ent countries. They reported that, while the overall
factor loadings for the ratings were very similar mean level of agreement between parent and child
across mother, father, and teacher ratings, even item ratings of mental health problems was .41,
though the correlations among the ratings from there was much greater consistency across parents
each informant were fairly low. Thus, the low and adolescents on whether any rating fell in the
level of agreement appeared to be more likely deviant range, with a 71–85% agreement on devi-
due to situation-specific behavior than due to the ance status across countries.
manner in which the behaviors were measured. There are two ways one can conceptualize this
This situational variability has important issue. On a psychometric level, classic measure-
implications for clinical assessments. It suggests ment theory states that an aggregate measure
that to make sense of information on a child’s shows higher reliability than its individual compo-
emotional and behavioral functioning, a clinical nents (Petersen, Kolen, & Hoover, 1989).
assessor must also assess the contextual demands Alternatively, the same issue can be framed as
that might influence a child’s behavior and, another aspect of the situational specificity of
thereby, account for the discrepant information behavior. That is, the same underlying construct or
one is obtaining. For example, a clinical assessor psychological dimension (i.e., ADHD) may be
might be quite puzzled over a teacher’s report of present across situations, but the behavioral mani-
significant problems of inattention, disorganiza- festation (i.e., how the symptoms are displayed)
tion, and hyperactivity at school, when it appears may change, depending on the situational demands.
to be in stark contrast to only minimal problems If one frames it in this way, the implication for
Why Is Interpreting Assessment Information So Difficult? 283
clinical assessment is the same as that discussed Cassidy, & Dykas, 2011; Moens, Weeland, van
previously. To understand the different behavioral der Glessen, Chhangur, & Overbeek, 2018;
manifestations, one must understand the demands Ringoot et al., 2015) Possibly because they are
of the different contexts under consideration. indicative of problems in the family system, par-
Karver (2006) provided data to suggest that ent-child disagreements in reporting of child
some of the disagreement between informants is adjustment predict later emotional and behavioral
related to the types of behaviors being assessed. problems in the child (Laird & De Los Reyes,
Using 20 judges of ratings of 59 child behaviors on 2013). Similarly, when parents disagree in their
11 dimensions, three primary dimensions of rating of their child’s personality (Tackett, 2011)
behavior accounted for 43% of the level agreement or parents and children disagree in their ratings of
between parent and child ratings. The first two a child’s personality (Tackett, Herzhoff, Reardon,
dimensions relate to the saliency of the behavior to Smack, & Kushner, 2013), these disagreements
the parent or child (e.g., ease of recall of the behav- are related to both internalizing and externalizing
ior, objective nature of the behavior, perceived problems in the child. This link between infor-
seriousness of the behavior). The third dimension mant discrepancies and both dysfunctional family
involved how observable the behavior was and processes and child adjustment suggest that, while
how willing the parent or child would be to report reported discrepancies on the rating of children’s
the behavior accurately due to social desirability. adjustment are common across most assessments
Another influence on informant disagreement contexts, they are likely to be especially common
is the fact that some apparent disagreement may in clinical assessments of children with emotional
be an artifact of different measurement techniques and behavioral problems.
used across informants. An example is a case in Thus, disagreement among informants on
which a child is showing indications of depression children’s emotion, behavior, and personality is a
on a projective technique but which is not reported common occurrence in clinical assessment that
by parents or teachers on rating scales. A clinical requires careful consideration of its causes. Using
assessor may consider this a disagreement across the available research on this issue, clinical asses-
informants. In fact, this is more a function of the sors should first try to determine if seemingly dis-
lack of agreement across methods, because stud- crepant information can be attributed solely to
ies have shown that the agreement between pro- varying situational demands, the level of analy-
jective and objective self- report measures is sis, the types of behaviors being assessed, or to
generally low (Ball et al., 1991). However, even differences in the assessment strategy. If the dis-
within the same method, what is assessed may dif- crepant information can be explained by these
fer across informants. For example, some rating factors, one has gone a long way in explaining
scales include very different items for parents and the reasons for a child’s problems, and one usu-
teachers (Richters, 1992). As a result, it becomes ally has obtained important information for
unclear whether disagreements are due to actual designing treatment. For example, if a child is
differences in perceptions between parents and only displaying highly anxious behavior in the
teachers or whether they are due to different ques- school setting, one should then look for what
tions being asked of the two informants. aspects of the school environment are leading to
A final issue related to informant discrepancies the anxiety (e.g., social-evaluative situations). In
is that the lack of agreement between informants, doing so, one has pointed the way to one route of
in addition to requiring explanation, can be indic- intervention (e.g., to systematically desensitize
ative of clinically important information (De Los the child to social-evaluative threats).
Reyes, 2011). In fact, when parents and children Unfortunately, in many situations one cannot be
disagree on the child’s level of behavior problems, confident that any of these factors are contribut-
this can be indicative of high rates of parent-child ing to the discrepant information. In such cases,
conflict (De Los Reyes & Kazdin, 2006), parent- one is confronted with the difficult decision of
ing stress (Penny & Skilling, 2012), and other what to do with conflicting information in devel-
problems in the parent-child relationship (Ehrlich, oping a case formulation.
284 15 Integrating and Interpreting Assessment Information
Box 15.2 An Attempt to Develop Rules for (b) Confirmation of antisocial behavior
Combining Multiple Sources of Information from at least one other source.
on Childhood Psychopathology 4. Major Depression
The struggle to develop standardized meth- (a) For children over 13, a diagnosis
ods of integrating information from multiple can be made from child report alone
informants is exemplified in an article by if (1) child reports dysphoric mood
Reich and Earls (1987). These researchers or anhedonia for 2 weeks or longer;
set out to develop “a replicable strategy to (2) at least four additional symp-
make psychiatric diagnoses when reports are toms are present for 2 weeks or lon-
obtainable independently from parents and ger; and (3) there is evidence of
children” (p. 601). The authors clearly state impairment, such as grades drop-
that the goal was not to achieve perfect par- ping, irritability, social withdrawal,
ent-child agreement but to learn how to eval- etc.
uate these different sources of information. (b) A diagnosis of depression would
The following is a summary (see appendix of normally not be made by parents
research article for full criteria) of the deci- report alone unless child report was
sion tree used by these authors to make diag- just under threshold.
noses from structured interviews. We present 5. Separation Anxiety and Overanxious
this approach not so much as a recommended Disorder
decision tree but as an example of strategies (a) Diagnosis can be made from either
that assessors have developed to resolve the parent or child report alone only if
problem of discrepant information. other informant provides some evi-
dence of anxiety and/or depressive
1. Attention-Deficit Disorder symptoms.
(a) Minimum of eight symptoms from
parent report, at least six symptoms Source: Reich, W., & Earls, E. (1987).
from child report, and an age of Rules for making psychiatric diagnoses in
onset before seven. children on the basis of multiple sources of
(b) Evidence of impairment in school information: Preliminary strategies.
from teacher’s report either of sig- Journal of Abnormal Child Psychology, 15,
nificant inattention, disruptive 601–616
behavior, academic underachieve-
ment, or peer difficulties.
(c) If parent reports six or seven symp- Piacentini, Cohen, and Cohen (1992) provide
toms, and child report shows four or an interesting discussion on developing ideal
five, diagnosis can be made with weighing systems for combining information
compelling evidence from teacher. across multiple informants. These authors label
2. Oppositional Disorder systems in which one source of information is
(a) Minimum of two symptoms from weighed more heavily than others (e.g., teachers’
both parent and child and a report of report of inattention given more weight than par-
at least 6 months duration. ents’ report) as complex schemes. In contrast,
(b) Teacher indicates a pattern of nega- simple schemes are those where information from
tive and defiant behavior. all sources is weighed equally. These simple
3. Conduct Disorder schemes can also be called the either/or rule,
since a finding is considered significant if it is
(a) Two or more symptoms from either reported by any informant (e.g., parent, child, or
parent or child. teacher). These authors provide both a theoretical
286 15 Integrating and Interpreting Assessment Information
and empirical rationale for why simple schemes A second issue was raised by Piacentini,
are either as good as or better than complex Cohen, and Cohen (1992), who noted that simple
schemes. First, these authors argue that most schemes are better only if informants are asked to
weighing systems are based on clinical judgment, provide “only that information they would ordi-
in the absence of any clear research base to guide narily be expected to know” (p. 59). In essence,
decisions. They cite numerous findings that this issue implies that there will be certain infor-
“expert judgments” are “almost always inferior mants who have better knowledge of a child’s
to equal weighing in predicting outcomes based adjustment than others. For example, teachers
on multiple variables” (p. 54). This rationale, may not have knowledge about certain behaviors
however, would only hold for weighing systems that occur outside of the school setting, such as
based on clinical judgment. What about empiri- fire setting or cruelty to animals or whether a
cally derived weighing systems? The authors cite child refuses to sleep in his or her own bed. In
a study in which a complex method of combining addition to the type of adjustment being assessed,
information was developed based on logistic we feel that there are other factors that may make
regression analyses. When Bird, Gould, and some informants better than others in a given
Staghezza (1992) compared their empirically case. In the next several sections, we discuss
derived weighing procedure with a simple either/ these factors that may affect the quality of infor-
or scheme, there were no differences in how well mation obtained from specific sources. From the
either system predicted clinicians’ diagnoses. outset, we should note that most of these factors
Moreover, weighing systems based on regression largely will affect the planning of the evaluation
analyses from one sample are likely to underper- rather than the interpretation. Specifically, many
form simple schemes when applied in a different of the factors will affect the choice of informant
sample, because the weighing system is at least and choice of the method used to obtain
partially dependent on the idiosyncratic charac- information.
teristics of the sample. In summary, Piacentini,
Cohen, and Cohen (1992) provide compelling
arguments for the use of simple schemes in com- Informant Discrepancies
bining discrepant reports. and the Type of Problem
Although in principle this approach seems
sound, and we advocate that it should be the pri- The survey by Loeber, Green, and Lahey (1990),
mary or default option for most clinicians, there cited in Box 15.1, indicated that clinicians tend to
are several issues that one needs to consider when weigh adult informants (e.g., parents, teachers)
using this approach. First, this approach assumes more heavily for observable behaviors and tend
that false positives (i.e., a clinically significant to weigh child self-report more heavily for emo-
finding from an informant that is not true) are tional problems. The logical basis to this argu-
rarer and less harmful than false negatives. For ment is that emotional distress is largely an
many clinical situations, this emphasis on ensur- internal subjective event that is more difficult to
ing that children in need of treatment are not tie to behavioral referents that can be observed by
missed is appropriate. However, we also feel that others. However, there is a growing body of
such a decision should not be made blindly. research that has largely supported this method of
Clinicians should always be careful in diagnos- clinical practice.
ing and classifying childhood problems and care- For example, mothers and fathers show a
fully consider both positive and negative higher level of agreement in rating children’s
consequences of any decision (see Chap. 3 for objective, observable, behavioral problems than
issues in diagnosis and classification). Therefore, in rating less observable emotional problems
one must carefully consider the risks and benefits (Christensen, Mergolin, & Sullaway, 1992). Also,
of both false positives and false negatives when research has found that adults tend to report more
developing case formulations. conduct problems and overactivity than is
Integrating Information Across Informants 287
reported by child self-report (Kashani, et al., decrease (e.g., Cantwell, et al., 1997; Elliott,
1985; Reich & Earls, 1987), and children tend to Huizinga, & Ageton, 1985).
report more emotional symptoms than do parents The validation of different informants requires
or teachers (Bird, Gould, & Stagheeza, 1992; the measurement of some independent and clini-
Cantwell et al., 1997; Edelbrock et al., 1986). cally important criteria against which to judge the
This differential prevalence in reporting would usefulness of information provided. This issue
support the common clinical practice of empha- has been very problematic for testing the differ-
sizing adult informants in the assessment of ential validity of informants’ reports of internal-
externalizing problems and emphasizing children izing child behavior because it is difficult to
in the assessment of internalizing problems, if determine what should be the validating criteria.
one assumes that false positives are rare. This One notable exception was a study by Frick,
view is consistent with the either/or approach dis- Silverthorn, and Evans (1994) that used a mater-
cussed in the previous section. nal history of anxiety as a way of validating
However, there are other studies that have informants’ reports of childhood anxiety.
attempted to go beyond simply documenting dif- Specifically, these authors studied the correla-
ferences in prevalence to test the differential tions between parent, child, and teacher reports
validity of various informants on child adjust- of anxiety in children and a history of an anxiety
ment. These studies test the differential validity disorder in mothers. In the sample of children
of various informants across behavioral domains ages 9–13, these authors found that teacher report
using some external and clinically important cri- of anxiety was not related to a family history of
teria. A good example of this approach is a study anxiety but both parent and child reports of anxi-
by Loeber et al. (1991). In a clinic-referred sam- ety were. However, there was additional support
ple of boys between the ages of 6 and 13, these for the relative importance of children’s self-
authors tested the correlations among parent, report of anxiety. Children’s self-reports of anxi-
child, and teacher reports of disruptive behaviors ety disorders were always accompanied by
and several important impairment criteria (e.g., confirmation by their parents. However, parental
school suspensions, police contacts, grade reten- report was not always confirmed by the child, and
tion, special education placement) assessed parent-child disagreements on the presence of an
1 year later. Supporting common clinical prac- anxiety disorder were systematically related to
tice, teachers’ reports of inattentive-hyperactive the parents’ own level of anxiety. These authors
behaviors were the best predictors of impairment concluded that there was evidence for parental
1 year later. For conduct problem behaviors, projection of their own anxiety onto their reports
parental report tended to show the most consis- of anxiety in their children. This finding provides
tent prediction of impairment, although teachers’ some preliminary support for the relative empha-
and children’s information also had utility in pre- sis on children’s self-report of anxiety, at least in
dicting school suspensions. These findings this age group (9–13 years) of children (see also
should be interpreted in light of the age of the Becker, Jensen-Doss, Kendall, Birmaher, &
sample studied. Specifically, this study focused Ginsburg, 2016).
on elementary school-age children. As we will
discuss in more detail in the next section, there is
some evidence that the age of the child may influ- I nformant Discrepancies and Age
ence the importance of different sources of infor- of the Child
mation. For example, the importance of child
report for assessing antisocial behavior, espe- Another factor that might affect the quality of
cially covert antisocial behavior (e.g., lying, information provided by different informants is
stealing), may increase in adolescents, while the the age of the child or adolescent being assessed.
importance of parental report may decrease as First, one would expect that, as a child grows
parents’ knowledge of their child’s behaviors older, parents would have less knowledge of the
288 15 Integrating and Interpreting Assessment Information
child’s emotions and behaviors, especially as with young children. However, clinical assessors
parent-child relationships change in adolescence who work largely with older children, adoles-
(Paikoff & Brooks-Gunn, 1991). Similarly, as a cents, or adults tend to use a traditional approach
child leaves early elementary school, the likeli- to assessment that relies heavily on self-report
hood of a single teacher spending a great deal of and may be uncomfortable with a reduced role of
time with a child decreases. Third, as a child self-report in the assessment of young children.
develops cognitively, he or she may become bet- A final issue related to informant discrepan-
ter able to report on such abstract concepts as cies and age of the child relates to the level of
emotions and thoughts. As a result of these fac- agreement between informants other than the
tors, one would expect the importance of parents child (e.g. parent and teachers). In their meta-
and teachers as informants to decrease with age analysis of 119 studies, Achenbach et al. (1987)
and the importance of children’s self-report to reported that parent and teacher ratings were in
increase. greater agreement for younger children than for
Edelbrock et al. (1985) tested these hypothe- adolescents. As noted above, part of this age
sized age effects on parent and child reports using trend may be due to adolescents spending less
structured diagnostic interviews. Using 2-week time with teachers and parents, and, as a result,
test-retest correlations as the index of reliability, these informants may have less complete infor-
these authors found that the reliability of parent mation on the adolescent’s adjustment. However,
report decreased with age as predicted, although Achenbach et al. (1987) also suggest that younger
the reliability generally remained at an accept- children’s behavior may be more cross-
able level into adolescence. Children’s self-report situationally consistent. Importantly, however,
showed a more dramatic age trend. Children’s these age-related findings have not always been
self-report on the structured interviews showed a consistently reported across samples (De Los
clear increase in reliability with age. Importantly, Reyes & Kazdin, 2005).
the reliability of child self-report on the struc-
tured interviews was quite poor before age 9.
Evidence for the predicted decrease in reliability I nformant Discrepancies: Other
of teachers’ reports also was obtained from Factors
another source. Specifically, in the initial devel-
opment of the Behavioral Assessment System for In addition to type of behaviors and age of the
Children, it was found that the reliability for the child, there are other factors that might influence
Teacher Report Form decreased with age the report of various informants and should be
(Reynolds & Kamphaus, 1992). considered when interpreting discrepant infor-
One question in interpreting these results is mation. In a previous chapter, we discussed fac-
whether the changes in reliability across age are tors within the family, such as parental adjustment
confined to structured techniques, like structured and marital conflict, that may affect the informa-
interviews and behavior rating scales. At least the tion provided by a parent on a child’s adjustment.
findings for children’s self-report are not con- For example, Foley and colleagues (2005)
fined to structured techniques. The reliability of reported on a sample of 2798 twins ages 8–17.
children’s responses to projective techniques has They found that maternal alcoholism and marital
also been found to increase with age (Exner, difficulties were characteristic of cases in which
Thomas, & Mason, 1985). Therefore, the hypoth- an anxiety disorder was reported by mothers but
esized age-related changes in the reliability of not children. Harvey, Fischer, Weieneth, Hurwitz,
various informants have generally been sup- and Sayer (2013) tested predictors of discrepan-
ported in research across a number of assessment cies in informant ratings of preschool age chil-
domains. These findings on the limited reliability dren’s behavior and reported that ethnicity was
of child self-report for very young children may one of the most consistent predictors of discrep-
not be surprising for many clinicians who work ancies, with African-American mothers and
Integrating Information Across Informants 289
fathers being more likely to rate their children’s judgment. Using this general problem orientation
hyperactivity, attention problems, and aggression as a basis, the following section outlines a step-
lower than teachers, whereas Latina mothers by-step strategy that can help to guide the clini-
were more likely to rate their children as more cian in the integration of assessment
hyperactive and inattentive than teachers. information.
Also, informants might have differing motiva-
tions, both conscious and unconscious, that can
affect the information they provide (Karver,
2006). For example, a child may not want to Box 15.3 Research Note: Common Errors in
admit to problem behavior, or a teacher may be Clinical Reasoning and a Problem-Solving
invested in getting a child placed outside of his Model for Developing a Case Formulation
class. Several rating scales discussed in previous Nezu and Nezu (1993) conceptualize a
chapters include validity scales that attempt to clinical assessment as a method of translat-
detect such response sets and aid in the interpre- ing a client’s complaints of distress into a
tation of information provided. meaningful set of target problems and
It is also important to note testing conditions treatment goals. This is what we have
when interpreting the report of different infor- referred to in throughout this book as
mants. For example, a child may have been developing a case formulation. Nezu and
administered a self-report questionnaire after a Nezu conceptualize this process as delin-
long testing session, and it is obvious in watching eating instrumental outcomes (target prob-
him complete the questionnaire that he is not lems) that are believed to directly or
reading the items carefully. indirectly relate to a desired outcome for
the client (treatment goals). For clinicians
to use assessment information in delineat-
Summary ing instrumental outcomes, they must go
through a complex problem-solving pro-
From this discussion, it is clear that, although the cess. Nezu and Nezu first present several
simple scheme of equally weighing the report of common cognitive strategies that can lead
different informants using an either/or approach to errors in the problem-solving process.
is a good starting point, a clinical assessor cannot Next, these authors outline an orientation
use this approach blindly. There are numerous to problem-solving that should limit these
factors that must be considered in trying to errors in clinical reasoning.
explain seemingly discrepant information from
different sources. Errors in Clinical Reasoning
The previous discussion outlines some of the 1. The availability heuristic occurs when
more important issues that have been uncovered clinicians attempt to estimate the prob-
in research that can help guide clinical decision- ability of an event based on the ease
making. However, the final case formulation that with which examples of that event come
results from the integration of multiple types of to mind. Nezu and Nezu use the exam-
assessment information involves a number of ple of a clinician who may overestimate
complex clinical decisions. To aid in this process, the risk of suicide in a new client, if the
in Box 15.3, we provide a summary of an article clinician was recently involved in a case
by Nezu and Nezu (1993) that outlines (1) some of a client who died by suicide but was
of the common cognitive strategies that are used judged to be of low risk.
by people in making decisions but which could 2. The representative heuristic occurs
lead to errors in clinical reasoning and (2) a gen- when a schema is accessed by a given
eral orientation to clinical reasoning that can characteristic to the exclusion of other
minimize the effects of these errors in clinical
290 15 Integrating and Interpreting Assessment Information
schemas. For example, the symptom of outcomes are brought about by a multiplic-
sadness may automatically lead the cli- ity of interacting factors. The basis of this
nician to consider major depression problem orientation is that:
(i.e., diagnostic schema) to the exclu- A variety of biological (e.g., genetic, neuro-
sion of other possible diagnostic sche- chemical, physical), psychological (e.g.,
mas (e.g., medically related mood affective, cognitive, behavioral) and social
disorder, personality disorders). (e.g., social and physical environment) vari-
ables can serve to act and interact as casual
3. The anchoring heuristic occurs when agents or maintaining factors. Such vari-
predictions or decisions are overly ables can influence the pathogenesis of a
dependent on initial impressions and symptom, disorder or both in either a proxi-
later information is discounted, even if mal (e.g., immediate antecedent stimulus)
or distal (e.g., developmental history) man-
the new information is in disagreement ner (p. 256).
with the initial impressions. For exam-
ple, if a depressed client discusses inter- Nezu and Nezu state that even if this
personal problems in the initial meeting model is incorrect in a given case (i.e.,
with the clinician, the anchoring heuris- there is a unitary cause), by starting with
tic would lead the clinician to select the planned critical multiplism orientation,
poor social skills as the primary target the clinician is less likely to selectively
of intervention to the exclusion of other focus on any single type or class of instru-
information (e.g., presence of negative mental outcomes. As a result, this problem
cognitions) obtained in subsequent orientation limits the potential for cogni-
assessment. tive biases.
4 . Confirmatory search strategies involve Source: Nezu, A. M. & Nezu, C. M.
using procedures that seek only to con- (1993). Identifying and selecting target
firm initial impressions, failing to seek problems for clinical interventions: A prob-
disconfirming evidence. For example, a lem-solving model. Psychological
child referred for problems of inatten- Assessment, 5, 254–263.
tion and motor restlessness is assessed
only for other symptoms of ADHD,
without considering other possible rea-
sons for his or her symptoms (e.g., Multistep Strategy for Integrating
A
learning disability, depression, Information
anxiety).
The following strategy assumes one has con-
Problem Orientation ducted a comprehensive clinical evaluation of the
Nezu and Nezu propose that a system- child or adolescent. The prerequisites are (1) hav-
atic problem-solving approach to clinical ing information on a child’s adjustment from
decision-making helps to limit the use of various sources and using multiple methods; (2)
these cognitive heuristics. The key to this having information on the child’s developmental,
approach is the clinician’s problem orienta- medical, and psychiatric history; (3) having
tion. This refers to the clinician’s overall information on multiple areas of functioning,
theoretical orientation for viewing problem such as information on academic capabilities and
behavior, which defines the proper content peer relations, in addition to information on mul-
and methods of assessment. These authors tiple areas of behavioral and emotional function-
recommend an orientation of planned criti- ing; and (4) having information on the important
cal multiplism, which assumes that clinical contexts (e.g., home, school, work) in which a
child functions. Once this information is obtained,
A Multistep Strategy for Integrating Information 291
mental health professional with extensive ioral functioning through structured inter-
training in family-based behavioral tech- views conducted with her and her mother
niques. Such consultation could help them and through rating scales completed by her,
develop strategies to carefully monitor his her mother, and her teacher.
behavior and to develop clear and consis- Step 1: Document All Clinically
tent strategies for encouraging positive Significant Findings on the Child’s
behavior and punishing negative behavior. Adjustment.
Given that he showed limited prosocial As part of the psychoeducational assess-
emotions, it was noted that this interven- ment, Jane was administered the Wechsler
tion should particularly focus on motivat- Intelligence Scale for Children-5th Edition
ing Jake through positive change (WISC-V; Wechsler, 2014). While her
strategies. overall IQ score was within an age norma-
Step 5: Determine Critical Information tive range (96; 39th percentile), her pro-
to Place in the Report cessing speed composite was 69 (2nd
Because the psychoeducational testing percentile) indicating that she was below
did not reveal many significant strengths or age norms in her ability to process visual
weaknesses in cognitive processing or in stimuli quickly. However, her scores on an
various areas of academic achievement, individually administered test of academic
this information was only briefly summa- achievement were all at or above an age
rized, indicating that it was not suggestive normative range. In an unstructured inter-
of the presence of a cognitive deficit or a view, Jane’s mother described Jane as
learning disability and that it indicated that “always being a worrier.” She described
Jake was capable of performing at or above Jane worrying at home when she had to be
grade level. separated from her parents, in school about
her performance and her peer relationships,
and around medical procedures. A struc-
tured interview led to a diagnosis of
Generalized Anxiety Disorder with a Panic
Box 15.5 Case Study to Illustrate Multistep
Attack specifier. On the BASC-3, Jane’s
Interpretive Procedure: A 9-Year-Old Girl
Anxiety subscale score was rated at a
with Generalized Anxiety
T-score of 82 (99th percentile) by both Jane
Jane was 9 years old and in the fourth grade and her teacher. Jane’s teacher also rated
when she was referred for a comprehensive her on the Withdrawal scale at a T-score of
psychological evaluation by her parents. 77 (also 99th percentile). Jane rated her
They were primarily concerned with her Social Stress (T-score of 88), Sense of
poor school performance, inattention, and Inadequacy (T-score of 91), and
moodiness. Jane had done well in school Hyperactivity (T-score of 71) as being ele-
but started showing a decline in her grades vated on the BASC-3 as well.
following an incident of being bullied and Step 2: Look for Convergent Findings
threatened by another student at school Across Sources and Methods
with a knife. The most consistent finding across all
Jane was administered a comprehensive sources of information was that Jane
battery of tests that included an unstruc- showed a high level of anxiety that was not
tured clinical interview with her mother, a specific to any single domain. Most of
psychoeducational assessment, and an Jane’s problems in school appeared to be
assessment of her emotional and behav- due to her high level of anxiety.
294 15 Integrating and Interpreting Assessment Information
ses of anxiety disorders reported, but the parent one area. As a result, the next step in the interpre-
has reported three symptoms of generalized anxi- tive process is to develop a profile of a child or
ety disorder and two symptoms of separation adolescent’s strengths and weaknesses across the
anxiety disorder. Further, a teacher rated the child different domains of psychological functioning
with a T-score of 64 on the anxiety scale of a that have been assessed. In addition, we feel that
behavior rating scale. If one just viewed scores this process needs to go beyond simply
that exceeded a clinical cutoff, this pattern would documenting the different areas of strength and
look like the child’s self-report of anxiety was weaknesses, to also prioritizing the different
unsupported by other informants, when, in fact, areas of concern. This prioritization should be
we would argue that there is fairly consistent sup- both a conceptual and practical endeavor.
port for the presence of anxious behaviors, Conceptually, one should consider what prob-
although the anxiety is not being perceived as lematic area may be primary and which areas
being as severe by parents and teachers. may be secondary, with secondary being defined
as areas that seem to be largely a result of some
other primary factor. For example, for a child
Step 3: Try to Explain Discrepancies who becomes depressed because of her frequent
school suspensions and police contacts for anti-
The complexity of the assessment process is dra- social behavior, the depression may best be con-
matically reduced when assessment information sidered secondary to the antisocial behavior.
is consistent across methods and sources. This is Practically, one needs to consider what area
the ideal case that is the dream of every clinical should be the primary focus of intervention. This
assessor. Unfortunately, existing research and may follow closely with the determination of pri-
clinical experience suggest that this is not likely mary and secondary problems, where interven-
to happen often. In most cases, there will be tion targeting primary areas (e.g., antisocial
numerous discrepancies between the information behavior) may also alleviate problems considered
provided by the different sources after steps 1 and to be secondary (e.g., depression). However, these
2. At this stage, one should take the information two levels of analysis do not necessarily have to
discussed in previous sections of this chapter and be congruent. For example, a child with ADHD
try to develop explanations for the discrepancies. may have developed conduct problems secondary
Can the discrepancies be explained by different to the ADHD because of the effects of a child’s
demands in the various settings in which a child behavior on parent-child interactions. However,
is observed? Can the discrepancies be explained given the current problems caused by the conduct
by differences in the measurement techniques disturbance and the relation to these problems to
used or by certain characteristics/motivations of future problems in adjustment, intervention tar-
the informants? Can the discrepancies be geting the conduct problems may be the primary
explained by differing knowledge of the child’s treatment recommendation, rather than treatment
behavior across informants? As mentioned previ- for ADHD (see Box 15.4 for a case example).
ously, if one can answer these questions and Most assessors would agree with the need to
account for discrepant information, one has gone prioritize problem areas, but most would also
a long way in developing a good case formulation agree that this is difficult to accomplish with any
and developing goals for treatment. degree of reliability and validity. We must admit
that prioritizing areas of need and determining
primary and secondary areas of dysfunction are
tep 4: Develop a Profile and Hierarch
S often largely based on clinical judgment in the
of Strengths and Weaknesses absence of a clear research base to guide this pro-
cess. However, we feel that there are three pri-
Both research and clinical practice indicate that mary factors to consider in prioritizing areas
children rarely have problems that are specific to within a child’s profile.
296 15 Integrating and Interpreting Assessment Information
First, one should look at the degree of impair- tep 5: Determine Critical
S
ment associated with different areas of dysfunc- Information to Place in the Report
tion. As a standard part of any assessment, one
should measure the degree to which various In the next chapter, we provide a detailed discus-
behaviors are affecting a child’s functioning in at sion of report writing. However, it is important to
least three major life areas: in school, with peers, note that the last stage of integrating information
and at home. Then several questions can be is a filtering process. After one has developed a
asked in an effort to determine if some problem profile of the child that explains the referral prob-
areas are contributing more to impairment than lems and suggests treatment goals, it is likely that
others. Are the problematic areas differentially many pieces of information gathered will have
affecting a child’s ability to learn or to perform either minimal or no bearing on the final case
well academically? Has one or more of the areas conceptualization. A good assessor will limit the
significantly affected a child’s ability to form or amount of unessential information that is col-
maintain meaningful relationships with same- lected in an evaluation by carefully tailoring the
age peers? Is any problem area causing a higher evaluation to the needs of an individual child.
level of conflict with parents and/or siblings than However, much information cannot be deter-
others? This method of prioritization is one of mined to be irrelevant until after it is collected.
the most important clinically, because areas that Many clinicians feel that, if information was
cause the greatest degree of impairment typi- collected, it should be included in the written
cally are the most important targets of and/or oral report of the assessment results. Our
interventions. feeling is that too much irrelevant information
Second, one can look at the temporal sequenc- detracts from the case formulation, making it
ing of when problem behaviors developed to help unduly confusing. Therefore, we feel that the
to determine what might be primary or second- final step in the interpretive process is determin-
ary. Did symptoms of depression develop after a ing what information is essential for understand-
chronic pattern of antisocial behavior? A good ing the case formulation and subsequent
example of the therapeutic implications of tem- recommendations for treatment and what infor-
poral sequencing comes from a study in which mation should be omitted or discussed only mini-
children with major depression who only showed mally in the report. This final step sets the stage
conduct problems after the onset of the depres- for the writing of a clear, concise, and readable
sion showed a significant reduction of conduct summary of the assessment results.
problems after the depression was successfully
treated (Puig-Antich et al., 1978).
Third, viewing family history data can help Conclusions
in making a determination of primary and sec-
ondary areas of disturbance. As we discussed in In this chapter, we discuss some of the issues
Chap. 12, in the section on assessing a child’s involved in integrating assessment information
family history, information on parental psychi- across informants, across methods, and across
atric history can often aid in making differen- psychological domains. One of the more difficult
tial diagnoses. This is largely due to the fact aspects of this endeavor is integrating discrepant
that (1) adult psychological disturbance tends information from different informants on the
to be more easily defined and (2) many forms of same dimension of functioning. We discussed
childhood disturbance show a familial link. As evidence that simple decision rules that equally
a result, assessing parental adjustment patterns weigh information given by various informants
can provide clues as to what may be the pri- may be as good or better than more complex
mary area of dysfunction in their child or methods of differentially weighing information
adolescent. provided by various informants. However, this
Conclusions 297
simple rule needs to be qualified by whether the 3. Unfortunately, in many cases it is difficult to
informant is knowledgeable about the child’s attribute discrepant information solely to
behavior and whether the informant can report varying situational demands, to the level of
the information effectively. These two factors can analysis, to the type of behavior being
be influenced by many factors, such as age of the assessed, or to differences in assessment strat-
child and the specific behavioral domain being egy. In these cases, there is evidence that a
assessed. We conclude the chapter by providing a simple either/or system for combining infor-
multistep procedure for integrating the various mation works best. That is, a finding is consid-
pieces of information from a comprehensive ered significant if reported by any informant.
evaluation into a clear case formulation that (a) This approach assumes that false posi-
answers the referral question and points the way tives are rarer and less harmful than false
to a recommended course of intervention. negatives.
(b) This approach also assumes that infor-
Chapter Summary mants are only asked to provide informa-
tion that they should be expected to know.
1. Because most clinical assessments of children 4. The quality of information provided by differ-
and adolescents must be comprehensive, one ent informants seems to vary depending on
of the most difficult parts of the assessment the domain of functioning being assessed and
process is integrating assessment information the age of the child being assessed.
into a clear case formulation that points the (a) Research suggests that teacher report of
way to the most appropriate interventions. inattentive-hyperactive behaviors, parent
2. The most difficult task is integrating informa- report of conduct problems, and child
tion on the same behavior provided by differ- report of emotional problems seem to be
ent informants. Research indicates that poor relatively most important in preadoles-
agreement among informants is the rule, cent children.
rather than the exception. (b) The usefulness of child self-report
(a) These discrepancies may reflect real dif- increases with age, and the usefulness of
ferences in a child’s functioning in differ- teacher report decreases with age. Age-
ent contexts (e.g., home, school, clinic). related changes in parental report are less
Therefore, assessing the characteristics of consistently found, although there are
the different contexts can help to explain some indications that the usefulness of
some discrepancies. parental report also decreases somewhat
(b) Some discrepancies may be an artifact of as the child enters adolescence.
the level of analysis. For example, indi- 5. Other factors, such as a parent’s emotional
vidual symptoms and behaviors may be adjustment, the degree of marital conflict,
very discrepant across informants, idiosyncratic motivations of an informant, and
whereas the presence of a disorder may testing conditions, can influence informant
show greater concordance. agreement.
(c) The level of agreement can be related to 6. A multistep procedure for integrating diverse
certain characteristics of the behaviors assessment information is recommended.
being assessed, such as the salience of the (a) Step 1: Document all clinically significant
behavior to the informant, how observ- findings on the child’s adjustment.
able the behavior is to others, and how (b) Step 2: Look for convergence across
willing the informant is to report on the sources and methods.
behavior. (c) Step 3: Attempt to explain discrepancies.
(d) Discrepancies can also be an artifact of (d) Step 4: Develop a profile and hierarchy of
different measurement techniques used to findings.
obtain information from different (e) Step 5: Determine what critical informa-
informants. tion to place in the formal report.
Report Writing
16
opinions of some stakeholders (e.g., teachers) In most situations, we recommend that the cli-
have been discussed in some work summarized nician attempt to make his or her reports acces-
below, and those findings help provide some, sible and useful to all pertinent audiences so that
although not extensive, guidance on advantageous interested parties do not have different reports for
ways to construct reports. One can get a sense of the same assessment of the same child. Our dis-
why teachers and other professionals are often dis- cussion will focus mainly on the expectation that
satisfied with psychological reports by reading the one report will be made available to parents,
following excerpt that was taken verbatim from a teachers, physicians, etc., with the understanding
report. All of the conclusions drawn by the evalua- that certain presentations of results and interpre-
tor in this case are based on one test requiring the tations will be most useful to certain audiences.
child to simply reproduce nine designs with pencil
and paper. We quote:
The Bender-Visual Motor Gestalt test suggests eport Writing as Part of Evidence-
R
delinquency and an acting out potential. He is anx- Based Assessment
ious, confused, insecure and has a low self-esteem.
He may have difficulties in interpersonal relation- Report writing has not specifically been addressed
ships and tends to isolate himself when problems
arise. ⋯. [He] also seems to have a lot of anxiety in recent writings on evidence-based assessment
and tension over phallic sexuality and may be in of children. Instead, the discussion has focused
somewhat of a homosexual panic. on the use of tools and methods that are valid and
that demonstrate clinical utility (see Hunsley &
This clinician was apparently using a cook- Mash, 2007). We feel that the move toward
book approach to interpretation, conveying no evidence-based assessment should, and will, be
clear evidence to support his/her case conceptual- reflected in the reports that result from psycho-
ization. Furthermore, it is safe to say that the test logical assessments.
in question was to be interpreted for purposes for Mash and Hunsley (2005) point out that
which it had not been validated. A report like this evidence- based assessment is not meant to
is of no help to anyone, especially not the child replace the clinician or the clinician’s judgment.
being evaluated. Similarly, the clinician will continue to be a key
One of the difficulties with report writing is figure in assessment reports. That is, it is unlikely
that different audiences require different reports. and undesirable that reports will be completely
For example, a psychometric summary (i.e., a boiler plate endeavors that do not allow for flexi-
portion of the report that presents only test scores bility based on the particular assessment approach
and is usually given at the beginning or as an used, the client’s particular presenting problems,
appendix at the end of a report) given without or the needs of the client and allied
context is likely to be of little use to parents but of professionals.
great potential use to colleagues and perhaps In contrast, not unlike a scientific manuscript,
teachers. An important decision that each psy- reports from an evidence-based approach to
chologist must make prior to report writing is to assessment can be seen as the means by which a
determine the primary audience for the report. client’s history and difficulties are described,
For example, a psychometric summary may be results are obtained and interpreted, and sugges-
confusing to parents who have contracted with tions for future (treatment) approaches to the dif-
the psychologist in private practice for an evalua- ficulties are discussed. The main sections of most
tion. In this case, it is more sensible to present psychological reports are discussed later in this
test results descriptively and with full attention to chapter, but in many ways, they are analogous to
context in order to communicate effectively with sections of journal articles in psychology and
the parents. A psychometric summary is more in other scientific fields. For example, the “Referral
order in a treatment team situation, where it is Questions” section is essentially a statement of
imperative that a psychologist communicate the evaluation’s purpose. “Background
effectively with knowledgeable colleagues. Information” in a report is similar to a literature
Pitfalls of Report Writing 301
For example, it may be said that a client “man- more positive reinforcement to bring his grades
ifests overt aggressive hostility in an impul- up to pre-trauma levels.
sive manner,” when, in fact, the client “punches Problems may also occur if a psychologist
you in the face.” draws conclusions that are clearly in conflict with
• They are not frequently enough written in lay the data collected for a child. A psychologist may
language. In particular, scores are overempha- decide not to make a diagnosis, in seeming con-
sized, and the fit between the results and the trast to rating scale findings of significant
child’s actual behavior is underemphasized. T-scores on the majority of scales. If a clear argu-
Tallent (1993) argues that an excessive focus ment for resolving this incongruity is not made,
on multiple scores or indices may be a method the consumer of the report may well suspect
to cover up the clinician’s lack of true under- biases. The psychologist who routinely does not
standing of the assessment findings. reconcile high T-scores with a lack of a diagnosis
• They include language that is so vague and may soon be labeled as unwilling to diagnose
unclear that it cannot be falsified or consid- regardless of assessment results. The reverse situ-
ered wrong. ation can also be problematic, wherein the psy-
chologist makes a diagnosis without any clear
These latter two points are critical if reports indications of significant symptomatology or
are to address the referral question in a manner impairment. Teachers, pediatricians, or other
that is amenable to subsequent, appropriate inter- referral sources who receive this interpretation
vention. Of course, psychology cannot be singled consistently from the same psychologist may
out as the only profession with a preference for eventually pay more attention to the data pre-
its own idiosyncratic terminology, as anyone who sented in the reports and ignore the psycholo-
reads a physician’s report or legal contract will gist’s conclusions, or they may simply refer
attest. Perhaps psychologists can, however, lead elsewhere.
the way toward competent reporting of findings.
Report Length
Faulty Interpretation
Psychologists, more so than other groups, com-
Faulty interpretations may be based on personal plain about the excessive length of reports
ideas, biases, and idiosyncrasies (Tallent, 1993). (Tallent, 1993). However, length may not be the
The problem is most readily seen when the psy- real issue. Perhaps long reports are used to dis-
chologist is clearly using the same theories or guise incompetence, fulfill needs for accountabil-
drawing the same conclusions in every report. A ity, or impress others. The possibility that length
psychologist may conclude that all children’s is a cover for other ills is offered in the following
problems are due to poor ego functioning, neuro- example:
psychological problems, family system failure, A business executive likes to relate the anecdote
or some other favored factor. A psychologist who about the occasion when he assigned a new
adheres exclusively to behavioral principles, for employee to prepare a report for him. In due time,
example, might attribute all child problems to a voluminous piece of writing was returned.
Dismayed, the executive pointed out that the
faulty reinforcement histories. The savvy con- required information could be presented on one,
sumer of this psychologist’s reports will eventu- certainly not more than two, pages. But sir, pleaded
ally become wary of the psychologist’s the young man, I don’t know that much about the
conclusions, as the relevance of the favored the- matter you assigned me to (Tallent, 1988, p. 72).
ory to some cases is questionable. One can imag-
ine the skepticism that may be engendered by a It may also be worth considering that the Ten
psychologist who concludes that a child whose Commandments are expressed in 297 words, the
school performance has just deteriorated subse- Declaration of Independence in 300 words, and
quent to a traumatic head injury merely needs the Gettysburg Address in 266 words.
Pitfalls of Report Writing 303
Number Obsession This very obvious point is all too frequently over-
looked. Psychologists should insist that referral
The clinician must always keep clearly in mind sources present their questions clearly, and if not,
that the child is the lodestar of the evaluation, and the psychologist should meet with the referring
the numbers obtained from personality tests and person to obtain further detail on the type of
the like are only worthy of emphasis if they con- information that is expected from the evaluation
tribute to the understanding of the child being (Tallent, 1993). Many agencies use referral forms
evaluated. One way to think of the scores is as a to assist in this process of declaring assessment
means to an end, with the end being better under- goals. A form similar to those used by hospitals is
standing of the child. The same numbers for two shown in Table 16.1, and one suitable for use by
children can mean two quite different things. Just school systems is given in Table 16.2.
as a high temperature reading can be symptom- On occasion, the referral question(s) can be
atic of a host of disorders from influenza to deceptive and, consequently, place the psycholo-
appendicitis, so too a pathognomic behavioral gist in the position of disappointing the referral
sign can reveal a host of possible conditions. source before the evaluation is even initiated.
One error often made when reporting test Under these circumstances, the psychologist may
scores is a psychologist reporting a score and feel helpless or even betrayed because of the neg-
then saying that it is invalid. Then why report it ative reaction of the referral source to the presen-
(Tallent, 1993)? If a test score is invalid, how tation of results and recommendations.
does it serve the child to have this score as part of Psychologists may often need to pursue the true
a permanent record? Reporting apparently invalid referral question. Some examples of stated and
scores is akin to a physician making a diagnostic true referral questions are shown below.
decision based on a fasting blood test when the
Stated referral True referral
patient violated the fasting requirements. In all
Question Question
likelihood, the flawed results would not be
A child’s teacher The teacher is convinced that
reported; rather, the patient would be required to wants to the child has
retake the test. We suggest that one does not have know if a child has ADHD and expects the
to report scores for a test just because it was ADHD psychologist to confirm it
administered. This stance applies to scores that A parent wants to The parent thinks the child is
are deemed invalid or circumstances in which the know why depressed and
a child is failing in would like for her to be on
psychometrics underlying the scores are ques- school medication
tionable. In these situations, disregarding the A psychiatrist wants The psychiatrist has made the
information from the measure or providing only to know diagnosis of
descriptions of the responses given may better if a child is depression and has placed the
inform case conceptualization. Schwean et al. depressed child on
(2006) describe a child-focused report writing medication. The referral was
made simply because a
approach that does not focus on test scores but
second opinion is required for
provides a clearer road map for readers to under- reimbursement purposes.
stand and then intervene. In so doing, the clini- A psychologist The psychologist is seeking a
cian has engaged in best practice and provided wants to know diagnosis of
clarity on the initial referral question. if a child is traumatic brain injury to
neurologically bolster her court
impaired testimony
Failure to Address Referral Questions In all of these scenarios, it would behoove the
psychologist to clearly determine the referral
Tallent (1993) points out that psychologists too source’s actual needs and/or desires early on in
often fail to clarify referral questions, and as a the referral/evaluation process and then deter-
result, their reports appear vague and unfocused. mine the most appropriate way to proceed.
304 16 Report Writing
Patient is a 15-year-old with type I diabetes who has poor adherence to treatment regimen. Parents are concerned
that John is aware of the risks of this poor adherence but seems apathetic. Patient’s affect is flat and may be
depressed. Patient to be discharged from hospital following psych. consultation.
Results of consultation: Patient appears depressed and seems knowledgeable about diabetes and his
diabetes regimen. In particular, his parents noted that he appears sad most of the time, lacks energy, has
reduced his contact with friends, and does not seem interested in activities that he used to enjoy. Rating
scales completed by patient and his mother showed moderate levels of depression. Family history of
depression is significant. Outpatient therapy focused on treatment adherence and depression is
recommended and has been scheduled to begin in 1 week.
COMMUNICATION PROBLEMS
CLASSROOM BEHAVIOR
ACADEMIC PROBLEMS
clear integration of information; (d) inclusion of questions; describe the person; organize the data;
client strengths; and (e) a direct link between the and recommend interventions” (p. 3). If informa-
referral question and the information provided tion is not relevant to these objectives and is very
that serves to answer the referral question. personal, the psychologist should consider care-
Taken together, the work in this area points to fully the decision to invade a family’s privacy by
reports that are approached in a comprehensive, including such information in the report.
client-specific manner and that are clear to a vari-
ety of audiences and focus on psychological con-
structs rather than technical jargon and test Define Abbreviations and Acronyms
scores. Approaching report writing in this man-
ner with a clear eye toward answering the referral Acronyms are part of the idiosyncratic language
question should increase the likelihood that a of psychological assessment. They can greatly
report translates to meaningful interventions for a facilitate communication among psychologists,
client. but they hinder communication with non-
psychologists. Psychologists, just like other pro-
fessionals, need to use nontechnical language to
Best Practices for Report Writing communicate with parents, teachers, and other
colleagues in the mental health field. A pediatri-
Report Only Pertinent Information cian would not ask a mother if her child had an
emesis; rather, the physician would inquire
An important consideration in any report is what whether or not the child vomited.
information to include. Because child assessment When writing a report, psychologists should
routinely involves collecting multiple types of limit their use of acronyms and should define any
data from multiple informants, synthesis of infor- acronym used in a report. For example, to simply
mation can be quite difficult. That is, clinicians use the acronym SAD for separation anxiety dis-
happen onto a great deal of information during order, for example, without defining, is question-
the course of an evaluation, some of which is tan- able practice, particularly when considering that
gential. Say, for example, a child is referred for the same acronym could refer to social anxiety
an evaluation of ADHD. During the course of an disorder.
interview with the child’s father, he recounts at
length his disappointment with his wife. He tells
the clinician that she is dating other men, and he mphasize Words Rather than
E
believes that she is not spending adequate time Numbers
with their children. When writing the report on
this case, the clinician has to determine whether Particularly in the test results section of a report,
this information is pertinent to the ADHD clinicians must resist a temptation to focus exclu-
evaluation. sively on numbers (i.e., test scores). Lichtenberger
Clinicians must think critically about the and colleagues noted that “some evaluators spend
information that they include in reports and con- too much time writing about the obtained test
sider its relevance to the case. Groth-Marnat scores rather than about what these scores mean”
(2009) explains it in this manner: “A poorly inte- (p. 5). Words often communicate more effec-
grated reported uses phrases such as ‘people with tively than numbers because they communicate
these profiles’ or ‘test scores indicate.’ The prob- more directly and in a more accessible manner to
lem with these … is that the referral source(s)… a variety of audiences. The typical question of a
want to know if your interpretations apply to this referral source has nothing to do with the obtained
client” (p. 304). Furthermore, as discussed by T-scores but, rather, the psychologist’s interpreta-
Lichtenberger et al. (2004), the objectives of psy- tion of these scores. Most laypersons will not
chological reports are to “answer the referral understand the T-score metric but can more eas-
Best Practices for Report Writing 307
focused on the word fresh. You wouldn’t sell stale interpretations include anxiety, cooperation,
fish, would you? Once more our hero bowed to the
strength of logic. But finally he was relieved that
dependent, hyperactive, and low self-esteem. One
he had a logic-tight sign for his business; FISH. His way to foster clarity is to use examples (i.e.,
ever alert friend, however, audibly sniffing the air behavioral referents) of the child’s behavior.
for effect, made a final observation: You don’t need Here, for example, are two ways to say that a
a sign (p. 88).
child, Jacob, was anxious.
Jacob exhibited considerable anxiety during the
Psychologists do not need to engage in such testing.
severe editing, but they should at least make an Or, alternatively:
attempt to think critically about their word usage Jacob appeared anxious during the testing. He fre-
quently asked whether or not he had solved an item
to reduce report, sentence, and paragraph length. correctly. He occasionally looked at the ticking
Judicious editing can go a long way toward clari- stopwatch during an item and then hurried, and his
fying meaning in a report. Sometimes new clini- face became flushed when it was obvious to him
cians are not used to critiquing their own writing. that he did not know the answer to a question.
One readily available option is to have a col-
league read reports. Confidentiality, however, Of course, efforts should be made to provide
should be protected if an editor is used. such examples in a concise manner, yet a bene-
fit of using examples of behavior generously is
that it forces the psychologist to consider the
Use Headings and Lists Freely extent of supporting evidence for a conclusion
about a child’s behavior. If a psychologist
Use of heading and clear formatting can enhance writes that a child is anxious but cannot think of
the readability of psychological reports behaviors to help explain this, then the conclu-
(Schneider et al., 2018). If, for example, a clini- sion should not be drawn, as it is insupportable
cian draws a number of conclusions about a child, by evidence.
the conclusions can sometimes lose their impact Direct quotes, to some extent, are also helpful
if they are embedded in paragraphs. for clarifying meaning. If a clinician concludes
As one would predict, the use of headings and that an adolescent is suicidal, a quote may help
lists to excess has a downside. A report that uses clarify this issue considerably. The adolescent
too many lists, for example, appears stilted, and it may have said, “I thought about taking some pills
may not communicate all of the texture and sub- once” or “I feel like I want to run out in front of a
tleties of the child’s performance. Report writers car tonight and if that doesn’t work, I will steal
should consider using additional headings if a my father’s gun and kill myself.” These state-
section of their report stretches for nearly a page ments convey varying degrees of suicidal intent
(single-spaced) without a heading. Clinicians that are most clearly differentiated by quotes.
should consider lists if they want to add impact to
statements and/or conclusions.
Reduce Report Length
but the time to write reports should decline tencies. If, for example, an adjudicated adoles-
with experience). cent who was referred for conduct problems
• The psychologist has difficulty organizing all obtains an elevated T-score on depression mea-
of the details for presentation. sures and no elevations on conduct problem
• Some of the content is not clear or useful. scales, then the score should be double-checked
• The detail is much greater than can be put to to see if a scoring error is the source of the incon-
good use. gruity. Quite simply, if a score doesn’t seem sen-
• Speculations are presented without a good sible, then the clinician should always check for a
rationale for them. scoring error to rule out this possibility.
• The writing is unnecessarily repetitious.
• The organization is not tight.
• The reader is irritated by the length or reads Check Grammar and Spelling
only a few sections such as the Summary or
Recommendations sections. Another problem with reports that detracts from
the credibility of the clinician is the presence of
Harvey (2006) added that “unedited computer- spelling errors. Clinicians are strongly advised to
generated reports can be excessively lengthy and take the time to electronically and visually check
have stilted narratives that appear ‘canned’” their spelling and grammar.
(p. 16). Thus, the ability of the clinical to synthe-
size information into a format that is user-friendly
and specific to the child’s unique circumstances Adaptive Reports to Audience
is a critical clinical skill. Moreover, the issue of and Setting
length is primarily a concern of other psycholo-
gists, and it is intertwined with other issues, such There is probably no optimal report format.
as clarity. Hence, the psychologist in training Psychologists often find that they have to adapt
should not assume that shorter is better. Quality their reports to meet the needs of an ever-changing
is a more important issue than quantity. At this audience. Audiences have varying characteris-
early point in training, the new report writer tics, such as literacy levels, and, more impor-
should keep the issue of length in mind while tantly, they have differing referral questions.
writing reports. Concerns about length, however, In a school setting, many referrals are for
should never interfere with the need to portray a learning problems. Teachers may also be seeking
child’s performance accurately and in the appro- information to assist them in curriculum deci-
priate context. sions. These are very different referral questions
than those that may be of interest in other set-
tings. For example, in a psychiatric hospital,
Check Scores issues such as suicide potential, safety, and cop-
ing strategies may be of greater concern. These
An all-too-frequent and grievous error is to report questions are very different than those in the
scores that are incorrect. Computerized scoring school setting, requiring a focus on topics such as
limits errors. However, in many cases, scoring diagnosis and implications for behavioral inter-
software is based on data that were initially ventions. Parents are yet another audience with
scored by hand by the clinician or responses that specific questions. When conducting an evalua-
were input by the clinician. Thus, scoring and tion for parents in a private practice setting, the
responses should be double-checked against the emphasis may be on advising the parents on what
data being scored by computer software. One they can do to effect change in their child’s
way of checking scores is to be alert to inconsis- behavior.
310 16 Report Writing
The report excerpts used throughout this book stated in terms of specific examples of the child’s
were taken from a variety of settings with differ- difficulties rather than general labels (e.g.,
ing referral questions. The reader is advised to “hyperactivity,” “academic problems,” “anxi-
think carefully about the needs of referral sources ety”). This section may also indicate (briefly) the
when reading these examples and writing reports. duration, severity, and/or frequency of the
problem.
he Sections of Psychological
T
Reports Background Information
His parole officer acknowledges that... statement should be accompanied by the behav-
If care is not taken to make clear the sources of iors that led to this assertion.
information, questions may arise at the time
that feedback is given to involved parties.
Assessment Results
Sensitive background information should also and Interpretation
be corroborated or excluded from the report if it
is inflammatory and cannot be corroborated. For This section is where the test results for the child
example, a 5-year-old may say something like are reported. Some report writers prefer to inte-
“My mother cuts people,” and later, the psychol- grate the results from various measures into a
ogist discovers that the child’s mother is a single section. Still others opt to divide this sec-
surgeon. tion into subsections according to domains
Previous assessment results should also be assessed. The domains may include cognitive/
included in this section (Schneider et al., 2018). intellectual, academic achievement, adaptive
Also, previous experiences with psychological or behavior, visual/motor, and behavioral/personal-
educational interventions should be noted here. ity. This latter section is of primary interest for
The clinician may also refer the reader to a previ- this text.
ous evaluation. Referring to previous evaluations, Organization within the behavioral/person-
without fully recapitulating them, can substan- ality section can be according to theoretical ori-
tially reduce written report bulk. entation, training, or other preferences of the
psychologist. We happen to recommend that
this section be organized from the most impor-
Behavioral Observations tant problem area to least important, such that
all evidence from multiple tools regarding the
In this section, the behaviors that the child exhib- most important domain of functioning for the
its during the assessment are recorded. A brief client (e.g., depression) is discussed first, fol-
description of the setting, particularly if the lowed by other comorbid issues, rather than
report writer is describing classroom observa- presenting information by each individual mea-
tions, is also appropriate. It is also important to sure and then trying again to integrate the infor-
specify the number of sessions during which mation from varied sources. This approach puts
observations were made and any significant con- the focus, in our view, where it belongs: the
trasts in the behavior of the child across those constructs/domains of functioning, not the
sessions. Domains that routinely should be cov- tests. Most importantly, this section should pro-
ered include “physical appearance, ease of estab- vide coherent interpretations of results that
lishing and maintaining rapport, response to relate logically to one another and to other sec-
failures, response to feedback, attention, tions, such as sections devoted to providing
problem-solving strategies, attitude toward self, diagnostic considerations. Hence, this section
unusual mannerisms or habits” (Schneider et al., should not simply report numerical findings
2018), as well as activity level and apparent that are devoid of interpretation. This approach
visual, auditory, language, and motor is reflected in the teacher format preferences
functioning. reported by Pelco and colleagues (2009) who
Care should be taken not to confuse observa- reported that the majority of teachers in their
tions with interpretations. In other words, it is sample preferred that reports be organized by
appropriate, for example, to state that the child themes and in nontechnical language rather
appeared motivated to perform well, but such a than test-by-test.
312 16 Report Writing
The Signatures
Recommendations
Reports typically require signatures attesting to
Recommendations should be specific and clearly their authenticity. An important component of
tailored to the client (Schneider et al., 2018). A this seemingly unimportant aspect of the report is
recommendation for individual therapy may be the necessity for clinicians to use titles that repre-
difficult to carry out, for example, if the specific sent them accurately. Some states, for example,
Communicating Results Orally 313
do not have specialty licensure, and the use of a Communicating Results Orally
title such as Licensed Pediatric Psychologist is
not appropriate. In this case, a more generic term Parent Conferences
such as Licensed Psychologist should be used,
especially if the psychologist lacks evidence of For the purposes of this text, parent is used
board certification of specialty training. Students generically to include any consistent caregiver in
should also be careful to represent themselves the child’s life. Examples of such caregivers
accurately. A title such as practicum student, include stepparents, residential caretakers, and
intern, trainee, or something similar should be grandparents, among others.
used. Psychological custom also dictates the Imparting assessment results to parents
inclusion of the highest degree obtained by the requires considerable skill, as the individual dif-
clinician. ferences between families are myriad. Because of
this diversity, there is not a singular methodology
that will be effective with all parents. This section
Report Writing Self-Test will present some ideas for sharing results with
parents or other caregivers. However, it is vital
A report writing self-test is provided in Table 16.3. for the psychologist to remain flexible in order to
This checklist allows the psychologist to periodi- adapt the format of the feedback session to the
cally and quickly review principles of report needs of the parent, other caregiver, or family, as
writing. well as the setting.
Appelbaum (2009) discusses the challenges daughter outgrow her ADHD?” Psychologists
that may occur during parent conferences, partic- can gauge the probability that such questions
ularly given that different parents may approach will arise by listening carefully in the intake
hearing about assessment results in quite different interview and throughout the assessment
ways. It is advised that clinicians be prepared for process.
a variety of responses and being planful in dis- 6. Schedule adequate time for the interview.
cussing ways that recommendations can be feasi- Parent conferences often become more
bly followed at home (Appelbaum, 2009). involved than one has planned. Adequate
Regardless of the potential pitfalls, parents time allows the psychologist time to use
must be informed of the results of a psychologi- counseling skills to bring a parent feedback
cal evaluation of their child (the legal, ethical, conference to adequate closure. Ideally, 1 to
and regulatory mandates for this practice are 2 hours could be allocated for such a parent
given in Chap. 4). session. If a session ends early, then the psy-
Some helpful suggestions for communicating chologist is the recipient of a precious gift—
test results to parents are given next. extra time.
7. It is often helpful to seek practice communi-
1. Avoid excessive hedging or deceit. The prob- cating with parents from a variety of back-
lem with hedging or failing to report bad grounds. Some parents can be addressed as if
news is that many parents sense this deceit they are colleagues, while others may have
and respond to the psychologist with appro- only a limited grasp of the issues being dis-
priate mistrust. Honesty is also easily sensed cussed. Translators, ministers, teachers,
by parents, which ultimately enhances the trusted family friends, and others may serve
credibility of the psychologist. as allies in the feedback process.
2. Be sure to always provide some discussion 8. Avoid questionable and/or overly explicit
of strengths of both the child and parents. predictions (Kamphaus, 2001). Phrases to be
This should be done early in the feedback avoided would be statements like “She will
because it helps parents to accept the results never go to college,” or “She will always
that may highlight some potential have trouble with school.” These types of
difficulties. statements can be offensive to parents, not to
3. Use percentile ranks heavily when describ- mention inaccurate.
ing norm-referenced test results. This metric 9. Use good, basic counseling skills. Every par-
is easier for parents to understand than other ent likes to talk about the trials and successes
norm-referenced scores such as standard of raising a child. Give parents at least some
scores or T-scores. opportunity to do so, as it allows you to show
4. Instead of lecturing, allow parents opportu- interest in the child by listening to the par-
nities to participate by asking about topics ent’s perspective.
such as their opinion of the results and how 10. Be careful about terms that may sound fine
they fit with their knowledge of their child. to a professional but interpreted badly by
Moreover, listening carefully to parents parents. For example, describing a child as
helps the psychologist determine the psycho- being in the “average range” is considered
logical needs of the parents that are relevant good to professionals. However, no parent
to the evaluation. Similarly, it is essential likes to think of their child as “average.”
that the parents be given frequent opportuni- Describing the score as indicating that the
ties to ask questions (Lichtenberger et al., child is “right where he/she should be” is
2004). often better accepted by parents than
5. Anticipate questions prior to the interview “average.”
and prepare responses. How would a psy- 11. Do not engage in counseling that is beyond
chologist answer the question, “Will my your level of expertise. Parents are often very
Communicating Results Orally 315
eager to obtain advice from a professional. It brief lunch, when they prefer to unwind with
is inappropriate (and unethical by most stan- colleagues and prepare for the remainder of
dards) for a psychologist to provide services the day. A clinician is unlikely to command a
for which he or she is not trained. If, for teacher’s undivided attention during such
example, a parent requests marital counsel- breaks. If a teacher has an additional free
ing and you have no training in this area, you period, it may be a good time for a conference.
should inform the parent of this fact and After school is frequently the best time to get
offer a referral. In fact, the psychologist is a teacher’s undivided attention for a meeting.
wise to have referral sources readily avail- Teachers are generally very busy people, so
able for such eventualities. the pace of the meeting will be quicker than is
12. Be aware that some parents are not ready to the case for parents.
accept some test results. Parents may impugn 2. Teachers are interested in schooling issues.
your skills because they cannot accept the fact The diagnosis of Conduct Disorder is of less
that their child has a severe handicap. They concern to teachers than getting specific rec-
may leave the session angry, and you may feel ommendations for helping the child in the
inept. The idea that every parent conference classroom (Pelco et al., 2009). If a psycholo-
will end on a happy note is unrealistic. gist is not trained and/or has little experience
Examine your skills critically in response to in teacher consultation, then the assistance of
parent feedback but realize that some parents someone like a qualified school psychologist
simply will not accept the results because of should be enlisted to assist with the teacher
their own personal difficulties with the results. conference.
An example of such a situation may involve a
parent with the same disability as the child. If In any assessment with a school-aged child, the
a parent was labeled and ridiculed by peers, clinician should be prepared to conduct a teacher
then this parent may become defensive and conference or at the very least be available to
angry at the suggestion that his or her child answer any questions and facilitate the imple-
may have similar problems. The session with mentation of classroom-based interventions.
such a parent will likely end on a tense note. Such a conference is desirable because teachers
In many of these cases, however, the parent are usually involved to some degree in the treat-
will adapt and accept the news after develop- ment of children and adolescents.
ing the psychological resources to cope with
the attendant stresses. The psychologist may
find this same parent to interact more posi- Child Feedback
tively in the next encounter.
13. Maintain a positive tone throughout the ses- Providing assessment feedback to a child is often
sion and discuss the child’s strengths and overlooked. However, as noted in Chap. 4, chil-
competencies. dren have the right to have the results of the eval-
uation described to them in ways that are
developmentally appropriate. Further, providing
Teacher Conferences the child with a clear explanation of the results is
important because the child will likely begin
Many of the principles used in parent confer- some interventions or experience some changes
ences also apply to teachers. Several nuances, in his or her environment directly related to the
however, will be outlined in the following assessment.
suggestions. The major decision that a clinician needs to
make before giving feedback to a child regards
1. Do not monopolize a teacher’s time. Some the type of information that is appropriate for a
teachers get few breaks in a day. Most get a child’s developmental level. Clearly, the kind of
316 16 Report Writing
feedback given to parents is inappropriate for a In most cases involving feedback to children
5- or 6-year-old, who may have extraordinary or adolescents, it is advisable to consult with a
difficulty understanding the concept of a per- fellow professional (e.g., teacher, counselor,
centile rank or even what a “disorder” is. A speech therapist, etc.) who knows the child
child this age, however, may be able to under- extremely well. This colleague can help the psy-
stand the consequences of the evaluation. In chologist gauge the ability of the child to deal
this situation, the child may be able to under- appropriately with the assessment results and
stand something like: “Remember those tests I associated interventions.
gave you? Well, some of them seemed hard for
you. Because of this, I suggested to your par-
ents that you be helped after school. So now, Conclusions
you will be going to visit a teacher after school
who will help you with schoolwork.” Also, Report writing and oral feedback are central, not
describing the signs of a diagnosis is often more ancillary, considerations in the assessment pro-
helpful to a child than using the name of the cess. The most insightful, comprehensive, and
disorder. For example, telling a child that he elegant evaluations are lost if not translated to
needs “to keep working hard at paying attention usable information in written reports and feed-
and staying in his seat” is easier to understand back sessions. Unfortunately, these central
than he “has ADHD.” assessment skills are underemphasized in the
The older the child, the more similar the feed- training of clinicians who are left to acquire these
back session becomes to the one for parents. One skills through trial and error. Clinicians are
dramatic difference, however, is that negative advised to seek out expert supervision in this
feedback to a child or adolescent can have the area, if it is not readily offered. In addition, enlist-
opposite of the intended effect. That is, in most ing the aid of a competent editor can markedly
cases, the goal is to improve variables such as enhance the quality of written work. Writing is
peer-related social skills. A child who is told that not easy. Writing skills, however, can be acquired
he or she has poor social skills may decide to stop and improved with diligence and patience.
trying to interact with peers or develop increased
stress concerning peer interactions. In some Chapter Summary
cases, the clinician’s honesty could harm the
child. A few options are available in cases where 1. Psychological reports are frequently made
a clinician is concerned about such negative con- available to parents, judges, lawyers, and
sequences. One option is to have someone who other non-psychologists, creating the oppor-
knows the child well and has a positive relation- tunity for improper interpretation of the
ship with the child to help the psychologist com- results by untrained individuals.
municate the results in a nonthreatening way. A 2. Past psychological reports can be useful to
good person to fill this role is a teacher, other pro- other clinicians who evaluate a child.
fessional caregiver, or possibly a parent. A sec- 3. Different audiences require different types of
ond possibility, if applicable, is to have the child’s written reports.
primary therapist or counselor eventually share 4. Some of the common problems with report
the results with the child in a counseling session, writing include:
when he or she could help the child cope with the (a) Vocabulary problems
results in a supportive setting. However, because (b) Faulty interpretation
children have a right to know the results of the (c) Report length
evaluation, it is important that the clinical asses- (d) An emphasis on numbers rather than on
sor ensures that such an explanation actually what the numbers mean
takes place. (e) Failure to address referral questions
Conclusions 317
5. Some research has shown that teachers pre- (j) Psychometric summary
fer reports that are organized by themes, (k) Signatures
rather than by tests, and that are nontechnical 9. Hints for communicating test results to par-
in language. ents include
6. Parents also tend to prefer a question-and- (a) Be direct and honest.
answer format to other formats when dis- (b) Use percentile ranks heavily when
cussing assessment results. describing test results.
7. Suggested report writing practices include: (c) Be sure to discuss strengths of the child
(a) Report only pertinent information. and parents.
(b) Define abbreviations and acronyms. (d) Allow parents opportunities to partici-
(c) Emphasize words rather than numbers. pate and ask questions.
(d) Reduce difficult words. (e) Anticipate questions prior to the inter-
(e) Describe the tests used. view and prepare responses.
(f) Edit the report multiple times. (f) Schedule adequate time for the
(g) Use headings and lists freely. interview.
(h) Use examples of behavior to clarify
(g) Practice communicating with parents
meaning. from a variety of backgrounds.
(i) Reduce report length. (h) Avoid questionable predictions.
(j) Check scores. (i) Use good, basic counseling skills to con-
(k) Check spelling and grammar. vey difficult information.
8. Psychological reports often include some or (j) Do not engage in counseling that is
all of the following headings beyond your level of expertise.
(a) Identifying information (k) Be aware that some parents are not ready
(b) Assessment procedures to accept some of the conclusions
(c) Referral question(s) offered.
(d) Background information 10. Teacher conferences are important for ensur-
(e) Behavioral observations ing cooperation with recommendations.
(f) Assessment results and interpretation 11. The major decision that a clinician needs to
(g) Diagnostic considerations make before giving feedback to a child is the
(h) Summary type of information that is appropriate for the
(i) Recommendations child’s developmental level.
Assessment of Attention-Deficit/
Hyperactivity and Disruptive 17
Behavior Disorders
we first provide a brief discussion of the most dren could have attention deficits in the absence
clinically relevant research findings and then dis- of motor hyperactivity. The next revision, DSM-
cuss specific recommendations for assessment III-R, softened this emphasis on attention defi-
procedures based on these findings. We divide cits, placing it on equal footing with motor
our discussion into two sections corresponding to hyperactivity in its definition of Attention-Deficit
the major subdivisions within the externalizing Hyperactivity Disorder (American Psychiatric
disorders. The first section involves a discussion Association, 1987).
of a syndrome of behaviors involving inattention- One source of contention in these early defini-
disorganization and impulsivity-motor hyperac- tions of ADHD was over the core features of the
tivity labeled as Attention-Deficit Hyperactivity disorder. For example, DSM-III proposed three
Disorder (ADHD) by the DSM-5 (American core dimensions of behavior associated with
Psychiatric Association, 2013). The second sec- attention-deficit disorders: (1) inattention (e.g.,
tion focuses on conduct problems and aggression very distractible, difficulty finishing things), (2)
subsumed under the categories of Oppositional impulsivity (e.g., acts without thinking, inter-
Defiant Disorder (ODD) and Conduct Disorder rupts others, loses things, makes careless mis-
(CD; American Psychiatric Association, 2013). takes), and (3) hyperactivity (e.g., fidgety and
restless, running around and climbing exces-
sively). In contrast, DSM-III-R eliminated any
Attention-Deficit/Hyperactivity distinctions among these behaviors and consid-
Disorder ered all three dimensions to be indicative of a
single domain of behavior. Research has fairly
Classification and Subtypes consistently suggested that the best method for
conceptualizing the symptoms is somewhere in
History The syndrome of ADHD has long been between. Specifically, factor analyses have gen-
recognized in the medical and psychological lit- erally been able to document two partially inde-
erature, with some descriptions dating back well pendent dimensions of behavior: inattention/
over 100 years (Smith, Barkley, & Shapiro, disorganization and impulsivity/overactivity
2007). However, over the years there has been (Frick & Nigg, 2012). The behaviors that are gen-
considerable disagreement over what are the core erally considered to be the core features of these
features of the disorder. As a result of this confu- two dimensions of ADHD are listed in Table 17.1.
sion, there have been numerous changes in diag- Based on the consistency of the research support-
nostic definitions (see Frick & Thornton, 2017). ing this two-dimensional approach, this has been
This evolution in our conceptualization of ADHD used to form the diagnostic criteria for ADHD in
is reflected in changes in the diagnostic criteria the two most recent versions of the manual: the
for ADHD over the most recent revisions of the DSM-IV (American Psychiatric Association,
Diagnostic and Statistical Manual of Mental 1994) and the DSM-5 (American Psychiatric
Disorders. Association, 2013).
The second edition of the DSM included a The second source of contention over the most
syndrome of the Hyperkinetic Reaction of recent versions of the DSM has been the debate
Childhood to emphasize the belief that motor over whether there are valid subtypes of ADHD.
hyperactivity is the core feature of the disorder Unfortunately, for this issue, there has not been
(DSM-II; American Psychiatric Association, strong consensus (Frick & Nigg, 2012). DSM-III
1968). In the third edition of the DSM, the disor- proposed the existence of two types of attention-
der was reconceptualized to emphasize deficits in deficit disorder: Attention-Deficit Disorder with-
sustained attention, acquiring the term Attention- out Hyperactivity (ADD/WO) and
Deficit Disorder (DSM-III; American Psychiatric Attention-Deficit Disorder with Hyperactivity
Association, 1980). Also in the DSM-III, it was (ADD/H). This distinction was made largely to
explicitly recognized for the first time that chil- reflect the fact that some children present with
Attention-Deficit/Hyperactivity Disorder 321
Table 17.1 Core dimensions of attention-deficit/hyper- marily by elevated rates of impulsivity and hyper-
activity disorder
activity, ADHD-Predominantly
Impulsivity- Hyperactive-Impulsive Type (ADHD-PHI).
Inattention-disorganization hyperactivity
Subsequent research on these subtypes called
Difficulty organizing thingsa,b Excessive running
and climbinga,b into question their validity for a number of rea-
Difficulty following through on Difficulty playing sons. First, the definition of ADHD-PI allowed
instructionsa,b quietlya,b children to have above-average levels of
Often loses thingsa,b Excessive talkinga,b impulsivity-hyperactivity (or hyperactivity/
Easily distracteda,b Frequently impulsivity), as long as they did not reach the
interrupts and
threshold of six symptoms. As a result, this sub-
intrudesa,b
Often does not listena,b Always on the goa,b type led to a heterogeneous grouping of children
Difficulty concentrating and Excessive fidgeting who varied greatly on the amount of impulsivity
sustaining attention a,b
and squirming a,b and hyperactivity they displayed (Diamond,
Difficulty finishing tasksa,b Difficulty staying in 2005; Milich et al., 2001). Second, a meta-
seata,b analysis of five longitudinal data sets by Willcutt
Often avoids or dislikes tasks Difficulty waiting
et al. (2012) reported a high degree of instability
requiring sustained mental turna,b
efforta in these subtypes over time, with the majority of
Often forgetful a
Frequently blurts children changing subtypes when reassessed at
out answersa,b later time points. Third, the Willcutt et al. (2012)
Misses details and makes Frequently calls out review also reported that there were few consis-
careless mistakesa in classb tent unique neuropsychological or cognitive
Needs a lot of supervisionb
problems associated with the different subtypes
a
Symptoms included in the DSM-5 (American Psychiatric
specified in the DSM-IV, and most of those that
Association, 2013) criteria for ADHD, although wording
may not be exactly as that included in the manual were found could be explained by children with
b
Behaviors included in factor analyses reviewed by Lahey ADHD-CT showing more symptoms than those
et al. (1997) with ADHD-PI. However, in studies that used a
more restrictive definition of ADHD-PI that sig-
inattention but without hyperactivity, and these nificantly limited the degree of hyperactivity that
children often show a different pattern of atten- could be present, some differences did emerge,
tion deficits (e.g., a sluggish cognitive style) and such as those with attentional problems without
different pattern of comorbidity (e.g., more inter- hyperactivity showing an atypical attentional
nalizing and learning disorders; Milich, blink response (a measure of early stage gating of
Balentine, & Lynam, 2001; Penny, Waschbusch, information; Carr, Henderson, & Nigg, 2010)
Klein, Corkum, & Eskes, 2009). After the sub- and showing a sluggish cognitive tempo
types were eliminated in the DSM-III-R, they (Derefinko et al., 2008) relative to those with
were again added in the DSM-IV definition of ADHD-CT. Thus, based on this available
Attention-Deficit Hyperactivity Disorder research, it was clear that there were some major
(ADHD). Specifically, similar to subtypes used limitations in how the DSM-IV subtypes were
in the DSM-III, the DSM-IV included a subtype defined. Unfortunately, there was not an alterna-
defined primarily by symptoms of inattention and tive method that had been used in research suffi-
disorganization, ADHD-Predominantly ciently to warrant its adoption in the DSM-5
Inattentive Type (ADHD-PI), as well as a sub- (Frick & Nigg, 2012). As a result, the DSM-5
type defined by elevated rates of both inattention- chose to rename the different symptom patterns
disorganization symptoms and as “presentations” rather than “subtypes” to doc-
impulsivity-hyperactivity symptoms, ADHD- ument the very different symptom presentations
Combined Type (ADHD-CT). The DSM-IV also that can occur but to be more consistent with the
added an additional subtype that was not included instability shown across time in these
in previous versions of the manual defined pri- presentations. However, it is important to note
322 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
that this issue of whether there are important sub- Table 17.2 Key parts of the DSM-5 criteria for attention-
deficit hyperactivity disorder
types of ADHD remains unsettled, and it will be
important for future research to determine if there Distinguishing normal
inattention-
are alternative methods for defining subtypes that disorganization and
might implicate distinct etiological factors across hyperactivity- Change in the criteria for
subtypes, especially those with attention deficits impulsivity from ADHD ADHD in the DSM-5
and no hyperactivity, and lead to more stability in Child must shows 6 or Older adolescents and adults
the presentations over time (Frick & Nigg, 2012). more symptoms of (age 17 and older) must only
either inattention- show 5 or more symptoms
disorganization or Symptoms were reworded to
DSM-5 Thus, the DSM-5 maintained the two- impulsivity- give examples relevant to
dimensional symptom list from the DSM-IV that hyperactivity older adolescents and adults
includes symptoms of inattention-disorganization Person must show Symptoms must be present
symptoms for at least prior to age 12 years
and impulsivity-hyperactivity. Further, the defini-
6 months starting in
tions of ADHD-PIT, ADHD-IH, and ADHD-CT childhood
all remained relatively unchanged, although they Several symptoms
were now labeled as presentations rather than must be present in two
subtypes. In Table 17.2, we provide a summary or more settings
Symptoms must
of the major changes that were made in the diag-
interfere with the
nostic criteria in the DSM-5. Also, we summarize child’s quality of
key parts to the DSM-5 criteria that help to dif- social, academic, or
ferentiate when a child shows what should be occupational
functioning
considered normal levels of these behaviors and
Specifiers of Predominantly
when the child should be considered as showing Inattentive, Predominantly
ADHD. Hyperactive-Impulsive, and
Combined are now called
One important criterion to differentiate nor- “Presentations” to recognize
instability across time
mal from disordered levels of inattention and
Severity specifiers are now
hyperactivity is the number of symptoms the added to include:
child shows. In the DSM-IV, a person was Mild: Few, if any symptoms
required to show six symptoms of either in excess of those required to
make the diagnosis and no
inattention-
disorganization or impulsivity-
more than minor
hyperactivity as the diagnostic threshold for the impairments in functioning
disorder. This level of severity was chosen based Moderate: Symptoms and
on evidence that it seemed to designate a level of impairment between mild
and severe
symptomology that predicted clinically signifi-
Severe: Many symptoms in
cant levels of psychosocial impairment (e.g., excess of those required to
poor academic performance, social rejection) for make the diagnosis, several
elementary school-aged children (Lahey et al., symptoms are particularly
severe, or the child shows
1994). However, there were three concerns raised
marked impairments in
about these criteria when applied to adults (Frick functioning
& Nigg, 2012). The first concern was whether the
symptoms were worded in ways that make them
applicable to adults (e.g., runs about or climbs on Third, there were concerns over whether the
things). The second concern was whether other number of symptoms to make the diagnosis
symptom domains not included in the current should be the same for children and adults. The
diagnostic criteria may be important for diagnos- DSM-5 chose to address two of these issues in
ing adults, such as problems in executive func- their criteria for ADHD. First, the wording of
tioning and problems in emotional regulation. symptoms was revised to include examples that
Attention-Deficit/Hyperactivity Disorder 323
are more relevant across the lifespan. For exam- et al., 1997). Importantly, prospective studies fol-
ple, the symptom “often has difficulty organizing lowing samples of children with ADHD into
tasks and activities” now includes the examples adulthood suggest that nearly all persons identi-
of difficulty managing sequential tasks; difficulty fied with ADHD over the lifespan could have
keeping materials and belongings in order; messy been identified by age 12–14 based on the symp-
and disorganized work; has poor time manage- toms they showed at that time (Kieling et al.,
ment; and fails to meet deadlines (American 2010). As a result, the DSM-5 chose to place the
Psychiatric Association, 2013). Second, the diag- minimum age of onset for a diagnosis of ADHD
nostic cut point for adults was lowered to 5 (but as being prior to age 12.
remained at 6 for persons ages 16 and younger) One part of the DSM-5 criteria for ADHD that
based on longitudinal research suggesting that did not change was the requirement of cross-
this cut point identified more adults who had situational consistency of symptoms. That is, to
been identified with ADHD in childhood be diagnosed with ADHD, a child must show
(Barkley, Murphy, & Fischer, 2008; Mannuzza impairment related to the symptoms in two or
et al., 2011). more settings. This criterion is consistent with
Since the earliest definitions of ADHD, the the view that ADHD should not be solely a func-
disorder has been viewed as starting early in tion of a single set of environmental circum-
development, relatively chronic throughout the stances (e.g., a disorganized classroom, a chaotic
lifespan, and not a transient reaction to a specific home environment) and clearly suggests that an
stressor or to the demands of a particular devel- adequate assessment of ADHD must involve an
opmental stage (Barkley, 1997a). Thus, there has assessment of the child in several different set-
always been a minimum age of onset required for tings. However, although this criterion appears
making the diagnosis, with the DSM-IV specify- quite basic, it is difficult to use in clinical assess-
ing a minimum age of 7 years (American ments for several reasons. For example, it is clear
Psychiatric Association, 1997). While the age of that children with ADHD will show variations in
onset criterion is consistent with this conceptual the level and severity of their behavioral prob-
framework, there have been several problems lems depending on the demands of the situation
with this part of the criteria. First, it is often dif- (e.g., time of day, level of structure, and complex-
ficult to gain accurate accounts of when the ity of the task; Barkley, 1997a). Therefore, an
symptoms first became problematic, especially assessor should not interpret the cross-situational
when assessing adolescents and adults, which criterion to imply that a child with ADHD must
involves recall of events over a long period of show the same level and severity symptoms
time (Barkley, 1997a). Second, the support for across different settings with different demands.
the validity of this cutoff has been questioned, Instead, one must consider whether the child
based on findings that adolescents and adults shows similar behaviors in situations with equiv-
who show enough symptoms to meet criteria for alent demands, and such judgments are very dif-
ADHD but whose symptoms emerge after age 7 ficult to make. For example, if a child is having
have the same symptom profile, course, impair- trouble associated with ADHD at school but not
ment, severity, treatment response, and neurobio- at home, the clinical assessor must judge whether
logical findings as those identified as having an or not this is due to the fact that demands for sus-
onset prior to age 7 (Applegate et al., 1997; tained attention and sitting still are not placed on
Kieling et al., 2010). Finally, it is not uncommon the child at home.
for many of the symptoms of ADHD, especially Finally, as in previous versions of the DSM,
the inattention ones, to only become problematic the DSM-5 criteria for ADHD require that the
once the demands for sustained attention and symptoms must lead to significant levels of
organization increase in later elementary school impairment in functioning for the child, such as
years. Thus, the majority of children with ADHD- causing problems completing schoolwork,
PIT do not show an onset by age 7 (Applegate leading to being rejected by peers, or leading to
324 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
substantial conflict with parents at home. child often is the best predictor of how well a
Importantly, the impairment associated with the child will do later in life, and the impairments
symptoms may be the most important consider- experienced in family functioning, peer relation-
ation in determining whether a child should be ships, and academic functioning are often the
diagnosed with a disorder. The types of impair- main reasons that a child with ADHD is referred
ments associated with ADHD may change across for an evaluation (Pelham, Fabiano, & Massetti,
development, as shown in Table 17.3. However, 2005). Also, new to the DSM-5 was a way to des-
the degree of impairment experienced by the ignate the severity of the child’s symptoms and
the impairments associated with them.
Table 17.3 Developmental changes in the core features Specifically, the DSM-5 included a severity spec-
and secondary characteristics associated with ADHD ifier, in which a child could be designated as hav-
across the lifespan
ing a “mild” form of ADHD, if he or she showed
Age Primary Features few, if any, symptoms in excess of those required
Characteristics Secondary to make the diagnosis and showed no more than
Preschool Restlessness, Noncompliance,
minor impairments in functioning. In contrast, a
excessive accidental injuries,
activity, aggression, and child is considered to have a “severe” form of
difficulty problems in toilet ADHD, if he or she shows many symptoms in
remaining training excess of those required to make the diagnosis,
seating, and
several symptoms are particularly severe, or the
acts without
thinking child shows marked impairments in functioning.
Elementary- Poor attention Difficulty following A child is considered to show “moderate” ADHD
school age span, rules, immaturity, if he or she shows a number of symptoms and
distractibility, social rejection, level of impairment between mild and severe.
impulsivity, needing a lot of
difficulty supervision, failure
remaining to do household
seated, and chores, and poor Comorbidities
can’t play school performance
quietly
Children with ADHD are an excellent example of
Adolescence Poor attention Poor school
span, performance the fact that children with problems in one area of
distractibility, (dropping out), low adjustment are at risk for problems in other areas
inability to self-esteem. as well. Problems that often co-occur with ADHD
finish things, depression,
are quite important clinically. They often cause
careless substance use,
mistakes, and delinquency, family more disruptions for the child and predict poorer
lack of conflict, risk taking, outcomes than the primary ADHD symptoms
forethought and automobile themselves (Frick & Lahey, 1991). As a result,
planning accidents, teenage
the secondary features are often a major focus of
pregnancy, and
rebelliousness intervention (Pelham et al., 2005).
Young adults Restlessness, Poor educational
distractibility, and occupational Conduct Problems/Aggression The most com-
difficulty performance, mon co-occurring problems experienced by chil-
completing depression,
work, substance use, poor dren with ADHD are conduct problems and
carelessness, social/marital aggression, with research suggesting that 60–75%
and lack of adjustment, and of children referred to clinics with ADHD show
forethought and poor anger significant levels of these problems (Hinshaw,
planning management
1987). This high co-occurrence between ADHD
Sources: Summarized from Barkley (1997a), Barkley,
Guevremont, Anastopoulis, and Fletcher (1993), DuPaul
and conduct problems is true for both boys and
and Stoner (1994), Nadeau (1995), Smith et al. (2007) and girls (Tung et al., 2016). The onset of symptoms
Wender (1995) of ADHD typically precedes the onset of conduct
Attention-Deficit/Hyperactivity Disorder 325
problems (Danforth, Connor, & Doerfler, 2016), influential and best articulated of such theories is
but it is often the conduct problems displayed by one proposed by Barkley (1997b), which defines
children with ADHD that lead to a great deal of “behavioral inhibition” as the capacity to inhibit
impairment for the child, leading to multiple dis- motivated behaviors, either prior to their initia-
ciplinary confrontations with parents and teach- tion or once they are initiated, which creates a
ers, school suspensions, and problems in peer delay between an impulse and action. This delay
relations (Waschbush, 2002). In addition, these allows the child to “think through” his or her
conduct problems are often predictive of poor actions and allows the behavior to be self-directed
outcomes in adolescence and young adulthood, and guided by the demands of any given situa-
especially for predicting delinquency and sub- tion. A deficit in this inhibition system would
stance abuse (Fischer, Barkley, Fletcher, & make it difficult for a child to sustain his or her
Smallish, 1993; Mannuzza et al., 1989). attention on a single task, it would make foresight
and planning difficult, and it would make it dif-
Other Comorbidities Another condition that ficult for the child to inhibit impulses for motor
often occurs with ADHD is academic under- movement, thereby accounting for the core
achievement or a learning disability, both of symptoms of the disorder. Barkley also outlines
which are frequently defined as school achieve- how such a deficit could account for many of the
ment below a level predicted by a child’s age and other characteristics typically found in people
intellectual level. Approximately 30% of chil- with ADHD, such as a poor sense of time, poor
dren with ADHD show such learning problems emotional self-control, deficits in problem-
(Frick et al., 1991; Massetti et al., 2008). In addi- solving, and an inability to modulate behavior
tion, children with ADHD tend to show a high based on changing situational demands.
rate of conflict with peers (Mikami & Hinshaw, While there is a growing consensus that a defi-
2003), with parents (Johnston & Mash, 2001) cit in the inhibitory control of behavior may be a
and with teachers (Cunningham & Boyle, 2002). primary or at least important deficit in children in
And not surprisingly, given the amount of diffi- ADHD, it is less clear what could cause this defi-
culty and conflict the child with ADHD often cit to develop. Many theories focus on structural
experiences in his or her environment, children neurological abnormalities in parts of the ner-
with ADHD often show high rates of anxiety, vous system involved in inhibitory control of
depression, and low self-esteem (Tannock, 2000) behavior (Castellanos et al., 2002). Other theo-
that persist throughout childhood and into ado- ries focus on abnormalities in the functioning of
lescence (Fischer et al., 1993). ADHD is also a these neurological regions with studies examin-
strong risk factor for the development of sub- ing the cerebral blood flow of children and adults
stance use disorders (Lee, Humphreys, Flory, with ADHD consistently showing areas of
Liu, & Glass, 2011) decreased activity in the prefrontal regions of the
brain (Hendren, De Backer, & Pandina, 2000).
There is also evidence that these neurological
Conceptual Model abnormalities can result from a number of differ-
ent influences. Specifically, there is evidence that
There have been numerous theories of ADHD ADHD symptoms are highly heritable, and, as a
that differ in terms of identifying the “core defi- result, these neurological abnormalities may be
cit” that underlies the symptoms of ADHD and inherited (Waldman & Gizer, 2006). In addition,
the etiological factors that lead to this deficit. A the neurological abnormalities could result from
growing number of researchers in this area have trauma to the developing nervous system, such as
begun to focus on a failure in a child’s inhibition through prenatal exposure to alcohol or other
system that influences his or her ability to regu- drugs, birth trauma, or exposure to environmental
late attention, actions, and emotions (e.g., Nigg, toxins (e.g., lead; Smith et al., 2007). It is impor-
2006; Whalen & Henker, 1998). One of the more tant to note that, although most theories of ADHD
326 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
emphasize potential neurological underpinnings Table 17.4 The nature of ADHD and implications for
to the disorder, there is currently no neurological assessment
test that has proven to be useful in diagnosing Focus of research Implications for assessment
ADHD. Instead, the diagnosis relies on a careful Key factors Assess for presence of two
assessment of the behaviorally based diagnostic differentiate normal core dimensions of
behavior from the behavior: inattention-
criteria using a process outlined in the next sec- disorders and there are disorganization and
tion of this chapter. Although social experiences several different impulsivity-overactivity
can influence the development of behavioral inhi- presentations Assess for presentations that
bition and can change brain functioning (Ciccheti differ on which of the two
core dimensions
& Walker, 2001), most theories have not empha- prodominate
sized environmental factors as primary causal Assess duration to determine
agents in the development of the core symptoms if behaviors are chronic and
of ADHD. Instead, most theories emphasize the stable and have been
evidence since at least age
role of environmental factors in determining how 12
the core deficit is expressed. For example, a child Assess situational variability
with problems in behavioral inhibition will be of behaviors
very difficult to socialize, but some parents will be Assess level of impairment
associated with symptoms
better than others in working with such a child to Presence of multiple Assess for presence of
develop compensatory strategies to minimize the comorbidities and conduct problems/
effects that the inhibitory deficit may have on the co-occurring problems aggression
child’s academic and psychosocial functioning. in adjustment Assess for presence of
learning problems
As a result, environmental factors can play a Assess anxiety and
large role in the development of some of the prob- depression
lems (e.g., poor school performance, social rejec- Assess self-esteem
Assess substance use
tion, conduct problems) that can develop
Assess social relationships
secondarily to the ADHD symptoms (Frick, 1994; and peer social status
Johnston & Mash, 2001). Therefore, although Assess level of parent-child
psychosocial influences may not be integral to and teacher-child conflict
many causal theories of ADHD, it is still impor- Potential alternative Assess for injuries, illnesses,
causes and medications
tant to carefully assess a child’s psychosocial con- Assess for intellectual and
text to (1) determine the degree to which the learning deficits
problems associated with ADHD have negatively Assess for emotional
impacted a child’s functioning and (2) guide difficulties
interventions designed to reduce or prevent many
of the secondary characteristics and co-occurring child with ADHD-Combined Type and Box 17.2
problems in adjustment that often develop in chil- describing the assessment of a child with ADHD-
dren with ADHD (Pelham et al., 2005). Predominantly Inattentive Type.
Box 17.1 Case Study: Evaluation of an rated at T-scores of 72 (98th percentile) and
8-Year-Old Boy with Attention-Deficit/ 67 (95th percentile), respectively. On struc-
Hyperactivity Disorder-Combined Type tured interviews, Justin’s mother also
reported significant problems with impul-
Justin was 8 years, 2 months old when his
sivity and hyperactivity. She reported that
mother requested a comprehensive psycho-
Justin moves around a lot, is often fidgety
logical evaluation from an outpatient men-
or restless, often has trouble staying in his
tal health clinic based on the
seat, often climbs or runs around when he
recommendation of Justin’s pediatrician.
isn’t supposed to, makes more noise than
Justin reportedly had been impulsive and
other children when he is playing, often
hyperactive throughout his life. This behav-
interrupts others, often blurts out answers,
ior has led to him having poor grades in
and often has trouble waiting his turn.
most subjects. He has reportedly become
Again, these problems appear to be more
aggressive with others recently, and he was
severe than would be expected for boys his
asked to leave three different after-school
age. On the BASC-3, Justin’s teacher rated
programs for hitting both classmates and
him on the Hyperactivity subscale at a
teachers. Further, he had received three
T-score of 87 (99th percentile), and his
detentions from school over the 3 months
mother’s ratings reached a T-score of 81
prior to the evaluation for not obeying
(99th percentile). Further, Justin’s mother
classroom rules, largely leaving his seat
reported that Justin has had problems with
without permission, and talking without
impulsivity and hyperactivity his whole
permission.
life, and he began having problems with
inattention when he was 4 years old. These
Assessment of Core Symptoms
problems appear to be causing significant
The core symptoms of ADHD were problems for Justin, as they have led to his
assessed through a structured diagnostic receiving several detentions at school, he
interview conducted with Justin’s mother has been receiving poor school grades that
(DISC; Shaffer, Fisher, Lucas, Dulcan, & appear largely due to his not turning in
Schwab-Stone, 2000) and behavior rating work, and he was asked to leave several
scales completed by his mother and his after-school programs due to conflicts with
teacher (BASC-3; Reynolds & Kamphaus, peers. Thus, Justin shows significant prob-
2015). On the structured clinical interview, lems with inattention, impulsivity, and
his mother reported that Justin often has hyperactivity that appear to be more severe
difficulty concentrating or keeping his than what would be expected for boys his
mind on what he is doing, often dislikes age, that have been displayed for much of
doing things where he has to pay attention his life, and that seem to be causing signifi-
for a long time, is often disorganized, often cant impairment at school and with peers.
has trouble finishing things, often loses As a result, a diagnosis of Attention-Deficit/
things, often forgets what he was supposed Hyperactivity Disorder, Combined
to be doing, often makes careless mistakes, Presentation (Severe) seems to be war-
and often doesn’t listen when people are ranted at the present time.
speaking to him. These problems appear to
be more severe than would be expected for Assessment of Comorbidities
children his age as indicated by teacher and
parent ratings on the Attention Problems Like many children with ADHD-
subscale of the BASC-3, where he was Combined Type, Justin also exhibited sig-
328 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
grade, and her teacher attributed this poor inattentive all her life, but it wasn’t until
performance primarily to problems in con- middle school when it started to signifi-
centration. At home, Claire’s mother also cantly affect her grades. In further support
reported that she had difficulty completing of the impairing nature of these problems,
things, was often daydreaming, and seemed she received a T-score of 71 (96th percen-
to have little motivation for anything. tile) on the Learning Problems subscale of
the BASC-3, as rated by her teacher. Also,
Assessment of Core Symptoms her mother reported that the problems with
schoolwork has led Claire to be somewhat
The core ADHD behaviors were nervous and anxious when required to per-
assessed by structured interviews form in school. In summary, Claire appears
(DISC-IV) completed with Claire and her to show a number of signs of inattention
mother and by rating scales completed by and disorganization that (a) are more severe
Claire, her parents, and her teacher (BASC- than would be typical for children Claire’s
3). From this information, it was clear that age, (b) have been present for much of
Claire shows signs of inattention and disor- Claire’s life, and (c) appear to be negatively
ganization. On the C-DISC, her mother influencing Claire’s academic achievement
indicated that Claire dislikes doing things and causing distress for her about her aca-
that require her to pay attention for a long demic performance. Thus, Claire currently
time, finds it hard to keep her mind on what meets criteria for a diagnosis of Attention-
she is doing when other things are going Deficit/Hyperactivity Disorder,
on, is disorganized, has trouble finishing Predominately Inattentive Presentation,
things, loses things she needs, makes a lot Moderate.
of careless mistakes, and often doesn’t
seem to be listening when people are Assessment of Comorbidities
speaking to her. Additionally, in an unstruc-
tured interview, her English/language arts On parent structured interviews, Claire
teacher reported that Claire has trouble was reported to get very nervous before
focusing, is often inaccurate and makes tests. However, her anxiety seemed to be
careless mistakes on her work, frequently confined to taking tests. Further, ratings by
misses details, and is easily distracted by Claire, her parents, and her teacher on the
noise or movement. Further, Claire’s inat- BASC-3 did not indicate anxiety outside of
tention seems to be more severe than would an age normative range (T-scores ranging
be expected for children Claire’s age at from 46 (Claire) to 59 (her mother)). Thus,
school, as indicated by a T-score of 75 it appeared that Claire’s anxiety seems to
(99th percentile) on the Attention Problems be largely a sign of the distress her atten-
subscale of the teacher-reported BASC-3. tional problems were causing for her at
However, Claire’s attentional problems at school. Also, there were no indications
home do not appear to be as severe at from any assessment source that Claire
school, with ratings on the Attentional exhibited significant conduct problems.
Problems subscale not reaching a clinical Finally, while Claire did not have a lot of
level with T-scores of 62 (87th percentile) friends, she did have “one or two close
and 53 (65th percentile) reported by her friends” according to both Claire and her
mother and father, respectively. According mother.
to her mother, Claire has been somewhat
330 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
Box 17.3 Commonly Used Behavior Rating Scales with Subscales Related to ADHD: An Illustration
of Item Heterogeneity
Many of the more commonly used parent and teacher rating scales have very heterogeneous
item content with respect to ADHD. Most of the scales that purport to measure constructs
related to ADHD contain a large number of behaviors not considered to be part of the core
symptoms of ADHD (see Table 17.1). To illustrate this, we list the item content of relevant sub-
scales from a few commonly used behavior ratings scales and highlight in boldface type the
items that are not directly tied to the two core dimensions of ADHD.
Attention-Deficit/Hyperactivity Disorder 331
--------------------------------------------------------------------------------
--------------------------------------------------------------------------------------------------
Behavior Assessment System for Children, 3rd Edition-ages 6 to 11 (Reynolds & Kamphaus, 2015)
------------------------------------------------------------------------------------------------------
Conners Rating Scales, 3rd edition (Conners, 2008)
Hyperactivity/impulsivity Hyperactivity/impulsivity
(14 items) (18 items)
----------------------------------------------------------------- -----------------------------------------------------------
Fidgeting Leaves seat when he/she should stay seated
Blurts out answers before the question has been completed Gets overly excited
Is constantly moving Fidgets or squirms in seat
Excitable, impulsive Restless or overactive
Gets over stimulated Blurts out answers before the question has
Acts as if driven by a motor been completed
Blurts out the first thing that comes to mind Excitable, impulsive
Has difficulty waiting for his/her turn Acts as if driven by a motor
Runs or climbs when he/she is not supposed to Runs or climbs when he/she is not supposed to
Is noisy and loud when playing our using free time Talks out of turn
Leaves seat when he/she should stay seated Interrupts others (e.g., butts into conversations
Fidgets or squirms in seat or games)
Restless or overactive Is noisy and loud when playing or using free
Interrupts others (e.g, butts into conversations time
or games) Gets over stimulated or “wound up”
------------------------------------------------------------------- Talks too much
7% not part of core ADHD symptoms Is constantly moving
Gets up and moves around during lessons
Has difficulty waiting for his/her turn
Talks non stop
-----------------------------------------------------
11% not part of core ADHD symptoms
---------------------------------------------------------------------------------------------------------
Attention-Deficit/Hyperactivity Disorder 333
Second, many measures simply do not have distinguishing the different ADHD presentations.
sufficient coverage of the core ADHD behaviors Typically, structured interviews that are tied to
to aid in the differentiation of the two subtypes of diagnostic classification systems and are updated
ADHD. Often behaviors indicative of inattention- as the classification system is updated have the
disorganization are intermixed with impulsive and best symptom coverage. Structured interviews
overactive behaviors, providing no method of also allow one to determine the duration and sta-
determining subtypes. For example, from Box bility of ADHD behaviors, which as discussed
17.3 one notices that the ASEBA (Achenbach, previously, is crucial in the assessment of
2001) Attention Problems scale includes items ADHD. There are several ratings scales that have
associated with both inattention-disorganization been designed solely to assess behaviors associ-
and impulsivity-motor hyperactivity. As a result, ated with ADHD and provide much more exten-
use of this overall clinical scale does not aid in sive coverage of behaviors associated with this
334 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
disorder. They include the ADHD Rating Scale widely found and accepted gender ratio and artifi-
(DuPaul, Power, Anastopoulos, & Reid, 1998) and cially equates the number of girls and boys with
the Disruptive Behavior Disorders Rating Scale significant levels of ADHD behaviors. These
(Pelham et al., 2005), both of which were designed norms will lead to more girls and fewer boys
to provide a complete coverage of the diagnostic being considered to have significant ADHD
criteria for ADHD. While they were designed to behaviors than if cross-gender norms are used.
assess the DSM-IV criteria, the similarity in crite- Fifth, researchers studying children with
ria between this version and the DSM-5 results in ADHD and attempting to define the core deficit
their continued ability to assess the diagnostic cri- that may underlie this disorder (e.g., response
teria. The Brown Attention-Deficit Disorder inhibition, sustained attention) have frequently
Scales for Children (Brown, 2001) and the used laboratory measures of inattention and
Attention-Deficit Disorder Evaluation Scales impulsivity in their assessments. These measures
(ADDES-3; McCarney, 2004) also provide exten- place children in standardized conditions and
sive coverage of ADHD symptoms, although they attempt to quantitatively measure their inatten-
were not explicitly tied to DSM-5 criteria. tiveness, impulsiveness, or related behaviors
Third, when selecting measures to assess the under these conditions. For example, one of the
core features of ADHD, one must obtain infor- most frequently used laboratory measures for the
mation from multiple sources (parents and teach- assessment of ADHD is the continuous perfor-
ers). This use of different informants helps one mance test (CPT). There are many variations of
determine the situational variability of behaviors. the CPT, but a prototypical CPT is the one devel-
Also, if the core features of ADHD can be oped by Gordon (1983). In the Gordon CPT, a
assessed through multiple modalities (e.g., rating child is told to view a screen on which numbers
scales and behavioral observations), this negates are presented. The child is instructed to press a
the need to rely on any single imperfect assess- button each time a predetermined number is pre-
ment instrument. Therefore, one must choose a sented. Two responses are measured. Omissions
set of assessment instruments that provides a are the number of times the designated number is
multi-informant and multi-method assessment of presented to the child, and the child fails to press
the core ADHD behaviors. the button. The number of omissions is considered
Fourth, one should have information that to be a measure of sustained attention, especially
allows comparison with age norms. Most defini- increases in omissions over time. Commissions
tions of ADHD either implicitly or explicitly state occur when the child incorrectly presses the but-
that the symptoms should be inconsistent with a ton when the designated number is not presented
child’s developmental level (American Psychiatric and are considered a measure of impulsivity.
Association, 2013). Therefore, an assessment There are many variations of the CPT (e.g.,
must provide information that allows one to com- Conners, 1995), with some CPT tasks presenting
pare a child’s behaviors to the behaviors of other auditory rather than visual stimuli and other tasks
children of a similar developmental level. presenting distracters (e.g., numbers flashing on
Typically, behavior rating scales are best suited either side of the target stimuli that need to be
for this task because of their extensive normative ignored). Also, in addition to the CPT, there are
base. However, one important caution is in order many other laboratory measures that have been
in using norm-referenced rating scales. Many used in research with children who have
scales (e.g., ASEBA) often only provide norm- ADHD. Rapport, Chung, Shore, Denney, and
referenced scores broken down by age and gen- Isaacs (2000) review 142 studies using over 40
der. Definitions of ADHD do not make the different laboratory measures to assess character-
restriction that the behaviors be inconsistent for a istics associated with ADHD. They identify sev-
child’s gender. In fact, it is well accepted that boys eral characteristics of the tasks that most
are two to four times more likely to show ADHD consistently differentiate children with ADHD
than girls (American Psychiatric Association, from other children. First, the tasks that most con-
2013). Using gender-specific norms ignores this sistently show group differences rely on recogni-
Attention-Deficit/Hyperactivity Disorder 335
tion, recall, or both (e.g., the letter or word that a or emotional disturbances show similar problems
child is supposed to look for is not continuously in performance. Without such information, it is
displayed for the child), and they often involve difficult to use these measures to make differen-
some speed of processing component. Also, most tial diagnoses between ADHD and other forms of
of the tasks that reliably differentiate children psychopathology.
with ADHD from other children place special Finally, scores from the laboratory measures
demands on a child’s working memory, and each generally show low correlations with behavioral
of the tasks is experimentally paced, not allowing observations and parent and teacher reports (e.g.,
the child to control the speed at which stimuli are interviews and rating scales) of ADHD symptoms
presented. Rapport et al. (2000) provide an inter- (Barkley, 1991; DuPaul, Anastopoulos, Shelton,
esting discussion as to how these different task Guevremont, & Metevia, 1992). Furthermore,
parameters may provide clues to the specific defi- there is very little information as to whether or
cits displayed by children with ADHD. not the laboratory measures add any clinically
It is evident from this rather extensive litera- useful information to these more ecologically
ture that these tasks have been quite useful in valid measures of behavior (Pelham et al., 2005;
studying children with ADHD, even those as Rapport et al., 2000). For example, if a child is
young as age 3 (Ezpeleta & Granero, 2015). reported as showing significant problems with
These tasks also have a number of appealing ADHD symptoms according to parent and teacher
qualities for clinical assessments as well. For reports and based on behavioral observations in
example, they can help bridge the gap between the classroom, interventions for his behavioral
assessment instruments being used in research problems in his natural environment will likely be
that guide our current conceptualizations of the same, irrespective of his performance on the
ADHD and those commonly used in clinical laboratory task. Because treatment decisions are
practice to make the diagnosis (Frick, 2000). largely based on the more ecologically valid mea-
Even more appealing, however, is that these labo- sures, the role of laboratory measures in clinical
ratory tasks provide a potential means of assess- assessments is uncertain at this point. In Box 17.4
ing the symptoms of ADHD that are not based on we summarize Barkley’s (1991) review of labora-
the perceptions of others that could be biased. tory measures of ADHD symptoms, which pro-
Unfortunately, despite this promise for their vides a more in-depth discussion of these
clinical utility, there are a number of limitations measures’ ecological and incremental validity.
in their development that make their clinical use-
fulness in the diagnosis of ADHD somewhat lim-
ited at present (Pelham et al., 2005; Smith et al.,
2007). For example, the primary validation of
these tasks has been to differentiate children with Box 17.4 Research Note: Ecological Validity
ADHD from children without ADHD or to show of Laboratory Measures of Attention and
the effects of stimulant medication on the task Impulsivity
performance of children with ADHD (Hall et al., Barkley (1991) provides a critical discus-
2016; Rapport et al., 2000). Unfortunately, both sion of laboratory measures (LM) used to
types of studies rely on group-level data that are assess attention and impulsivity, two
difficult to translate into findings that are mean- aspects of the core symptoms of
ingful for interpreting an individual child’s score ADHD. The focus of much of Barkley’s
(e.g., How many children with ADHD do not discussion is on reaction time tasks (RTT),
score high on the laboratory measure? What per- continuous performance tasks (CPT), and
centage of treated children show response to the matching familiar figures test (MFFT).
medication on the task?). Furthermore, it is not Although Barkley also discusses analogue
clear whether performance differences on these observations of motor activity, these were
tasks are specific to children with ADHD or discussed previously in Chap. 8.
whether children with other types of behavioral
336 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
co-occurring problems are best assessed through ough developmental and medical history be
multiple informants and using multiple formats. obtained on a child when assessing for
Given the overlap with learning problems, psy- ADHD. When the history is suggestive of the
choeducational testing should also be a part of possible presence of a medical or neurological
most ADHD assessments. Learning difficulties disorder, the child can be referred to a physician
are not reliably assessed through rating scales, for a more comprehensive medical and/or neuro-
interviews, behavioral observations, or projective logical exam. However, most experts on ADHD
testing. As a result, standardized intelligence and feel that a full medical exam need not be a stan-
achievement tests are often required as part of a dard part of a diagnostic assessment of ADHD,
comprehensive evaluation for ADHD. because the diagnosis is primarily based on
behavioral data (Barkley, 1997a; Pelham et al.,
Some omnibus behavior rating scales include 2005).
items that assess the degree of family conflict and The greatest difficulty in ruling out alternative
other aspects of a child’s family context. However, explanations for ADHD symptoms is the prob-
as discussed in Chap. 12, some assessments may lem of determining what is primary and what is
require a more in-depth assessment of specific secondary. Based on the research on comorbidi-
areas of family functioning that are better ties, we know that children with ADHD are at
obtained through methods that specifically focus risk for many other problems in adjustment, most
on the child’s family context. Each aspect of fam- of which can also mask as ADHD (e.g., learning
ily functioning that was highlighted in Chap. 12 disabilities, emotional disorders). The distinction
(parenting styles and behaviors, parenting stress, between primary and secondary is largely a clini-
marital conflict, and parental adjustment) is cal one, based on a complex weighing of various
important in understanding the family context of pieces of assessment information. To summarize
a child with ADHD. our suggestions from Chap. 15, several types of
information may be helpful in making this diffi-
Assessing Potential Alternative Causes One of cult case formulation. Considering the level of
the most difficult aspects of assessing for ADHD impairment associated with different areas of
is ruling out alternative causes for the symptoms. dysfunction (i.e., which problem areas seem to be
Probably the most important pieces of informa- causing the most problems for the child) and the
tion for this purpose have already been discussed. temporal sequencing of problem behaviors (i.e.,
If one has adequately determined that the symp- which problems seem to predate others) can help
toms of ADHD are of sufficient number, severity, in distinguishing primary or secondary areas of
and duration and cause enough impairment to dysfunction. In addition, viewing family history
warrant a diagnosis, most of the alternative expla- data and determining if a child might be at-risk
nations for symptoms can be ruled out as a sole for a certain type of problem given its occurrence
explanation for the behaviors. Many of the alter- in relatives may also aid in making differential
native explanations for ADHD symptoms result diagnoses. Specifically for ADHD, determining
in behaviors similar to ADHD but of lower inten- if a child’s parents or other first-degree relatives
sity and of shorter duration than is typical for had childhood histories of ADHD symptoms can
children with ADHD. provide valuable information in making an
ADHD diagnosis. In contrast, a positive family
Given that some medical and neurological dis- history for bipolar illness might warrant further
orders can manifest in problems of inattention- assessment to determine if a child’s motor rest-
disorganization and impulsivity-hyperactivity, lessness and problems of impulsive control are
and given that such behaviors can be side effects early indicators of an affective disturbance (Smith
of certain medications, it is important that a thor- et al., 2007).
338 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
when examining the same list of conduct prob- from “mild” if the behavior problems are con-
lems across multiple large samples (n = 27,861), fined to one setting (e.g., only at home), to “mod-
the correlations between the two types of behav- erate” if the behaviors are present in two settings,
iors were 0.68 for boys and 0.63 for girls (Burt and “severe” if the behaviors are present in three
et al., 2015). Thus, it is very common for children or more settings (American Psychiatric
with conduct problems to show both aggressive Association, 2013). For CD, severity is deter-
behavior and rule-breaking behavior. mined by the number of symptoms present in
One of the most common ways to differentiate excess of those required to meet criteria for the
between types of CP is the distinction used by the diagnosis (i.e., three) and the level of harm to
Diagnostic and Statistical Manual of Mental others associated with the conduct problems.
Disorders (5th Ed.) (DSM-5; American Specifically, a child is considered to have “mild”
Psychiatric Association, 2013). In its broad cate- CD if he or she shows few, if any, additional
gory of Disruptive, Impulse Control, and Conduct symptoms beyond those required for the diagno-
Disorders, it distinguishes between oppositional sis and the behavior causes only relatively minor
defiant disorder (ODD) and conduct disorder harm to others. A “severe” rating is given if the
(CD). ODD is defined by three types of behav- child shows many symptoms in excess of those
iors: argumentative/defiant behaviors (e.g., defy- required for diagnosis and the conduct problems
ing or not complying with grown-ups’ rules or cause considerable harm to others, such as in the
requests), angry/irritable behaviors (e.g., losing cases of rape or physical cruelty. There is also a
temper), and spiteful/vindictive behaviors. To “moderate” level of severity, which is a level of
distinguish the symptoms of ODD from behav- symptom severity and degree of harm in between
iors shown to some degree in typically develop- mild and severe.
ing children (Frick & Nigg, 2012), a child must
show at least four of the behaviors over the pre-
ceding 6 months with a degree of persistence and Comorbidities
frequency that exceeds what is normative for his
or her age, sex, and culture. Also, the behavior As noted above, ADHD and conduct problems
must lead to substantial impairment for the child frequently co-occur (Waschbush, 2002). In fact,
at home, at school, or with peers. In contrast, CD the vast majority of children who show severe
consists of more severe antisocial and aggressive conduct problems prior to adolescence often also
behavior that involves serious violations of oth- meet criteria for ADHD (Frick & Viding, 2009).
ers’ rights or deviations from major age- Importantly, ADHD usually precedes the devel-
appropriate norms. The children must show at opment of conduct problems (Waschbusch,
least three behaviors from four categories that 2002). In fact, for conduct problems that start in
largely fit within the covert-overt and destructive- childhood, there is a typical developmental pro-
nondestructive dimensions summarized above: gression that starts with the impulsivity and
aggressiveness to people and animals (e.g., bully- hyperactivity associated with ADHD that then
ing, fighting); property destruction (e.g., fire set- progresses into the oppositional and argumenta-
ting, other destruction of property); deceptiveness tive behaviors associated with ODD that then
or theft (e.g., breaking and entering, stealing progresses into the more severe conduct prob-
without confronting victim); and serious rule vio- lems associated with CD (Lahey & Loeber,
lations (e.g., running away from home, being tru- 1994).
ant from school before age 13). Internalizing disorders, such as depression
While ODD and CD are the major diagnoses and anxiety, also co-occur with ODD and CD at
of conduct problems in the DSM, the DSM-5 rates higher than expected by chance
also includes specifiers to designate different lev- (Cunningham & Ollendick, 2010; Nock, Kazdin,
els of severity within these diagnoses. Within the Hiripi, & Kessler, 2006). However, unlike
ODD category, children can range in severity ADHD, the conduct problems most often precede
Oppositional Defiant and Conduct Disorder 341
the onset of depressive and anxiety symptoms more likely to receive school discipline referrals
(Burke, Loeber, Lahey, & Rathouz, 2005; beginning as early as the first grade (Pas,
Hipwell et al., 2011; Nock et al., 2006). This tem- Bradshaw, & Mitchell, 2011; Rusby, Taylor, &
poral pattern has led some to suggest that anxiety Foster, 2007). This can perpetuate an escalating
and depression may often develop as a conse- cycle in which the discipline problems eventually
quence of the problems a child with serious con- lead to suspensions (Tobin & Sugai, 1999) and
duct problems experiences at home, at school, eventually early school dropout, which can have
and with peers (Frick, Lilienfeld, Ellis, Loney, & long-lasting consequences for the person’s
Silverthorn, 1999). However, it also possible that adjustment into adulthood (Odgers et al., 2007).
the co-occurrence between internalizing disor-
ders and ODD/CD reflect common underlying
problems in emotional regulation (Burke, orrelates with Potential Causal
C
Hipwell, & Loeber, 2010). This idea is supported Roles
by the finding that internalizing problems are
most common in children who present with the Most researchers agree that conduct problems are
angry/irritable mood symptoms of ODD (e.g., the result of a complex interaction of multiple
Burke et al., 2010; Rowe, Costello, Angold, causal factors (Frick, 2016), and prior research
Copeland, & Maughan, 2010; Stringaris & has identified a large number of risk factors that
Goodman, 2009). A finding from research that is have been associated with either ODD or CD and
critical to clinical assessments is that children may a play a role in the development and/or
with both conduct problems and depression show maintenance of the behavior problems associated
a high rate of suicidal ideation (Capaldi, 1992). with these disorders (see Dodge & Petit, 2003;
The combination of suicidal ideation, depression, Frick & Viding, 2009; Moffitt, 2018 for reviews).
and poor impulse control that often is present in These risk factors include dispositional risk fac-
children with conduct problems has been associ- tors, such as neurochemical (e.g., low serotonin)
ated with increased risk for suicide (Shaffer, and autonomic (e.g., low resting heart rate) irreg-
Garland, Gould, Fisher, & Trautman, 1988). ularities, neurocognitive deficits (e.g., deficits in
Both longitudinal and cross-sectional studies executive functioning), deficits in social informa-
have documented that ODD and CD constitute a tion processing (e.g., a hostile attributional bias),
significant risk factor for substance use (e.g., temperamental vulnerabilities (e.g., fearlessness;
Stone, Becker, Huber, & Catalano, 2012). The poor emotional regulation), and personality pre-
comorbidity with substance abuse is important dispositions (e.g., callous-unemotional traits;
because, when youth with serious conduct prob- impulsivity). In addition, they include risk factors
lems also abuse substances, they tend to show an in the child’s prenatal (e.g., exposure to toxins),
early onset of substance use and they are more early child care (e.g., poor quality child care;
likely to abuse multiple substances (Lynskey & poor nutrition), family (e.g., harsh and/or incon-
Fergusson, 1995). Although most of the research sistent discipline), peer (e.g., peer rejection, asso-
on the association between conduct problems and ciation with deviant peers), and neighborhood
substance abuse prior to adulthood has been con- (e.g., high levels of exposure to violence)
ducted with adolescents, the association may environments.
begin much earlier in development (Van
Kammen, Loeber, & Stouthamer-Loeber, 1991).
ODD and CD can also lead to a number of ultiple Developmental Pathways
M
school-related problems. Approximately 20–25% to Serious Conduct Problems
of children with serious conduct problems under-
achieve in school relative to a level predicted by Thus, from this research, it is clear that there are
their age and intellectual abilities (Frick et al., many risk factors for serious conduct problems
1991). In addition, children with ODD or CD are that are important to consider when conducting
342 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
clinical assessments. Unfortunately, there has not The most widely accepted model for delineat-
been much agreement as to a good framework for ing distinct pathways in the development of seri-
organizing these many variables into clear causal ous conduct problems distinguishes between
theories that explain how serious conduct prob- childhood-onset and adolescent-onset subtypes
lems develop. However, there are a few points of of CD (Moffitt, 2018). That is, the DSM-5
agreement (Frick, 2016). (American Psychiatric Association, 2013) makes
the distinction in the criteria for CD between
• First, to adequately explain the development youth who begin showing the symptoms of the
of ODD and CD, causal models must consider disorder before age 10 (i.e., childhood onset) and
the potential role of multiple risk factors and those who do not show serious behavior prob-
not focus on any single causal variable. lems before age 10 (i.e., adolescent onset). This
• Second, causal models must consider the pos- distinction is supported by a substantial amount
sibility that subgroups of antisocial youth may of research documenting important differences
differ in the mechanisms leading to their con- between these two groups of children with seri-
duct problems. ous behavior problems (see Frick & Viding,
• Third, causal models need to integrate research 2009; Moffitt, 2018, for reviews). Specifically, as
on the development of conduct problems with noted earlier, children in the childhood-onset
research on normally developing youth. group are highly likely to start with behaviors
associated with ADHD and ODD early in devel-
To illustrate this third point, research has sug- opment and eventually go on to develop the more
gested that the ability to adequately regulate severe behaviors associated with CD later in
emotions and behavior and the ability to feel childhood (Lahey & Loeber, 1994). However, it
empathy and guilt toward others seem to play is important to note that the reverse is not true in
important roles in the development of serious that many children with ADHD and/or ODD do
conduct problems in children (Frick & Viding, not go on to develop CD later in childhood.
2009). As a result, understanding the processes However, this developmental trajectory is quite
involved in the normal development of these different from youth who show an adolescent
abilities is critical for recognizing how they onset to their CD, who are much less likely to
may go awry in some children and place them at show ADHD or behavior problems prior to onset
risk for acting in an aggressive or antisocial of puberty. In addition, the childhood-onset group
manner. shows more severe conduct problems in child-
Thus, current conceptualizations of ODD and hood and adolescence and are more likely to con-
CD recognize that there are likely multiple causal tinue to show antisocial and criminal behavior
pathways leading to these behavioral problems, into adulthood (Moffitt, 2018). For example, in a
with each pathway involving multiple interacting New Zealand birth cohort followed from birth
risk factors. Further, these risk factors disrupt into adulthood, the rate of official convictions for
critical developmental processes that make a violent acts in adulthood (prior to age 32) was
child more likely to act in an antisocial and 32.7% for males who began showing serious con-
aggressive manner. Consideration of these path- duct problems prior to adolescence, 10.2% for
ways could be quite critical to clinical assess- males who began showing serious conduct prob-
ments because they could explain some of the lems in adolescence, and 0.4% for males who did
variations in the type and severity of conduct not show serious conduct problems in either
problems, the co-occurring problems, and the childhood or adolescence (Odgers et al., 2008).
multiple risk factors displayed by children with In addition to these differences in life course
ODD/CD. However, most importantly, these between childhood- and adolescent-onset CD,
pathways could be critical for guiding individual- most of the dispositional (e.g., temperamental
ized approaches to treating children with these risk, low intelligence) and contextual (e.g., fam-
disorders (Frick, 2012). ily dysfunction) correlates of ODD and CD are
Oppositional Defiant and Conduct Disorder 343
more strongly associated with the childhood- child and problems in his or rearing environment.
onset subtype. This led Moffitt (2018) to propose However, there may be several different types of
that youth in the childhood-onset group develop vulnerabilities within the childhood-onset group
CP through a transactional process involving a that reflect different developmental consequences
difficult and vulnerable child (e.g. impulsive, of this interaction. In the DSM-5, another speci-
with verbal deficits, with a difficult temperament) fier for CD was added that is labeled “With
who experiences an inadequate rearing environ- Limited Prosocial Emotions” and is applied to
ment (e.g. poor parent supervision, poor quality persons who meet the criteria for CD and who
schools). This dysfunctional-transactional pro- show a significant level of callous-unemotional
cess disrupts the child’s socialization, leading to (CU) traits. According to the DSM-5, significant
poor social relations with persons both inside levels of CU traits are defined as two or more of
(i.e., parents and siblings) and outside (i.e., peers the following characteristics that are shown by
and teachers) the family, which further disrupts the child persistently (12 months or longer) and
the child’s socialization. These disruptions lead in multiple relationships or settings:
to enduring vulnerabilities that can negatively
affect the child’s psychosocial adjustment across • Lack of remorse or guilt
multiple developmental stages (see also Frick & • Callous lack of empathy
Viding, 2009). In contrast, Moffitt (2018) pro- • Lack of concerned about performance in
posed that youth on the adolescent-onset path- important activities, and
way showed an exaggeration of the normative • Shallow or deficient affect
developmental process of identity formation that
takes place in adolescence. Engagement in anti- Although only a minority of children and ado-
social and delinquent behaviors is viewed as a lescents with CD show elevated rates of CU traits
misguided attempt to obtain a subjective sense of (Kahn, Frick, Youngstrom, Findling, &
maturity and adult status in a way that is mal- Youngstrom, 2012), research suggests that sepa-
adaptive (e.g., breaking societal norms) but rating this group from other youth with CD is
encouraged by an antisocial peer group. Given critical for designating important potential differ-
that their behavior is viewed as an exaggeration ences in the severity and stability of the child’s
of a process specific to adolescent development behavioral problems, for documenting potential
and not due to enduring vulnerabilities, their con- differences in the causes of the child’s behavioral
duct problems are less likely to persist beyond problems, and for designing individualized treat-
adolescence. However, they may still have ment for the child with CD (Frick, 2016).
impairments that persist into adulthood due to the Specifically, Frick, Ray, Thornton, and Kahn
consequences of their behavioral problems (e.g., (2014) reviewed over 200 published studies and
a criminal record, dropping out of school, sub- found relatively consistent support for the poten-
stance abuse; Moffitt, 2018). tial role of CU traits in designating a subgroup of
This distinction between childhood- and youth who show serious antisocial behavior that
adolescent-onset CD was first recognized in the is more stable, aggressive, and less responsive to
DSM-IV (America Psychiatric Association, typical treatments. Further, Frick et al. also
1994) and even earlier in research on juvenile reviewed research suggesting that these traits
delinquency (Frick & Viding, 2009). More classify a group of youth with serious conduct
recently, there has been increasing recognition of problems who show very different genetic, cog-
another distinction that could be important for nitive, emotional, and social characteristics.
designating important developmental pathways These findings seem to suggest that the causes of
to serious conduct problems. As noted above, the behavioral problems are different in those
childhood-onset CD is considered to be indica- with and without elevated levels of CU traits.
tive of an enduring vulnerability that results from This research led by Frick et al. proposed that
an interaction between dispositional traits in the children with serious conduct problems and ele-
344 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
vated CU traits, but not other children with seri- perament characterized by strong emotional reac-
ous conduct problems, have a temperament (i.e., tivity combined with inadequate socializing
fearless, insensitive to punishment, low respon- experiences that combine to make it difficult for
siveness to cues of distress in others) that can them to develop the skills needed to adequately
interfere with the normal development of con- regulate their emotional reactivity (Frick &
science and place the child at risk for a particu- Morris, 2004). These emotional regulation prob-
larly severe and aggressive pattern of antisocial lems can result in these children committing
behavior. In contrast, children and adolescents impulsive and unplanned aggressive and antiso-
with childhood-onset antisocial behavior with cial acts, for which they may feel remorseful
normative levels of CU traits do not typically afterward, but may still have difficulty control-
show problems in empathy and guilt. In fact, they ling in the future.
appear to be highly reactive to emotional cues in
others and are highly distressed by the effects of
their behavior on others (Frick & Viding, 2009). Implications for Assessment
Thus, the conduct problems in this group do not
seem to be easily explained by deficits in con- This body of psychological research forms the
science development. Further, they display higher basis for designing an appropriate assessment for
levels of emotional reactivity to provocation from children with conduct problems. In Table 17.5 we
others (Frick & Morris, 2004), and their behav- summarize the critical areas of research and their
ioral problems are strongly associated with hos- relevance to the assessment process. As is evident
tile/coercive parenting (Frick et al., 2014). As a from Table 17.5, assessment of conduct problems
result, children in this group seem to show a tem- shares several important characteristics with the
Table 17.5 Key research findings and their implications for the clinical assessment of children with oppositional defi-
ant disorder and conduct disorder
Research findings Implications for assessment
1. Core symptoms and subtypes: ODD and CD 1a. Assess a wide range of conduct problems
designate a very heterogeneous group of 1b. Assess the level of impairment associated with the child’s
children who can vary widely in the type and behavior in multiple situations and settings (e.g., home, school,
severity of their behavior problems work, interpersonal relationships)
1c. Assess the amount of harm to others that is caused by the
behavior
2. Common comorbidities: ODD and CD are 2a. Assess for the presence of ADHD
often accompanied by several comorbid types 2b. Assess for the presence of anxiety and depression (including
of problems that can influence the course and suicidal ideation)
treatment of children with serious conduct 2c. Assess for substance use and abuse
problems 2d. Assess for problems in child’s academic performance
3. Correlates with potential causal roles: A 3a. Assess a child’s intellectual level
number of different risk factors within the child 3b. Assess critical temperamental and personality risk factors
and his or her psychosocial environment can that have been associated with ODD and CD, such as problems
play role in the development and maintenance regulating emotions, rebelliousness, fearfulness, and response to
of the child’s behavior problems and may be punishment
important targets for treatment of children and 3c. Assess a child’s peer interactions, social status, and
adolescents with ODD or CD associations with a deviant peer group
3d. Assess a child’s family environment, especially the type of
parenting used
3e. Assess critical aspects of child’s broader social ecology (e.g.,
economic disadvantage, exposure to violence)
4. Multiple developmental pathways: ODD and 4a. Assess the developmental history of the child’s conduct
CD develop through multiple different problem behavior to determine if the child behavior problems
pathways each involving distinct started early in childhood or whether their onset coincided with
developmental mechanisms that can influence adolescence
the stability of the child’s behavior and his or 4b. Assess for presence of elevated callous-unemotional traits
her response to treatment
Oppositional Defiant and Conduct Disorder 345
mother also reported significant problems ing seems to have worked” and punishing
with impulsivity and hyperactivity. She her just “leads to more problems” because
reported that Charlie moves around a lot, is of her temper tantrums. It also appeared
often fidgety or restless, often has trouble that Charlie’s impulsivity led to a lot of her
staying in her seat, often climbs or runs behavioral problems because she acts with-
around when she isn’t supposed to, makes out thinking and will do things impulsively.
more noise than other children when she is Finally, Charlie’s estimated level of intelli-
playing, often interrupts others, often blurts gence on the WISC-5 was 97 or well within
out answers, and often has trouble waiting an age normative range. On the WJ-IV, her
his turn. Again, these problems appeared to Broad Reading and Broad Writing com-
be more severe than would be expected for posites showed an age standard score of 95
girls her age. On the BASC-3, Charlie’s (36th percentile) and 97 (43rd percentile),
teacher’s ratings on the Hyperactivity sub- respectively. These scores indicated that
scale reached a T-score of 87 (99th percen- she performed at a level similar to other
tile), and her mother’s ratings reached a children her age and at a level predicted by
T-score of 81 (99th percentile). Further, her estimated intellectual level in reading
Charlie’s mother reported that Charlie has and writing. However, her Broad Math
had problems with impulsivity and hyper- composite was an age standard score of 87
activity her whole life and began having (19th percentile), indicating that she per-
problems with inattention when she was formed slightly below average compared to
4 years old. These problems appeared to be other children her age in math. While this
causing significant problems for Charlie at result suggested that Charlie is somewhat
school, where much of her discipline prob- weaker in math, potentially due to her
lems were related to leaving her seat and ADHD, this score was not low enough to
talking out of turn. Thus, Charlie showed be considered a learning disorder.
significant problems with inattention,
impulsivity, and hyperactivity that appeared Assessment of Multiple Developmental
to be more severe than what would be Pathways
expected for her age, that have been dis-
played for much of Charlie’s life, and that Charlie showed a typical pattern of early
seemed to be causing significant impair- emerging conduct problems in which the
ment at school and with peers. As a result, inattention, impulsivity, and hyperactivity
a diagnosis of Attention-Deficit/ related to ADHD emerged early in child-
Hyperactivity Disorder, Combined hood, followed by the oppositional and
Presentation (Severe) was given. angry behaviors associated with
ODD. Thus, without intervention, she
Assessment of Correlates with Potential appears to be at risk for developing more
Causal Roles severe behavioral problems associated with
childhood-onset CD. Much of Charlie’s
Charlie lived alone with her mother, behavior problems appear to be due to
who worked full-time outside of the home impulsivity and getting angry when she
cleaning houses. Her parents divorced doesn’t get her way. According to her
when Charlie was 4 years old, and she has mother, Charlie seems to feel bad when she
had minimal contact with her father since “calms down” and that she typically is a
then. Her mother reported having no idea very caring child who is sensitive to others
of how to discipline Charlie because “noth- feelings.
Oppositional Defiant and Conduct Disorder 347
Box 17.6 Case Study: A 13-Year-Old Boy with his age. Of note, the ratings of his conduct
Behavior Problems problems by his teacher were not outside of
Hunter was 13 years, 8 months old when he an age normative range (i.e., T-scores of 52
was referred for a comprehensive psycho- and 59 on the Aggression and Conduct
logical evaluation by his mother, who was Problems subscales, respectively).
concerned about his serious behavioral However, Hunter was on medication dur-
problems and limited social relationships. ing the school day, and he was attending a
She reported that Hunter had been having structured residential school for children
serious behavioral problems since at least with behavior problems, all of which may
age 8. He was expelled from school in the have influenced his teacher’s ratings.
seventh grade due to fighting and inappro-
priate sexual behavior (i.e., unwanted Assessment of Comorbidities
touching of girls) and was enrolled in a
residential school for children with behav- As is typical for children who show
ioral problems at the time of the evaluation. childhood-onset conduct problems, Hunter
He had been hospitalized on an inpatient also showed a history of behaviors associ-
psychiatric unit twice for aggression, with ated with ADHD. On both structured and
the latest time for threatening to kill his unstructured interviews, his mother
younger sister for not sharing her toys. reported that Hunter often had trouble
keeping his mind on what he is doing, dis-
Assessment of Core Features and Subtypes liked and tried to get out of work that
required sustained attention, was disorga-
On both unstructured and structured nized, had trouble finishing important
(DISC-IV) interviews, both Hunter and his tasks, often lost important items, was for-
mother reported a long history of serious getful (e.g., forgets to turn in homework),
behavioral problems. Specifically, Hunter made careless mistakes, often did not listen
reportedly lies, steals (e.g., family mem- when people were speaking to him, and
ber’s jewelry, unauthorized use of parents’ often did not finish tasks before moving on
credit card), and deliberately destroys oth- to the next one. Further, Hunter’s mother
ers’ property. In addition, Hunter report- reported that he often was on the go, was
edly often bullies and threatens others (e.g., restless, had difficulty staying in his seat,
posted online threats toward teacher excessively talked and made noises, and
online), initiates physical fights, and is interrupted other people. Importantly, he
physically cruel to people. Hunter’s aggres- also engaged in impulsive, inappropriate
sion and behavioral problems appeared to sexual behavior including repeatedly
be more severe than other children his age, watching sexually explicit material on
as indicated by T-scores on the Aggression school computers and touching female
subscale of the BASC-3 at 64 (92nd per- classmates. Hunter’s level of inattention
centile) and 66 (93rd percentile) as rated by was more severe than would be expected
his mother and father, respectively, and on for children his age, as indicated by
the Conduct Problems subscale with T-scores on the Attention Problems sub-
T-scores at 74 (97th percentile) and 70 scale of the BASC-3 at 66 (92nd percen-
(95th percentile), as rated by his mother tile) and 66 (92nd percentile) as rated by
and father, respectively. These elevations his mother and stepfather, respectively.
indicate that Hunter’s behavior problems Hunter’s hyperactivity was more severe
are well above what is typical for children than other children his age, as indicated by
348 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
(Frick & Nigg, 2012). Therefore, it is necessary makes them difficult to capture through behav-
to determine whether or not the youth is exhibit- ioral observations.
ing levels of conduct problems that are atypical in
type and frequency for his or her age. In addition, Assessment of Comorbidities Co-occurring
it is important to assess the degree of impairment problems can be assessed conjointly with the
that is associated with the youth’s conduct prob- assessment of the conduct problems themselves,
lems. As noted above, youth can range greatly in through the use of omnibus rating scales, struc-
the severity of their behavior problems, ranging tured interviews, or multi-domain observational
from youth who show oppositional and defiant systems. Given the association between conduct
behaviors only at home to youth who show severe problems and learning disabilities, a psychoedu-
aggression that results in substantial harm to oth- cational evaluation that includes a standardized
ers in the community. Assessing the severity of intelligence test and academic achievement
the youth’s behavior not only can determine screener should also be a part of most evaluations
whether or not the youth needs treatment but also of children and adolescents with conduct
the treatment intensity and the most appropriate problems.
treatment setting.
The primary methods of assessing the number, Assessment of Correlates with Potential
types, and severity of the conduct problems and Causal Roles Given that research has clearly
the level of impairment that the conduct prob- suggested that there are typically many different
lems are causing for the child or adolescent are risk factors that play a role in the development of
structured interviews, behavioral rating scales, serious conduct problems, it is important to con-
and behavioral observations, which have each duct a comprehensive assessment of a child or
been reviewed in other chapters of this text. Many adolescent to screen for a wide range of individ-
structured interviews and behavioral rating scales ual and contextual risk factors that may be
provide good coverage of the conduct problem involved in the development of a youth’s serious
behaviors and allow for multi-informant assess- conduct problems. This is true across the differ-
ments. However, they each offer unique advan- ent developmental pathways that can lead to seri-
tages in other respects. Behavioral rating scales ous conduct problems. In Table 17.6, we include
are more time-efficient and provide some of the a summary of some of the most common risk fac-
best norm-referenced information for determin- tors associated with each of the developmental
ing whether the level and severity of conduct pathways.
problems are more severe than is typical for a Many of these risk factors can be assessed
child’s age. In contrast, diagnostic interviews through unstructured interviews with the parent,
often provide important information on the teacher, or youth. Also, various risk factors can
degree of impairment associated with the con- be assessed through some of the same behavior
duct problems, and they provide a structured rating scales that are used to assess the level and
means of assessing the age of onset of the severity of a child’s conduct problems. For exam-
problem behaviors. Behavioral observations in a ple, the BASC-3 assesses attention problems and
child’s natural setting can make a unique contri- hyperactivity that would likely be common in
bution to the assessment process by providing an youth with childhood-onset CD, as well as anxi-
assessment of a child’s behavior that is not fil- ety and depression, which are common in those
tered through the perceptions of an informant and children with CD who do not show the Limited
by providing an assessment of the immediate Prosocial Emotions specifier. One risk factor that
environmental context of a child’s behavior. deserves particular attention in the assessment of
Unfortunately, for older children and adoles- serious conduct problems is parenting practices.
cents, many of the common conduct problems The importance of parenting to the assessment
are by nature covert (e.g., lying and stealing) or processes is due to two main reasons. First, as
only occur infrequently (e.g., fighting), which noted in Table 17.6, parenting plays a critical role
350 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
Table 17.6 Individual risk and contextual risk factors to consider when assessing for ODD and CD
Developmental pathway Individual risk factors Contextual risk factors
Adolescent onset Rebelliousness Association with deviant peers
Authority conflict Poor parental monitoring of child’s
behavior
Childhood onset Low intelligence Family conflict and instability
ADHD Harsh and inconsistent parenting
Problems regulating emotions (e.g., anger, anxiety, Peer rejection
and depression) Neighborhood disorganization
Low self-esteem Exposure to violence
Limited prosocial ADHD Insensitivity to punishment Low parental warmth
emotions Fearlessness/thrill and adventure seeking Poor attachment between parent
Planned aggression for gain (i.e., instrumental and child
aggression)
activities and a failure to experience many types Another important issue when assessing age
of prosocial emotions (e.g., guilt and empathy). of onset of a child’s CP relates to the accuracy of
Further, assessing the level and severity of parent and youth reports. Three research findings
aggressive behavior, especially the presence of can help in interpreting such reports. First, the
instrumental aggression, would be critical for longer the time frame involved in the retrospec-
children and adolescents with elevated CU traits tive report (e.g., a parent of a 17-year-old report-
(Frick et al., 2014). ing on preschool behavior vs. a parent of a
As most clinicians recognize, people do not 6-year-old reporting on preschool behavior), the
often fall neatly into the prototypes that are sug- less accurate the report is likely to be (Green,
gested by research. Therefore, these descriptions Loeber, & Lahey, 1991). Second, although paren-
are meant to serve as hypotheses around which to tal report of the exact age of onset may not be
organize an assessment based on the available very reliable over time, typical variations in years
research. They also highlight several specific are usually small, and the relative rankings within
important pieces of information that are needed symptoms (e.g., which symptom began first) and
when assessing children and adolescents with within a sample (e.g., which children exhibited
conduct problems. One of the most critical pieces the earliest onset of behavior) seem to be fairly
of information in guiding assessment, and per- stable (Green et al., 1991). Therefore, these
haps ultimately intervention, is determining the reports should be viewed as rough estimates of
age at which various conduct problems began. the timing of onset and not as exact dating
This information provides some indication as to procedures. Third, there is evidence that combin-
whether or not the youth may be on the childhood- ing informants (e.g., such as a parent or youth) or
onset pathway. combining sources of information (e.g., self-
Both unstructured and structured clinical report and school/clinical/police records), and
interviews provide some best ways of obtaining taking the earliest reported age of onset from any
this information on the onset of a youth’s serious source, lead to an estimate that shows somewhat
conduct problems. However, one complicating greater validity than any single source of infor-
factor is that research has not been consistent mation alone (Lahey et al., 1999).
about the exact age at which to make this distinc- In addition to age of onset of the child’s con-
tion or even whether this distinction should be duct problems, the presence of elevated levels of
based on chronological age or on the pubertal sta- CU traits is also important for treatment plan-
tus of the child (Frick & Nigg, 2012). For exam- ning. As noted above, the DSM-5 included in the
ple, the DSM-5 criteria for CD (American diagnosis of CD a specifier called “With Limited
Psychiatric Association, 2013) makes the distinc- Prosocial Emotions,” and the criteria for this
tion between children who begin showing severe specifier provides guidance for the clinical
CP before age 10 (i.e., childhood onset) and those assessment of CU traits. These symptoms were
who do not show severe CP until age 10 or older selected for inclusion in the specifier, and the
(i.e., adolescent onset). However, other research- diagnostic cutoff used to designate elevated lev-
ers have used age 11 (Robins, 1966) or age 14 els of these traits (i.e., two or more symptoms)
(Patterson & Yoerger, 1993; Tibbetts & Piquero, was chosen based on extensive secondary data
1999) to define adolescent onset. Thus, onset of analyses across several large samples of youth in
severe CP before age 10 seems to be clearly con- different countries (Kimonis et al., 2015).
sidered childhood onset and onset after age 13 Specifically, these four symptoms (lack of empa-
clearly adolescent onset. However, classifying thy and guilt; unconcerned about performance in
children whose severe CP began between the important activities; and shallow or deficient
ages of 11 and 13 is less clear and is likely depen- affect) consistently were the best indicators of the
dent on the level of physical, cognitive, and social overall construct of CU traits in factor analyses
maturity of the child. across samples, and the presence of two or more
352 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
In this chapter, we discuss two specific applica- 1. Externalizing behaviors are the most common
tions of the assessment procedures and techniques reason for referral to child mental health
reviewed in previous chapters. We discuss the clinics.
assessment of two related types of childhood psy- 2. Based on research on ADHD:
chopathology: Attention-Deficit Hyperactivity (a) Assessments should include a multi-
Disorder and conduct problems, which form the informant and multisource assessment of
diagnoses of Oppositional Defiant Disorder the core ADHD behaviors: inattention-
(ODD) and Conduct Disorder (CD). To continue disorganization and impulsivity-
our basic premise that clinical assessments should hyperactivity. Assessment of these core
be guided by basic psychological research, we features must be placed within a develop-
provide a brief overview of some of the more mental perspective to determine if the
important research findings with particular rele- child’s behavior is more severe and
vance to the assessment process for these com- impairing that would be expected for his
mon disorders of childhood. For all of these or her age.
disorders, assessments should be structured (b) Assessments should screen for the pres-
around our current knowledge of the core features ence of the most common co-occurring
of each domain. In addition, the most frequent co- problems that may accompany ADHD:
occurring problems associated with both types of conduct problems/aggression, emotional
psychopathology should be routinely assessed, disturbance, low self-esteem, problematic
because these comorbidities often have important social relationships, learning difficulties,
prognostic and treatment implications. and family conflict.
For ADHD, a difficult part of the assessment (c) Assessments should rule out alternative
is ruling out other medical or psychological dis- causes for the core symptoms: medical/
orders that could solely account for the ADHD neurological disorders, mental handicaps,
symptoms. Also, in assessing ADHD, one must learning disorders, and adjustment reac-
be knowledgeable of educational laws related to tions to environmental stressors.
legally mandated services for children with (d) Because ADHD often has a major impact
ADHD, so one can work with educators in on a child’s or adolescent’s school func-
designing a treatment plan for the child or adoles- tioning, the assessment should be con-
cent with ADHD. ducted in collaboration with school
For conduct problems, the myriad of potential personnel and with a knowledge of local
causal factors should be assessed to determine educational statutes relevant to services
which ones may be operating for a given child for students with ADHD.
and which ones should therefore be a focus of 3. Based on research on severe conduct prob-
intervention. Also, different causal factors may lems in children and adolescents:
be involved in the various subgroups of children (a) Assessments should provide a multi-
with conduct problems. Understanding the differ- source and multi-method assessment of
ent pathways through which children and adoles- conduct problems including determining
cents develop serious conduct problems can be the types and severity of conduct prob-
critical for designing assessments and interpret- lems and how much harm they cause
ing the information provided by the evaluation. It others.
is especially important to distinguish between (b) Assessments should screen for the most
childhood and adolescent-onset conduct prob- common co-occurring problems that
lems and to determine when serious conduct often accompany conduct problems:
problems are also accompanied by elevated lev- ADHD, anxiety, depression, substance
els of callous-unemotional (CU) traits. use, and learning disabilities.
354 17 Assessment of Attention-Deficit/Hyperactivity and Disruptive Behavior Disorders
(c) Assessments should assess known corre- designating unique pathways to the devel-
lates of conduct problems that could play opment of conduct problems.
a role in causing or maintaining the prob- (e) Applying the research on the most com-
lem behavior and therefore should be a mon developmental pathways to serious
major focus of intervention. conduct problems can help organize the
(d) The age at which the serious conduct
assessment results into a clear description
problems began and the presence of of potential causes of the child’s behavior
callous-unemotional traits should be problems and help to individualize treat-
assessed because of their importance in ment for the child.
Assessment of Depression
and Anxiety 18
may be present (see Klein, Dougherty, & Olino, 3. Significant weight loss or failure to make
2005). Still, research has supported a two-factor expected weight gains
model of internalizing problems, one indicative 4. Frequent insomnia or hypersomnia
of depression and one that can be conceptualized 5. Motor agitation or retardation
as anxiety (Anderson & Hope, 2008; Burgess & 6. Frequent fatigue or lose of energy
Nakamura, 2014). Klein and colleagues (2005) 7. Feelings of worthlessness or guilt
provide a clear review of the ways in which the 8. Impaired concentration
symptoms of either problem can be mutually 9. Recurrent thoughts of death, suicidal ideation,
influential. It is beyond the scope of this chapter or suicide attempt
to consider all of the evidence for a common dis-
order for anxiety and depression. However, it is For many years, the need for diagnostic crite-
important for clinical assessors to be aware that, ria for childhood depression was not clear, as
despite the fact that we see anxiety and depres- researchers debated whether or not the disorder
sion as if they differ in many respects (including even existed in childhood. There is increasing
etiology and course), there is substantial comor- evidence that children and adolescents display
bidity between the two, and many children and much of the same symptomatology as adults
adolescent show signs of both (Anderson & (e.g., cognitive distortions) but with some clear
Hope, 2008; Lamers et al., 2011). differences as well (e.g., vegetative symptoms
The present discussion is influenced primarily such as insomnia being more common in adoles-
by the structure of the DSM-5, which focuses on cents; difficulty concentrating being more com-
anxiety and depressive disorders as separate enti- mon in adults; Rice et al., 2019). It is now clear
ties, although co-occurrence is recognized with that depression exists in childhood and that its
the “with anxious distress” specifier for major course is more persistent and more impairing
depressive disorder (American Psychiatric than thought previously. Although there are some
Association, 2013). Ultimately, it will be up to developmental influences on the manifestation of
the clinician to determine, through the evidence depression (see below), most evidence points to
gathered, if interventions targeting both anxiety the continuity in the signs of depression from
and depression are warranted. The important childhood to adulthood (Stark et al., 2006).
reminder is that both areas of concern should be Despite some of the previous controversy
assessed as either primary or secondary influ- regarding the existence of childhood depression,
ences on the child’s functioning. clinicians are fairly reliable in their diagnosis of
the condition. When using structured diagnostic
interviews such as those described in Chap. 11,
Childhood Depression the interrater reliability of the diagnosis ranges
from fair to excellent (see Klein et al., 2005).
Diagnostic Criteria Emerging agreement regarding the existence of
the depression syndrome in childhood, the wide-
To meet the most accepted diagnostic criteria for spread application of the DSM criteria, and evi-
major depressive disorder, at least five of nine dence of interrater reliability have all contributed
symptoms must be present for at least a 2-week to forming a consensus that childhood depression
period, and one of the symptoms must be either is an important public health problem that war-
depressed mood or loss of interest or pleasure rants significant research attention.
(American Psychiatric Association, 2013). Regardless of the accuracy of diagnosis, diag-
These symptoms include: nostic decisions for children and adolescents are
not always clear. In fact, a meta-analysis indi-
1 . Depressed or irritable mood cated that “recurrent screening of depressive
2. Diminished interest or pleasure in daily
symptoms does not improve detection of MDD
activities but that cognitive and physical symptoms (such
Childhood Depression 357
as sleeping problems) may be better predictors” appetite disturbances and loss of energy as
(Stockings et al., 2015, p. 460). Consequently, a aspects of their depressive presentation but less
clinical assessor should develop some strategies likely to exhibit anhedonia or concentration prob-
and/or cues (e.g., change in sleep patterns) that lems (Rice et al., 2019). Adolescents are more
trigger further consideration of depression during likely than young children to experience feelings
an evaluation. Strategies for assessing for depres- of hopelessness as part of depression (Weiss &
sion can be gleaned from an understanding of the Garber, 2003).
nature of the syndrome, including its risk factors Depression is also distinctive as a syndrome
and course. Some methods for identifying the because of the high rate of family resemblance
need to assess for the presence of depression are that has been identified. Heritability estimates
discussed next. vary widely, but it appears that children with a
parent with depression are three to four times
more likely to experience depression them-
Characteristics of Childhood selves (Maughan, Collishaw, & Stringaris,
Depression 2013). In addition, family history of depression
is a significant predictor of recurrence of
Although prevalence rates are often debated, depression for children and adolescents (Klein
research has generally concluded that less than et al., 2005). Therefore, knowledge of family
3% of preadolescent children experience depres- history is often helpful for conceptualizing
sion but that rates of depression increase substan- cases. Family history may affect assessment
tially in adolescence to around 12–13% (Center and intervention planning for a child who has
for Behavioral Health Statistics and Quality, several symptoms of depression but does not
2017), although a larger proportion of youth suf- meet diagnostic criteria. If this child’s mother
fer from at least some symptoms of depression at has a history of depression, more assessment of
some point (see Hankin, Wetter, & Cheely, 2008 mood difficulties and intervention would be
for review). A prevalence rate this large warrants warranted than if no family resemblance was
vigilance for the existence of depression in youth, present. This information indicates that the psy-
particularly adolescents, referred for chologist should routinely and carefully gather
assessments. family history data regarding a history of
Research consistently indicates that the preva- depression for the child suspected of having an
lence of childhood depression may also be age- internalizing disorder.
related, with symptoms as well as the disorder The duration of a depressive disorder can be
itself, emerging more in older youth. A higher substantial, and youth depression can be consid-
rate of depression may be expected in adoles- ered a chronic condition (Maughan et al., 2013).
cence because of the attendant stressors associ- Research has shown that episodes of major
ated with this age range (Cicchetti & Rogosch, depression tend to average around 10 months in
2002; Hankin, Mermelstein, & Roesch, 2007). 12–15-year-olds (Sund, Larsson, & Wichstrøm,
Adolescents are presented with increasing devel- 2011). As discussed elsewhere in this text, the
opmental demands and stressors, their interpreta- developmental onset and course are critical areas
tion of which may be important for their to assess for understanding the manifestation of
emotional functioning (Midgley et al., 2017). psychological problems, the associated features,
Several studies have suggested a few develop- and the potential outcomes. This issue is illus-
mental changes in the expression of depression. trated quite well for youth depression in a study
Although children and adolescents both report by Hammen, Brennan, Keenan-Miller, and Herr
dysphoria as a central symptom of depression, (2008). They considered the outcomes at age 20
younger children may experience episodes of for individuals with an early onset of depression
shorter duration (Klein et al., 2005). Adolescents (i.e., by age 15) versus a late onset (i.e., after age
are more likely than adults to exhibit sleep and 15), with an additional consideration of whether
358 18 Assessment of Depression and Anxiety
or not the depressive symptoms were recurrent those described in this text, reliably differentiate
by age 20. In general, the early onset, recurrent between the two syndromes.
group had the worst outcomes in many psycho- Also of interest is the comorbidity of depres-
logical and social domains of functioning. In par- sion and disruptive behavior disorders, specifi-
ticular, these individuals: cally ADHD, ODD, and CD (Costello, Mustillo,
Erkanli, Keeler, & Angold, 2003). The pattern of
• Were more likely than individuals in the other comorbidity also appears somewhat different
groups to have had comorbid anxiety by age based on developmental level. More specifically,
20 younger children with depression are likely to
• Were more likely to have comorbid eating dis- have comorbid anxiety disorders, whereas ado-
orders and externalizing problems than indi- lescents with depression are more likely to have
viduals with nonrecurrent depressive comorbid substance abuse problems or eating
symptoms disorders (see Hankin et al., 2008).
• Were more likely to report not having a close The comorbidity of youth depression also
friend and not being involved in a romantic extends into the realms of academic and social
relationship and report having problems in functioning. In short, youth internalizing prob-
family relationships lems, including depression, are related to poorer
• Were less likely to have a history of close academic performance (Owens, M., Stevenson,
friendships and good family relationships Hadwin, & Norgate, 2012), peer problems, and
(Mondimore et al., 2006) school refusal (LeBlanc, Sautter, & Dore, 2006).
Furthermore, some research suggests that one
These findings suggest that early onset of depres- should often screen for depression when a medi-
sion, and particularly recurrent depression during cal illness is present (Benton, Blume, Crits-
adolescence, have a significant impact on other Cristoph, Dubé, & Evans, 2013). Although a
areas of functioning. strong case can be made for depression being
Thus, knowledge of the developmental mani- either the cause or the effect in these relations,
festations of depression, recognition of heteroge- the important issue in assessment is to approach
neity of depressive presentations (Klein et al., referrals for depression comprehensively so as to
2005), and comprehensiveness in the assessment determine all of the relevant targets of interven-
of depressive symptoms and associated features tion. The research on childhood depression
are essential for providing appropriate and mean- clearly indicates a need to assess for symptom-
ingful recommendations for treatment. atology and difficulties in a variety of domains,
as well as to assess indicators of impairment
related to depressive symptoms (e.g., truancy).
Comorbidity Issues of comorbidity can also be confused
with the notions of primary and secondary condi-
When a child is diagnosed with depression, there tions. The primacy of the condition can be espe-
is a high probability that the child will also meet cially important for internalizing disorders, as
criteria for another disorder, especially an anxi- they often have to be treated simultaneously
ety problem (Melton, Croarkin, Strawn, & because of considerable comorbidity. It may also
Mcclintock, 2016). This overlap is striking, but it be important to differentially intervene based on
does not indicate that depression and anxiety knowledge of primacy. A child with both depres-
comprise the same syndrome. Indeed, whereas sion and anxiety may require intervention for
most youth with a diagnosis of depression have a depression first, for example, if the child is pre-
history of anxiety, only about 25% of children senting risk factors for suicidal behavior.
and adolescents with history of anxiety would Relatedly, a child suffering from anxiety and
qualify for a diagnosis of depression (Melton depression may require intervention for anxiety
et al., 2016). In addition, many measures, such as symptoms initially if compulsive hand washing
Childhood Depression 359
Reliability The reliability of the overall CDI 5. Good discriminant validity based on initial
score is good with internal consistency of the research
total score reported at 0.91 for the total standard-
ization sample. Subscale internal consistency The CDI-2, however, has some weaknesses that
coefficients ranged from 0.73 to 0.77. Test-retest should be noted:
coefficients for 2–4 weeks between administra-
tions were good at 0.89 for the total score and 1. The representativeness of the norm, particu-
0.76–0.89 for subscales (see Kovacs, 2011). larly in regard to race/ethnicity, is question-
able, particularly in light of changing
Validity Results of confirmatory factor analyses demographics in the United States
are reported in the CDI-2 manual and indicate 2. Unclear validity and utility for youth with
good fit for the two broad factors (i.e., emotional cognitive and other developmental disabilities
problems and functional problems) as well as (Bae, 2012)
moderate to high correlations among the sub-
scales (Kovacs, 2011).
Reynolds Adolescent Depression Scale, 2nd
The validity analyses in the development of the Edition (RADS-2; Reynolds, 2002) The
CDI-2 focused largely on discriminant validity, Reynolds Adolescent Depression Scale, 2nd
wherein the ability of the CDI-2 to correctly dis- Edition (RADS-2; Reynolds, 2002) is designed to
tinguish between youth with depression versus assess symptomatology associated with depres-
other difficulties was examined. These analyses sion via self-report in adolescents ages 11–20.
revealed that youth diagnosed with major depres- The RADS-2 is closely based on its predecessor
sion scored higher on the CDI-2 total score and its (Reynolds, 1986), although it may be used with a
subscales than matched controls, youth with gen- slightly wider age range than was recommended
eralized anxiety disorder, ADHD, or oppositional for the original version. The RADS-2 is not
defiant/conduct disorder. Scores on the self-report designed to provide a diagnosis of a specific
CDI-2 were moderately correlated (i.e., 0.28– depressive disorder. However, it is designed for
0.37) with the Beck Depression Inventory-Youth use as a screening measure or for research.
and somewhat higher correlations (i.e., 0.38–0.58)
with the DSM-IV Major Depression subscale Scale content The RADS-2 provides a thorough
from the Conners Behavior Rating Scale. In initial sampling of depressive symptoms that are
analyses, Hispanic youth tended to score lower on included in the DSM system and other nosologies
the overall CDI and Emotional Problems scales and takes approximately 5–10 min to administer.
than White or Black youth. This pattern held for
the parent and teacher forms (Kovacs, 2011). Administration and scoring The RADS-2 uses
30 items to which the client responds with 1 of 4
Strengths and weaknesses The CDI-2, like its choices of frequency: almost never, hardly ever,
predecessor, continues to be a widely used mea- sometimes, and most of the time. Six of the items
sure. Among the CDI-2’s strengths are: are considered critical items. A Total Depression
score and scores on four scales, Dysphoric Mood,
Anhedonia/Negative Affect, Negative Self-
1. A long research history that has contributed to evaluation, and Somatic Complaints are avail-
considerable popularity within the practice able. The items are placed on a single page where
and research communities a template is used to easily sum item scores to
2. Ease of administration and scoring obtain raw scores. Raw scores are then converted
3. Wide range of relevant content for the assess- to percentile ranks based on the total sample and
ment of youth depression by gender and grade, and unlike the previous ver-
4. Good reliability sion of the RADS, T-scores are offered. The
362 18 Assessment of Depression and Anxiety
T-scores are based on the original distributions of Validity Like the earlier RADS, the RADS-2 has
the total score and scale scores in the norm sam- good content validity in that it systematically
ple (i.e., linear T-scores). This change represents assesses many of the symptoms commonly asso-
a significant improvement and brings the ciated with the syndrome. Reynolds reports
RADS-2 in line with the scores available for higher correlations between the RADS-2 and
many other measures of child functioning. A Hamilton Depression Rating Scale Clinical
T-score of 65 or higher is considered clinically Interview than were found for the RADS. In
significant and worthy of close consideration addition, in an independent study, the RADS-2
(along with other assessment data) in case demonstrated good reliability and construct
conceptualization. validity. Specifically, the subscales had strong
inter-item correlations, and the total score and
Norming The RADS-2 was standardized on a Dysphoric Mood and Somatic Complaints sub-
sample of 3300 adolescents with sampling aimed scale scores were moderately related to suicidal
at obtaining a sample representative of the US in ideation, suicidal behaviors, low self-esteem, and
terms of gender and ethnicity at the time. social disengagement. The Anhedonia/Negative
Reliability and validity studies were conducted Affect subscale was moderately related to bore-
on 9052 participants from 7 states and 1 Canadian dom susceptibility. RADS-2 subscale scores
province recruited from schools. The sample is were weakly related to hopelessness (Osman,
described as socioeconomically diverse, with Gutierrez, Bagge, Fang, & Emmerich, 2010).
representation reported by age, gender, and eth- However, additional research suggests that the
nicity. The match of the sample to meaningful RADS-2 would not be a useful or sensitive
national criteria such as census data is not given. enough screener for adolescent suicidality spe-
The author reports increased representativeness cifically (Gutierrez & Osman, 2009).
of Hispanics in the RADS-2 norm sample; how-
ever, Hispanics (i.e., 4.5% of the norm sample)
and African-Americans (7.4% of the norm sam- Strengths and weaknesses The RADS-2 has
ple) both appear to be underrepresented. A clini- many apparent strengths including:
cal sample consisting of 297 participants was
also recruited. This sample was mostly female
(55.6%), with the vast majority of the clinical 1.
A thorough assessment of depressive
sample being Caucasian (86.1%). symptomatology
2. Ease of administration and scoring
Of note, the downward extension of the 3. Good recommendations regarding interpreta-
RADS-2, the Reynolds Child Depression Scale-2 tion in the manual
(Reynolds, 2010) is also available for children
ages 7–13. It has been studied much less exten- Weaknesses include:
sively than the RADS-2 but follows a very similar
approach to evaluating youth depressive symp- 1. More information of the norming sample is pro-
toms. Thus, we do not review it in detail here. vided than was the case for the RADS, but it still
does not appear to be representative of the US
Reliability Reliability indices for the RADS-2 population, particularly in regard to ethnicity.
total score are good, with the internal consistency 2. The length of time since the last update which
coefficients of the total score and scales ranging includes a need to update the norm sample.
from 0.86 to 0.93 in the norming sample 3. The response scale (i.e., almost never, hardly
(Reynolds, 2002). Two-week test-retest reliability ever, sometimes, most of the time) may be dif-
was also good, ranging from 0.77 for Somatic ficult for some adolescent respondents to
Complaints to 0.85 for the Total Depression score. interpret.
Childhood Depression 363
Peer-referenced assessments of depression An Table 18.1 The assessment questions related to child-
alternative approach to self-report rating scales hood depression and implications for assessment
for assessing depression is the use of peer nomi- Assessment question Implications for assessment
nations (discussed further in Chap. 9). For exam- Screening Administer screening measures or
screen during interview. Assess for
ple, Weiss, Harris, and Catron (2002) described
critical symptoms (i.e., suicidality,
the development and initial validation of the psychosis). Determine need for
Peer-Report Measure of Internalizing and further assessment
Externalizing Behavior. The data supported the Classification Assess for presence of symptoms
existence of six independent factors, including that meet DSM-5 criteria. Determine
stability and duration of symptoms
Depression. The Depression factor was signifi- Determine onset
cantly associated with teacher ratings of depres- Assess family history of depression
sion and more strongly related to such ratings Determine functional impairment
than teacher reports in other domains. As dis- and degree of distress
cussed further in Chap. 9, peers may provide Comorbidities Assess for presence of anxiety
disorders, ADHD, LD/intellectual
unique insight into some clinical constructs, disability, eating disorders, or
including depressive symptoms or negative oppositional defiant disorder/
affect, but a clinician must consider the relative conduct disorder. Determine the
costs and benefits of incorporating such a tool influence of depression on school
performance, absenteeism, etc.
into assessment batteries. Assess social relationships and peer
social status
Assess for presence of substance
abuse, particularly in adolescents
n Assessment Strategy
A Alternative Obtain developmental and medical
for Depression causes histories
Rule out persistent depressive
As we have in previous chapters, we recommend disorder and post-traumatic stress
disorder
a five-stage assessment process for the assess-
Rule out medical problems that are
ment of depression: screening, classification, associated with depression
comorbidities, alternative causes, and treatment Treatment Assess for presence of maladaptive
considerations (see Table 18.1). considerations cognitions. Assess for presence of
In light of the substantial epidemiology and chronic stressors. Assess parenting
style, especially level of parental
comorbidity of depression, its adverse effects on involvement with child
development in a variety of domains, and its less- Evaluate response to previous
than-flagrant symptomatology, we recommend interventions
that every child who is seen by a clinician should
be screened for depression. Research findings
support the need for screening all referrals that Any one of these screening methods takes only a
are seen not only by psychologists but also by few minutes. In fact, querying both the child and
other professionals (e.g., pediatricians, teachers). a parent would take only a few minutes. Screening
The screening process can be time-efficient and efforts may allow the clinician to implement
may include: intervention in order to avert considerable impair-
ment and suffering (see Box 18.2). A somewhat
• Administering a brief, self-report inventory to more comprehensive case example is presented
the child who has adequate reading ability in Box 18.3.
• Asking parents whether their child exhibits Classification of a child as depressed requires
symptoms given in the DSM-5 criteria; meeting the DSM-5 criteria, and by this we mean
• Querying teachers, nurses, grandparents, or meeting all criteria based on apparently valid
others about symptoms of depression information. Possessing the necessary symptom-
364 18 Assessment of Depression and Anxiety
Box 18.2 Case Example: A 15-Year-Old Boy opmental disorders, one with a reading dis-
Screened for Depression ability and the other with ADHD.
Greg gave the impression of being well-
Screening for depression is crucial because
adjusted and socially skilled. He was
of the nature of internalizing symptomatol-
impeccably groomed and a good conversa-
ogy. Such screening is time-efficient in most
tionalist. He even asked the examiner ques-
settings. This case example illustrates the
tions about his interests and offered his full
importance and practicality of screening.
cooperation with the assessment process.
Greg was referred for an evaluation
Greg described many successes in his life,
because of his parents’ concerns about his
including being chosen as captain of the
academic progress. He is a 15-year-old
junior varsity baseball team at his school.
sophomore in high school. His parents and
He apparently has an active and successful
teacher suspect that he has an undiagnosed
social life. Aside from athletics, he is
reading disability.
reportedly involved in a variety of other
Greg’s mother reported a long history of
extracurricular activities.
academic problems for Greg, including
The results of the assessment produced
considerable speech delays as a youngster.
clear evidence of below average achieve-
He was reportedly enrolled in speech ther-
ment in reading, even though Greg’s intel-
apy throughout his elementary school years.
lectual functioning was slightly above
He was reportedly often ridiculed by his
average. Following reports of moderate
peers for stuttering. His speech problems
levels of depression on the parent and self-
have reportedly declined somewhat in high
report versions of the BASC-3, Greg was
school. He has, however, had long-standing
asked to complete the Child Depression
problems with reading and spelling.
Inventory-2 (CDI-2)
Greg’s parents expressed some frustra-
Greg acknowledged enough symptoms
tion because they feel that Greg has been
on the CDI to fall into the mild/moderate
passed along from grade to grade because
range of depressive symptoms. He
he is very quiet and never causes any
acknowledged guilt about past failures,
problems.
uncontrollable sadness, insomnia, hope-
He continues to have trouble in most of
lessness, decreased appetite, increased
his classes. He is performing adequately in
fatigue, low self-esteem, concerns about
math, but he complains that he has trouble
his appearance, and occasional thoughts of
taking notes in class and comprehending
hurting himself, among other symptoms.
them after he takes them. He reportedly
The incongruence between the symptoms
failed his English class during the first
endorsed on rating scales and his presenta-
semester of the current academic year. He
tion was striking. During the feedback ses-
is very distraught about these failures
sion, Greg and his parents were apprised of
because of his desire to go to college fol-
these findings and asked to verify their
lowing graduation and his concerns about
validity, which they did. He was then
disappointing his parents. Greg also
referred for follow-up assessment and
reported some other recent stressors in his
intervention for his depression.
life. He recently broke off a 6-month rela-
This case illustrates how time-efficient
tionship with a girlfriend, and he is receiv-
screening, followed by more detailed
ing rehabilitation for a severe back injury
assessment, for depression can be of poten-
that he incurred in a bicycle accident
tial importance. In this scenario, depression
2 months ago. Greg also has two sisters
was not suspected by the referral sources,
who were reportedly diagnosed with devel-
the examining psychologist, or the client.
Childhood Depression 365
Box 18.3 Case Example: A 16-Year-Old Girl Megan reportedly received her first fail-
with Behavior Problems and Depression ing grade in the ninth grade and was moved
from algebra back to pre-algebra, which
Megan, a 16-year-old female, was referred
she had in eighth grade. According to her
for a psychological evaluation by her
mother, her grades continued to deteriorate,
mother due to academic problems.
and currently she is receiving an “F” in
According to Mrs. Johnson (mother),
History and English; a “C” in science, alge-
Megan has trouble concentrating in school,
bra, and Spanish; and an “A” in
and she thought that Megan’s school prob-
photography.
lems may be caused by ADHD.
Megan’s teachers feel that she is smart
Megan lives with her mother and
but that she does not put forth significant
younger brother. Megan’s parents are
effort. They reported that she exhibits
divorced, and she reportedly rarely sees her
behavioral problems in the classroom. She
father. Mrs. Johnson stated that there were
is reportedly constantly talking and dis-
no difficulties during her pregnancy with
rupting the class. Her algebra teacher stated
Megan. During birth, however, Megan was
that she is fond of Megan but that she plays
in fetal distress, and Mrs. Johnson had an
around with her classmates too much. She
emergency cesarean delivery. Megan was
reportedly talks back to her teachers, which
reportedly born full-term and weighed 8
is consistent with the reports from last
pounds, 7 ounces. Megan’s mother
year’s teachers. Megan’s teachers also
described her development as normal in
reported that she frequently sleeps through
that she reached all motor and language
class.
developmental milestones at age-
According to records kept by the school
appropriate times. Megan is reportedly in
for the past 2 years, Megan is often exces-
very good health. She is supposed to wear
sively tardy and absent from class. In the
contact lenses, but her mother reports that
fall of this year, she served 2 days of sus-
Megan rarely wears them.
pension in study hall for being rude to a
According to Mrs. Johnson, Megan’s
new teacher. Subsequent to this incident, a
parents divorced when Megan was 5 years
parent conference at school resulted in
old. In addition, Megan was reportedly
Mrs. Johnson agreeing to have Megan eval-
very close to her maternal aunt who died
uated to rule out ADHD.
earlier this year. Megan stated that her
Megan has reportedly been seen in
mother is constantly on her back and that
counseling at a community clinic for
she has to do too much work around the
2 years. Her current counselor reported that
house. There is a family history of alcohol
Megan remains under a great deal of stress
abuse on both her mother’s and father’s
due to her parents’ divorce, her aunt’s
sides of the family. Megan’s mother and
death, and her continuing academic prob-
father also both have their own reported
lems. Mrs. Johnson mentioned that Megan
history of alcohol abuse.
also often has stomachaches and dizzy
Megan is currently in the 11th grade.
spells, which she attributes to stress. Megan
Megan’s mother noted that Megan did well
complains of not being able to go to sleep;
in elementary school, always receiving
once asleep, she frequently has nightmares
“As” or “Bs.” Starting in the seventh grade,
about dying. She reported that she worries
Megan’s academic performance reportedly
about getting cancer. She told the examiner
deteriorated. Her mother noticed that she
that one time, she felt a bump on her leg
had problems reading aloud and problems
and thought that she had cancer.
with mathematics.
366 18 Assessment of Depression and Anxiety
Mrs. Johnson reported that Megan Several measures were given to assess
drinks alcohol regularly, and that at times, Megan’s behavioral and emotional func-
she drinks quite heavily (i.e., eight to ten tioning. Rating scales were completed by
beers). Megan told the examiner that she Mrs. Johnson, one of Megan’s teachers,
sometimes blacks out on the weekends and Megan. Megan’s mother and teacher
from drinking. She also admitted to fre- completed the ASEBA Child Behavior
quent use of marijuana. Mrs. Johnson also Checklist (CBCL) and Teacher Rating
expressed concerns about some of Megan’s Form (TRF), respectively. The teacher
online relationships and wonders if she rated Megan as having significant attention
might be sexually active. Megan down- problems, delinquent behaviors, and
played these concerns. somatic complaints. Mrs. Johnson did not
A few weeks before the evaluation, indicate any significant problems on the
Megan was detained by police because of CBCL. Despite the reports of attention
skipping school and being intoxicated. In problems, Megan’s teacher stated that she
two interview sessions, Megan described does not believe that Megan has ADHD.
herself as being depressed. She told the Megan completed several self-report
examiner that, for the past 3 weeks, she has measures that were used to determine her
felt angry, sad, and hopeless. Throughout the self-perception. She was given the Piers-
interview Megan appeared to be agitated. Harris Self-Concept Scale (PHCSCS) and
Megan’s intellectual functioning was the ASEBA Youth Self Report (YSR). On
evaluated with the Wechsler Adult the PHCSCS, Megan rated herself high on
Intelligence Scale-4th Edition (WAIS-IV). anxiety, depression, somatization, and
Her overall intelligence score of 115 indi- (external) locus of control. On the YSR,
cates that she is functioning slightly above Megan’s responses were similar. Megan
her same-aged peers. Her Perceptual also endorsed items that indicate problems
Reasoning Index score of 126 meets or with impulse control and compliance.
exceeds 96% of her same-aged peers. Her On the self-report measures, Megan
Verbal Comprehension score of 96 meets or reported some positive qualities for herself.
exceeds 48% of her age mates. She obtained She described herself as has having good
scores of 108 (70th percentile) on both the peer relationships. She reported that all of
Processing Speed and Working Memory her friends like her because she is funny.
indexes. These results indicate that her non- Megan also indicated that her peer relation-
verbal reasoning abilities are better devel- ships and social life at school are the areas in
oped than her abilities in verbal reasoning. her life that are not currently stressful for her.
She had the most difficulty with the As a follow-up measure, Megan was
Vocabulary and Arithmetic subtests. This given the KSADS. The KSADS is a struc-
result is consistent with both her mother’s tured diagnostic interview in which respon-
and her own report of problems with math. dents state yes or no to indicate the presence
Megan was administered the Woodcock- of symptoms relating to DSM diagnoses.
Johnson Psychoeducational Battery Tests Questions relating to major depression and
of Achievement (WJ-IV). Megan’s perfor- anxiety disorders were administered.
mance on all areas of achievement were all Despite feeling that she has some positive
in the Average range, indicating that her qualities, Megan endorsed all of the items
academic achievement is consistent with in the area of major depression with the
what would be expected for her age. Her exception of suicidal ideation. Megan
achievement was also similar to her described herself as being anxious, and she
WAIS-IV Verbal Comprehension score. told the examiner that she often has stom-
Childhood Depression 367
Post-traumatic Stress Disorder (PTSD) (PTSD) (Grills & Ollendick, 2003). These results
as separate classes of disorders from anxiety. should caution the clinician to attend to rea-
However, given the prominence of anxious dis- sons for, or domains of, convergence and
tress in each disorder (e.g., repetitive thoughts divergence among informants.
that evoke stress and significant arousal and/or • The phenomenology of childhood anxiety dis-
avoidance of stimuli related to a trauma, respec- orders appears to be very similar to that of
tively), they are discussed here alongside the pri- adults (Silverman, 1993). However, the very
mary anxiety disorders. Obsessive Compulsive assessment methods used to make diagnoses
Disorder (OCD) consists of persistent, recurrent of childhood anxiety disorders have not been
thoughts and behavioral compulsions associated developed with potential developmental influ-
with impaired functioning. Post-traumatic Stress ences on the manifestation of anxiety in mind
Disorder (PTSD) is based on re-experiencing of a (Silverman & Ollendick, 2005).
traumatic event and increased arousal. • Although anxiety often is apparent in cases of
The measures of anxiety that are often used in depression, anxiety disorders also often occur
child assessments typically screen aspects of all in the absence of depression. Moreover, anxi-
of these problem areas broadly. It is then up to the ety with comorbid problems is associated with
clinician to obtain the necessary information to poorer outcomes than anxiety alone (see
best conceptualize the specific anxiety problem Andover, Izzo, & Kelly, 2011).
or disorder and then make recommendations • Separation Anxiety Disorder typically occurs
accordingly. in children younger than adolescence (Eisen
et al., 2011), whereas social anxiety becomes
more common leading up to and during ado-
haracteristics of Childhood Anxiety
C lescence (Morris & Ale, 2011).
Disorders • Specific phobias, particularly of animals, the
dark, and heights, are also more likely to occur
There are several important issues from research at younger ages than social anxiety which, in
on anxiety in children adolescents that should be turn, occurs at younger ages than agoraphobia
considered in clinical assessments (Moretz & or generalized anxiety (Beesdo et al., 2009).
McKay, 2011; Silverman & Ollendick, 2005): • There are two main developmental courses for
school phobia, or school refusal: sudden or
• In many respects, anxiety disorders can be acute onset in a child who has functioned rea-
conceptualized as exaggerations of normal sonably normally and a gradual increase in
responses to developmental demands (e.g., distress in a child who has had similar prob-
separation anxiety in young children; anxiety lems from an early age and has had ongoing
in social situations for adolescents). It is when difficulties adjusting to the school environ-
the anxiety response is impairing that inter- ment (Tyrrell, 2005).
vention is needed (Silverman & Ollendick, • The average age of onset for childhood
2005). Obsessive Compulsive Disorder (OCD) is
• Evidence regarding the stability and prognosis approximately 9–12 years, with family history
of anxiety disorders in children is quite mixed, of OCD placing a youngster at-risk for an
as childhood anxiety can be transient or quite even younger onset of OCD symptoms (see
chronic (Beesdo et al., 2009). Baldwin and Dadds, 2008 for review).
• Research using the Anxiety Disorders • The central symptoms of Post-traumatic
Interview Schedule for Children (ADIS) has Stress Disorder (PTSD), increased arousal,
found strong parent-child interrater reliability emotional numbing, and re-experiencing the
for individual anxiety disorder diagnostic cat- trauma, occur in children as well as adults
egories (Lyneham, Abbott, & Rapee, 2007) who are diagnosed with the condition
but also substantial parent-child disagreement (American Psychiatric Association, 2013).
370 18 Assessment of Depression and Anxiety
However, the frequency and intensity of these Inconsistent Responding Index by which careless
symptoms may be important indicators of or otherwise invalid response patterns are identi-
impairment for children (Carrion, Weems, fied by contradictory responses to similar items
Ray, & Reiss, 2002), and the outcomes (e.g., (e.g., “I am nervous” and “I get nervous around
externalizing problems, risky behaviors, people”).
depression) associated for children who have
been exposed to trauma are quite varied Administration and scoring Children have sim-
(Sharma-Patel et al., 2011). ply to respond yes or no to RCMAS items that are
read from a one-page response form that is
accompanied by a template for scoring. It takes
Specialized Measures of Anxiety approximately 5–15 min to administer, and a
short form is also available. A high score is indic-
Revised Children’s Manifest Anxiety Scale, ative of higher anxiety.
2nd Edition (RCMAS-2; Reynolds &
Richmond, 2008) The Revised Children’s The Total Anxiety and subscale scores are
Manifest Anxiety Scale-2 (RCMAS-2; Reynolds T-score conversions of raw scores. Separate
& Richmond, 2008) measures the expression of norm-referenced scores are calculated for youth
anxiety symptomatology. We review the ages 6–8, 9–14, and 15–19. Unlike the previous
RCMAS-2 in relative depth because it and its version, the RCMAS-2 offers norm-referenced
predecessor have been widely used and provides scores for combined samples of boys and girls.
a good example of a measure that screens multi-
ple domains of anxiety in a relatively short for- Norming The RCMAS was normed using 2368
mat. The RCMAS-2 also has many unique children aged 6 through 19 years. The sample
features for a single-construct rating scale. was collected from 13 states. Age, gender, race/
ethnicity, parental education level, and geo-
Scale content The RCMAS-2 includes 49 items graphic region are provided in the manual
distributed among four scales: Physiological (Reynolds & Richmond, 2008). The RCMAS-2
Anxiety (12 items), Worry (16), Social Anxiety norm sample is an improvement over that of the
(12 items), and Defensiveness (9 items). The RCMAS in that Hispanic youth are better repre-
content of the subscales appears diverse. Items sented and information concerning parent educa-
from the physiological anxiety subscale, for tion level suggests that there is a reasonable
example, range from having trouble making up inclusion of children from different socioeco-
one’s mind (note that the physiological nature of nomic status levels.
this item is not apparent) to hands feeling sweaty
to waking up scared. Reliability Internal consistency reliability of the
The Defensiveness Scale is a rather unique composite score is good with overall coefficients
feature of a single-construct measure. The in the low 0.90s for the overall sample and across
RCMAS-2 Defensiveness Scale measures chil- age and gender subgroups. Subscale reliabilities
dren’s tendency to portray themselves in a favor- are somewhat lower but still adequate. The inter-
able light (e.g., “I am always kind”). The items nal consistency coefficient for the Physiological
on this scale are likely to be transparent to many Anxiety subscale was 0.75. All others were 0.79
adolescents. Research has shown that younger and higher. One-week test-retest reliability coef-
children tended to score higher on the Lie ficients ranged from 0.64 (Social Anxiety) to
(Defensiveness) scale of the earlier version of the 0.76 (Composite Score).
RCMAS Lie scale than older youth (Pina,
Silverman, Saavedra, & Weems, 2001). No Validity Considerable validity evidence is pro-
research has investigated this issue with the vided in the RCMAS-2 manual. Factor studies of
RCMAS-2. The RCMAS-2 also has an the standardization data do not suggest that
Anxiety Disorders of Childhood 371
RCMAS-2 scores are captured by a single factor. Additional self-report measures In addition to
The manual reports correlations of RCMAS-2 the RCMAS-2, there is a vast array of rating
scale scores with parent and teacher ratings on scales for child anxiety (see reviews by Silverman
the Conners Rating Scale and self-reports on the & Ollendick, 2005; Tulbure, Szentagotai,
Child Depression Inventory Short Form. As Dobrean, & David, 2012). A few measures that
might be expected cross-informant correlations focus on social anxiety include:
involving RCMAS-2 scales and analogous sub-
scales on other measures were moderate (i.e.,
0.20s–0.40s). Within informant correlations • Leibowitz Social Anxiety Scale for Children
between the RCMAS-2 and CDI Short Form and Adolescents (LSAS-CA; Masia-Warner
ranged from 0.45 to 0.68. et al., 2003)
• Social Phobia and Anxiety Scale for Children
An independent study (Lowe, 2014) demon- (SPAI-C; Beidel, Turner, & Morris, 1995)
strated confirmatory support for a five-factor • Social Anxiety Scale for Adolescents (SAS-A;
model (i.e., Physiological Anxiety, Worry, Social LaGreca & Lopez, 1998)
Anxiety, and two defensiveness factors) consis- • Social Phobia Inventory (SPIN; Connor et al.,
tent with that reported by Reynolds and Richmond 2000)
(2008). Group measurement invariance (e.g., for
males and females) was further suggestive of Two additional self-report ratings of anxiety
construct validity (Lowe, 2014). Ang, Lowe, and focus on aspects of the individual’s cognitions as
Yusof (2011) also found support for the conver- they contribute to, or be a sign of, marked anxi-
gent (i.e., high correlations with the Anxiety sub- ety. The Anxiety Control Questionnaire for
scale of the BASC-2, moderate correlations with Children (ACQ-C; Weems, Silverman, Rapee,
a measure of academic stress) and divergent (i.e., and Pina, 2003) contains 30 items that assess the
low, negative correlation with a measure of nar- child’s perceived control over threatening events
cissism) validity of the RCMAS-2 in a large sam- or stimuli. The Children’s Automatic Thoughts
ple of children from Singapore. (Schniering & Rapee, 2002) assesses whether or
not the respondent has experienced particular
Strengths and weaknesses Strengths of the negative thoughts about anxiety-provoking situa-
RCMAS-2 include: tions. An extension of this measure includes posi-
tive automatic thoughts (Hogendoorn et al.,
1.
A broad sampling of anxiety 2010). Research on the psychometrics of these
symptomatology scales has had promising results (see Hogendoorn
2. Ease of administration and scoring et al., 2010, 2014; Schniering & Lyneham, 2007;
3. A larger and more geographically diverse Silverman & Ollendick, 2005), and the unique
norming sample than is typical for single- focus of these scales also appears to translate
construct measures well to cognitive-behavioral interventions for
4. A reliable Total Anxiety score childhood anxiety.
5. The presence of a validity (defensiveness and
inconsistent responding) scales Interviews and clinician ratings The typical
way for anxiety to be assessed in children is
Weaknesses of the RCMAS-2 include: through interviews (including some of the diag-
nostic interviews reviewed in Chap. 11) and on
1. The yes/no format provides limited informa- subscales within broadband ratings (usually com-
tion about the frequency/severity of the symp- pleted by parents, teachers, or youths them-
toms it measures. selves). However, the Anxiety Disorders
2. Some content in the Physiological Anxiety Interview Schedule for Children (ADIS-C;
subscale appears to align with other constructs Silverman & Albano, 1996) is a unique struc-
(e.g., negative affect, depression). tured interview in that it provides an in-depth
372 18 Assessment of Depression and Anxiety
assessment of symptoms and impairment across clinical use. Observational methods potentially
anxiety disorder categories. Although it has not serve the clinician, even in experimental form, by
been updated for DSM-5, yet, the ADIS-C is still providing corroborating data for test-based find-
commonly used in research on clinical assess- ings and by providing insights that may be useful
ment of childhood anxiety. for intervention design. Still, the incremental
validity obtained by incorporating such proce-
The Yale-Brown Obsessive Compulsive Scale dures into assessment batteries for childhood
for Children (Y-BOCS; Goodman et al., 1989) is anxiety is uncertain (Silverman & Ollendick,
a clinician rating that specifically focused on 2005).
symptoms of obsessive compulsive disorder. It The preceding discussion in no way provides
has demonstrated good psychometric properties an exhaustive description of the available meth-
for children as young as age 5 (Freeman, Flessner, ods for assessing anxiety in children. Instead, we
& Garcia, 2011). It is very quick and time- have attempted to highlight some measures that
efficient (ten items) and thus provides a very are well-known, widely used, and/or unique in
quick screening for these types of symptoms. The their implementation or focus. The reader is
adult version of this scale is well-known and encouraged to remain up-to-date on the burgeon-
extensively used in research and practice ing body of research on evidence-based assess-
ment of anxiety (as well as other forms of
Cognitive-behavioral assess- psychopathology) to determine which approaches
ment Mychailyszyn and Whitehead (2016), possess clinical utility.
among others, emphasize the importance of cog-
nitive processes as central in the development
and expression of anxiety symptoms. Thus, the An Assessment Strategy for Anxiety
clinician should be prepared to assess for these
factors, which include negative self-statements, We again recommend a five-stage assessment
negative expectancies, evidence in support/not in process for the assessment of anxiety: screening,
support of the client’s expectations (Mychailyszyn classification, comorbidities, alternative causes,
& Whitehead, 2016). Of course, this type of and treatment considerations (see Table 18.2).
questioning may be more difficult for young chil- Screening for anxiety can be accomplished
dren who may have relatively little insight or ver- using both informal and formal means. A few
bal reasoning relative to adolescents or adults. As questions about anxiety symptoms may be more
noted above, there are several rating scales that appropriate for many settings than the use of self-
have been used to assess cognitions that have report or other more formal psychometric devices.
been related to anxiety in children. However, The reader should note, however, that this advice
cognitive influences on the experience and is based on practical experience rather than on a
expression of anxiety will often be assessed research base. As noted elsewhere in this text,
through clinical interviews, self-monitoring/jour- many omnibus rating scales provide a screening
naling on anxiety-provoking situations, and of anxiety, but they do not differentiate among
informant descriptions. different types or classes of anxiety problems.
Measures such as the RCMAS-S and domain-
In addition, numerous other behavioral obser- specific anxiety measures such as those men-
vation methods, including parent-child interac- tioned above provide a more in-depth assessment
tions and social interaction scenarios, have been of anxiety symptoms in a relatively efficient man-
used in the assessment of youth anxiety (see ner. Nevertheless, the most important screening
Silverman & Ollendick, 2005). principle to remember is to do it, because failure
These informal methods of assessment vary in to screen for anxiety may result in under-serving
the degree to which they have been developed for children with less obvious problems.
Anxiety Disorders of Childhood 373
Table 18.2 The assessment questions related to child- elicit or alleviate the anxiety. Attention to the
hood anxiety and implications for assessment
content of different measures of measures is
Assessment important for differential diagnosis and treatment
question Implications for assessment
recommendations.
Screening Administer time efficient rating
scales or ask specific questions The differentiation of separation anxiety dis-
about anxiety order, various phobias, adjustment disorders, and
Determine need for further other syndromes requires careful study of the
assessment DSM as well as considerable experience. Specific
Classification Assess for presence of symptoms
differential diagnoses across anxiety disorders
and subtypes that meet may DSM-5 criteria
through specific rating scales or and with other internalizing problems, such as
interviews. Determine stability and depression, are critical for informing treatment
duration of symptoms strategies. Moreover, the nature of the symptom-
Determine onset
atology may vary over time. A child may, for
Comorbidities Assess for presence of depression,
academic difficulties, attention example, initially be diagnosed with separation
problems, externalizing behaviors, anxiety disorder and not have school difficulties.
or other anxiety problems School refusal and subsequent academic prob-
Determine the influence of anxiety lems, however, may appear for this child at a
on school performance,
absenteeism, etc. later time.
Assess social relationships and peer Medical problems should be ruled out as alter-
social status native causes because of the role that physical ill-
Alternative Obtain developmental and medical ness and medical procedures can play in initiating
causes histories
or exacerbating a child’s anxiety symptoms. In
Rule out stressful life events or
other psychopathology as primary fact, one of the first hypotheses to consider when
explanation for anxiety symptoms somatic symptoms are present (e.g., headaches,
Rule out medical problems that are stomachaches, diarrhea) is the possibility that the
associated with anxiety
child is physically ill. Thus, although somatic
Treatment Assess for presence of maladaptive
considerations cognitions complaints are frequently present in children
Assess parental depression and with anxiety disorders, a medical evaluation
parenting style, especially should be conducted when they are evident to
overprotection rule out a clear physical illness.
Evaluate response to previous
interventions Further assessment on the factors that may
influence treatment outcome is also important.
Research on this area of assessment is limited,
Classification of anxiety difficulties is compli- but an approach that incorporates behavioral
cated by debate regarding the appropriateness of observation, rating scales, and/or self-monitoring
existing criteria for youth. DSM-5 attempted to is recommended (Silverman & Ollendick, 2005).
better account for developmental influences in Maladaptive cognitions, family stressors, and
the manifestation of anxiety, yet these influences traumatic events are but some of the variables
are not necessarily reflected in other classifica- that may play a role in the client’s manifestation
tion systems or reflected in the assessment tools and maintenance of anxiety symptoms and that
that professionals use (Mohr & Schneider, 2013). will undoubtedly play a role in the progress of
These issues notwithstanding, it is critical for treatment. See Box 18.4 for a case example of a
treatment planning purposes for the clinician to child with co-occurring anxiety and learning
determine the phenomenology of the client’s problems.
anxiety and the situations/stimuli that tend to
374 18 Assessment of Depression and Anxiety
Box 18.4 Case Report: A 9-Year-Old Boy with responded impulsively. Several times, he
Learning Problems and Anxiety remarked, “I’m an idiot.” He was resched-
uled to complete testing a week later.
Mark is a 9-year-old male who was referred
During the second assessment session,
for and assessment due to academic diffi-
Mark cooperated fully. He again pleaded
culties subsequent to being transferred to a
with his mother to join him, but she refused.
new school. He was referred in March of
He did not display any of the previous
the school year. His grades are reportedly
behavioral problems, and he did not make
lower in his new school than in the previ-
self-deprecating statements.
ous school. He makes frequent self-
Mark’s intellectual functioning was in
deprecating statements and reportedly
the low average range as was his measured
requires much attention and supervision to
academic achievement in reading, math,
complete his homework. He tends to give
and writing. His academic difficulties do
up easily on academic activities, often
not appear to be the result of a specific
refusing to complete them. He reportedly
learning disability, although it is reason-
has trouble separating from his mother, and
able to expect that Mark may struggle with
he often seeks attention in the classroom.
some academic tasks unless he is given
On the positive side, he is described as kind
some assistance in acquiring further aca-
to teachers, creative, and artistic.
demic skills.
Mark was reportedly born 3 months pre-
Based on findings from rating scales and
mature and weighed approximately 3
historical information, Mark was diag-
pounds at birth. He was treated in neonatal
nosed with Separation Anxiety Disorder.
intensive care for 2 months after delivery.
This diagnosis was based on the rationale
His mother reported that he did not walk
that he displayed the minimum of three
until 16 months, and he did not speak in
symptoms for a minimum of 4 weeks. His
single words until 24 months of age.
symptoms included:
He was reportedly diagnosed at age 5
with a motor delay, ADHD, and ODD. His
Excessive distress when separated from
teachers have reportedly always indicated
family members
that he responds better in a structured class-
Worry about the well-being of family
room. His grades have usually been above
members
grade level with the exception of mathe-
Worry about events that might cause sepa-
matics. He has not been retained nor has he
ration from family members
been served in special education. According
to his mother, Mark often complains of
An interesting aspect of this case is that
headaches and stomachaches at school and
the previous diagnoses of ADHD and ODD
has expressed significant worry about
were not confirmed based on detailed inter-
events such as severe weather that might
view information and rating scales. For
lead him to be stuck at school and not able
example, many of Mark’s oppositional
to be with his family.
behaviors seemed better explained by his
Mark’s test behavior was also remark-
refusal to separate from his mother, his dis-
able in that he refused testing initially. He
tress in the face of novel situations, and his
clung to his mother in the waiting room. He
difficulty working on tasks to their comple-
cried and requested that his mother join
tion when he was upset. His teacher
him for the testing. During testing (with his
reported that Mark only seems inattentive
mother in the room), he cried, threw temper
and distracted when faced with difficult
tantrums, refused to answer questions, and
tasks. On new or difficult tasks, Mark
Conclusions 375
8. A five-step method for assessing for depres- (b) Research is mixed on the degree of
sion and anxiety involves: parent-child agreement for anxiety
(a) Screening symptoms on structured interviews.
(b) Classification (c) The phenomenology of childhood anxi-
(c) Comorbidities ety disorders is very similar to that of
(d) Alternative causes adults.
(e) Past treatment (d) Anxiety disorders often occur in the
9. Each of the anxiety disorders described in absence of depression.
the DSM-5 has a different set of core 11. The Revised Children’s Manifest Anxiety
symptoms or problems. Thus, differential
Scale-2 is one example of a self-report rating
diagnosis is often quite important for opti- scale that offers a broad assessment of anxiety.
mal treatment planning. 12. Cognitive-behavioral assessment, behavioral
10. Some research findings regarding anxiety
observations, and clinician ratings have also
disorders include: been espoused as important assessment tools
(a) Many anxiety problems can be concep- for anxiety that have implications for inter-
tualized as exaggerations of normal vention planning and that enable progress
developmental processes (e.g., separa- monitoring. Structured diagnostic interviews
tion anxiety for young children). are often useful for differential diagnosis and
treatment planning as well.
Assessment of Autism Spectrum
Disorder 19
Association, 2013). With relatively high comor- 2018) and neurological (e.g., Courchesne et al.,
bidity, the differential diagnosis of ASD and 2001) underpinnings for ASD; however, specific
intellectual disability can be challenging. The etiological agents have not been identified. Much
DSM-5 emphasizes two classes of symptoms that remains to be learned about risk factors that may
are central to the disorder: social communication be etiologically involved in autism. The lack of
problems and repetitive and stereotyped behav- scientific clarity implies that clinicians should
iors. These symptoms are typically apparent dur- not place undue importance on etiological agents
ing the second year of life, yet more severe cases when conceptualizing cases of autism or when
may be noticeable earlier. The deficits present in providing feedback to others. Instead, the focus
relatively less severe cases may not be noticeable of assessment should be on symptomatology,
until later (American Psychiatric Association, developmental (e.g., language) history, strengths,
2013). The DSM criteria require that a total of and impairments.
five symptoms (three from the area of social Differentiation of ASD from intellectual dis-
communication and interaction and at least two ability can be quite difficult. Stereotyped behav-
from the area of restrictive, repetitive behaviors iors, poor social reciprocity, and impaired social
or interests) be present in order to make a interactions are among the symptoms that should
diagnosis. be closely assessed, and if present, signal that the
The social communication and interaction child may have ASD (Pedersen et al., 2017),
symptoms include: independent of whether the child would also
meet criteria for an intellectual disability. Aside
1 . Deficits in social-emotional reciprocity from differential diagnosis, assessment of intel-
2. Deficits in nonverbal communicative behav- lectual functioning is essential in evaluations for
iors used for social interaction ASD, as results of this testing provide an indica-
3. Deficits in developing, maintaining, and
tor of the child’s verbal and nonverbal function-
understanding relationships (American ing relative to peers in ways that should directly
Psychiatric Association, 2013, p. 50) inform treatment planning. Also, the child’s level
of intellectual functioning is quite important for
Problems with repetitive or stereotyped play determining how likely the child with ASD will
include: be able to function independently later in life
(Ozonoff et al., 2005).
1. Stereotyped or repetitive motor movements, Some of the assessment tools discussed in this
use of objects, or speech chapter may provide useful information in mak-
2. Insistence on sameness, inflexible adherence ing such distinctions between ASD and related
to routines, or ritualized patterns of verbal or problems. Nevertheless, it is incumbent on the
nonverbal behavior clinician to use a variety of methods to gain a
3. Highly restricted, fixated interests that are
comprehensive picture of the client’s difficulties
abnormal in intensity or focus and history such that the most appropriate inter-
4. Hyper- or hypo-reactivity to sensory input or ventions can then be sought.
unusual interest in sensory aspects of the envi-
ronment (American Psychiatric Association,
2013, p. 50) Comorbidity
not meet criteria for ASD. Needless to say, the child’s pediatrician. For relatively mild cases
inclusion of cognitive functioning in assessments and/or older children, such concerns might first
of ASD is critical for case conceptualization and come to the attention of a classroom teacher or
treatment planning. The same, then, can be said other professional. If a child demonstrates, or is
for the inclusion of a measure of adaptive func- reported to demonstrate delays that are consistent
tioning in the assessment battery, as adaptive with possible ASD (e.g., limited social communi-
functioning is part of the consideration for a diag- cation, limited language development, repetitive
nosis of intellectual disability but may also point behaviors), a more comprehensive assessment
to specific areas of intervention for youth with for ASD and other potential delays (e.g., intel-
ASD, independent of their intellectual lectual disability) would be warranted. The tools
functioning. discussed below are typically utilized in such an
As noted by Ozonoff and colleagues (2005), assessment.
new difficulties may emerge over the course of
development for youth with ASD such that initial
diagnosis as well as treatment planning and prog- utism Diagnostic Interview-Revised
A
ress monitoring are complicated. A common (ADI-R)
comorbid area of concern, particularly for high
functioning individuals with ASD, is anxiety The ADI-R is a semi-structured diagnostic inter-
(van Steensel & Heeman, 2017). However, the view designed to be used with a child’s primary
validity of commonly used measures of child- caregiver and can be used for anyone with a men-
hood anxiety—such as those discussed in the tal age of 2 years or higher (Lord, Rutter, & Le
preceding chapter—for individuals with ASD is Couteur, 1994). The ADI-R assesses behaviors
unknown (Ozonoff et al., 2005). relevant to the diagnosis of ASD in individuals
Regular, comprehensive assessment across from 18 months to adulthood. The ADI-R focuses
domains of functioning will aid in tracking treat- on three domains of behavior that generally align
ment gains as well as other important areas of with the domains of focus in the DSM-5 criteria:
difficulty apart from the primary problems asso- reciprocal social interaction, communication,
ciated with ASD. For example, many children and repetitive, restricted, or stereotyped behavior.
with ASD may also present with significant It is considered the “gold standard” of develop-
externalizing behavioral problems that occur mentally based interviews to aid in diagnostic
when expectations are not met, routines change, decisions concerning ASD (Walsh & Rau, 2018).
or due to frustration from communication diffi- Research has demonstrated good overall interra-
culties. These behaviors may themselves be a pri- ter reliability (Zander et al., 2017). Moreover,
mary focus of treatment. In addition, sleep scores on the ADI-R have been found to differen-
difficulties that may accompany ASD may war- tiate children with autism from children with
rant specific interventions (Ozonoff et al., 2005). other forms of delay (Mildenberger, Sitter,
Noterdaeme, & Amorosa, 2001) and good stabil-
ity in diagnostic status from less than 36 months
Assessment Tools of age to reassessment during elementary school
(Moss, Magiati, Charman, & Howlin, 2008).
Developmental Surveillance At the item level, one study evaluated interra-
ter reliability of the ADI-R across seven examin-
The first assessment for primary care physicians ers for one case (Cicchetti, Lord, Koenig, Klin, &
or others who may work with very young chil- Volkmar, 2008) and found perfect agreement for
dren and their families is developmental surveil- 74% of the items and “poor” agreement on less
lance (Ozonoff et al., 2005). This general than 10% of the items. A large-scale study of the
assessment typically involves screening of young ADI-R found support for a two-factor model
children for clear developmental delays by the (i.e., social communication and stereotyped
380 19 Assessment of Autism Spectrum Disorder
behaviors), which fits current DSM criteria, as scales is not documented. Moreover, it has not
well as a three-factor model (i.e., social commu- been updated to directly align with the DSM-5
nication, peer relationships/play, and stereotyped ASD criteria.
behaviors; Frazier, Youngstrom, Kubu, Sinclair,
& Rezai, 2008).
The strength of the ADI-R is its strong research hildhood Autism Rating Scale-2nd
C
base, including the validity evidence of the cur- Edition (CARS-2)
rent version and its predecessor which goes back
three decades (Le Couter et al., 1989). However, The CARS-2 (Schopler, Van Bourgondien,
the ADI-R is quite lengthy, taking anywhere from Wellman, & Love, 2010) is a widely used rating
90 min to 3 h to administer and score (Ozonoff form that is typically completed by an observer
et al., 2005). Therefore, it may be less practical of a child. The CARS-2 is designed for children
than other tools in many clinical settings. 2 years of age and higher. It takes about 5–10 min
to complete after sufficient observation of the
child has occurred (Schopler et al., 2010). It con-
Parent Interview for Autism (PIA) sists of a clinician form and a parent/caregiver
form. The clinician form consists of two 15-item
The PIA-Clinical Version (see Stone, Coonrod, versions: a standard form and a high-functioning
Pozdol, & Turner, 2003) is an alternative to the form to be used depending on the apparent cogni-
ADI-R. It is unique in its focus on tracking tive functioning of the child.
changes in symptoms over time, with parents rat- To date, most of the existing, and even recent,
ing the frequency of symptoms on a 5-point scale. research has focused on the original CARS rather
Research has shown good internal consistency than the CARS-2. Still, one study demonstrated that
and differential validity for the PIA; in addition, the CARS-2 did equally well in assisting with diag-
changes in PIA ratings, particularly in the areas nostic decisions under DSM-5 criteria as with
of social and communicative functioning, were DSM-IV criteria, which were the criteria in place at
related to changes reflected in other ratings of the the time of the development of the CARS-2
child symptomatology (Stone et al., 2003). While (Dawkins, Meyer, & Van Bourgondien, 2016). It
generally based on core dimensions of ASD, the has been proposed that the CARS, and by exten-
PIA has items that were developed and rationally/ sion, the CARS-2, is valid and useful for assessment
theoretically grouped into the following scales of ASD, even with the advent of new criteria and a
(Stone & Hogan, 1993): single diagnosis in DSM-5 (Park & Kim, 2016).
Social Relating
Affective Responses utism Diagnostic Observation
A
Motor Imitation Schedule (ADOS-2)
Peer Interactions
Object Play The ADOS-2 (Lord et al., 2012) is designed to
Imaginative Play assess symptoms of ASD in a wide range of indi-
Language viduals based on age and language skills. The
Understanding ADOS-2 consists of four modules, with the mod-
Nonverbal Communication ule selected based on the child’s current expres-
Motoric Behaviors sive language and developmental level (i.e.,
Sensory Responses Module 1 preverbal/single words; Module 2 for
Need for Sameness phrase speech; Module 3 for verbally fluent chil-
dren and young adolescents; Module 4 for
Although the PIA has shown some promise, verbally fluent adolescents and adults; Lord et al.,
research is limited, and the validity of the 12 sub- 2012). Module 2 is described in Box 19.1.
Assessment Tools 381
Box 19.1 A Closer View of the Autism 4. Joint Interactive Play: Assesses the
Diagnostic Observation Schedule (ADOS-2) extent to which the child initiates inter-
Together with the ADI-R, the ADOS, now action with the examiner during joint
available in a revised second edition, has play above and beyond responding to
been referred to as the current “gold stan- the examiner’s statements or requests.
dard” for the diagnosis of ASD (Ozonoff 5. Conversation: Evaluates extent to
et al., 2005, p. 524). The ADI-R, of course, which the child responds to the exam-
allows the clinician to gain a detailed his- iner’s statements or questions in a way
tory of the child’s development and symp- that leads to further back-and-forth
tomatology in a semi-structured interview conversation.
format. This approach is not particularly 6. Response to Joint Attention: The
unique among assessment practices. examiner observes the extent to which
However, the structured observation format the child follows the examiner’s shift
of the ADOS-2 is rather unique to child in gaze (with or without pointing).
assessment in that there are no well- Children with ASD may have diffi-
established analogous procedures for culty engaging in joint attention in this
assessment of internalizing or externalizing manner, particularly without other
problems. prompts such as pointing.
To further illustrate the content of the 7. Demonstration Task: Designed to
ADOS-2, Module 2, designed for children assess how well the child demonstrates
who speak in phrases, is described below. common activities (e.g., brushing
Module 2 consists of 14 tasks as follows teeth, getting dressed) without the use
(see Lord et al., 2012): of objects.
8. Description of a Picture: For this task,
1. Construction Task: Designed to assess the examiner attempts to elicit sponta-
how the child communicates a need for neous language from the child and to
more objects to complete a task. For observe what kinds of stimuli (from
example, some children with ASD pictures) are of interest to the child.
may grab the examiner’s hand to reach 9. Telling a Story from a Book: In addi-
for more objects rather than making a tion to assessing the child’s spontane-
verbal request. ous language and interest in a story,
2. Response to Name: Evaluates how this task, in requiring the child to tell a
readily the child responds to his/her story from a picture book, evaluates
name (e.g., by making eye contact) as whether the child can provide continu-
said by the examiner or a parent or ity in the story by sequencing events in
other stimuli directed to the child (e.g., the story in a sensible manner.
phrases, touching). Children with ASD 10. Free Play: This task is designed to assess
may require more direct stimuli or whether the child spontaneously seeks
prompting to respond. interaction or involvement from the
3. Make-Believe Play: Assesses whether examiner during free play. In addition,
child’s play includes imaginative use the extent to which the child maintains
of objects beyond their usual purpose attention to one task for an appropriate
and sequences of action and the extent interval is observed. Repetitive behav-
to which dolls are used to depict social iors—a symptom of ASD—may also be
interaction. evident during this task.
382 19 Assessment of Autism Spectrum Disorder
“what the child is trying to communicate through In situations in which interdisciplinary assess-
the behavior” (Ozonoff et al., 2005, p. 534), ment and intervention teams are not as readily
which takes on added importance in light of the available, it is still incumbent on the clinician to
behavioral problems and communication diffi- utilize methods that assess ASD comprehensively
culties that present with ASD. It is difficult to and specifically, rather than rely on subjective
imagine a comprehensive assessment of ASD clinical opinion or measures (e.g., IQ tests) that
that does not include some form of behavioral do not differentially diagnose ASD from other
observation and functional analysis as a means to developmental delays. Having said that, the
determine targets of intervention. Direct observa- incremental validity of specific ASD measures is
tion, coupled with functional analysis, allows the largely unknown (Ozonoff et al., 2005). As
clinician to also confirm or disconfirm the notions described above, the ADI-R and ADOS-2 pro-
of parents, teachers, or other caretakers as to the vide essential ASD-specific information, and
communicative intent of the child’s behavior and each has good sensitivity and specificity
on appropriate ways to reduce inappropriate (Falkmer, Anderson, Falkmer, & Horlin, 2013),
communicative behaviors (e.g., using another’s but they require a relatively large amount of time
hand to obtain a desired object; biting oneself to administer and significant training on the part
from frustration). Functional analysis, then, is of the clinician. As with other areas of child
conducted squarely with behavioral interventions assessment, more research is needed as to the
in mind. It is not a tool to differentially diagnose most sound, parsimonious battery that would
ASD from other problems. address referrals for ASD and lend itself to useful
recommendations.
Nevertheless, the severity and associated
An Assessment Strategy for ASD impairments of the symptomatology of ASD
require us to emphasize different aspects of our
ASD is recognized as having tremendous hetero- five-stage paradigm: screening, classification,
geneity in its symptoms and presentation (hetero- comorbidities, alternative causes, and treatment
geneous etiologies and highly variable considerations (see Table 19.1).
symptomatology among those affected (Masi, The screening process for ASD is aimed pri-
DeMayo, Glozier, & Guastella, 2017). Such marily at the population of children with
within-syndrome variability suggests that chil- developmental delays as opposed to the general
dren who are either suspected of, or diagnosed population. We advise that clinicians consider the
with, ASD should have access to multidisci- possibility of ASD in all young children with
plinary assessment procedures. The severity of developmental disabilities. Because the nature of
ASD symptoms, as demonstrated by the overlap ASD usually does not allow the child to serve as
with intellectual disability, their tremendous het- a source of screening information, parent and
erogeneity, and heterotypic continuity (i.e., pre- teacher ratings are likely to be the most fruitful.
senting differently across development; The time efficiency of such ratings argues for
Georgiades, Bishop, & Frazier, 2017), further their routine use with these populations.
emphasizes the need for a range of professionals Classification of the child as having ASD
to be involved in assessing the child. Such inter- requires meeting the DSM-5 criteria outlined ear-
disciplinary work, however, can be challenging if lier. The classification process will depend heav-
there are no adequate structures for encouraging ily on caretaker information such as detailed
interdisciplinary practice. Consequently, the psy- histories/interviews and possibly child behavior
chologist may have to make a concerted effort to ratings. Furthermore, medical, language, and
go beyond the typical practice regimen and com- other assessment data will need to be integrated
municate more frequently with colleagues who in order to make differential diagnoses. Lastly,
practice related specialties. direct observational tools (e.g., ADOS-2) are
384 19 Assessment of Autism Spectrum Disorder
Table 19.1 Assessment questions related to childhood lar, should be ruled out as primary etiologies for
autism spectrum disorders and implications for
the child’s behavior.
assessment
A central aspect of psychological intervention
Question Implications for assessment
for children with ASD is skill development in a
Screening Administer screening measures;
target young children with signs of variety of domains. A structured behavioral
developmental delays, particularly observation (e.g., ADOS-2, functional analysis)
in social communication and a comprehensive measure of adaptive behav-
Determine need for further ior are typically very useful for defining needed
assessment
skills and evaluating the effectiveness of inter-
Classification Assess for presence of symptoms
that meet DSM-5 criteria vention. See Box 19.2 for a case example of an
Determine stability and duration of assessment for a 5-year-old child.
symptoms
Determine age of onset
Comorbidities Assess for presence of intellectual
disability Conclusions
Determine the influence of autism
on school performance, adaptive ASD is among the most difficult of the develop-
behavior, development, etc. mental disorders to reliably classify because of
Assess social relationships and peer
social status its heterogeneous presentations. However, con-
Alternative Obtain developmental and medical siderable progress has been made in defining the
causes histories. core dimensions of impairment that separate this
Differential diagnosis/rule out of condition from other developmental disabilities
language disorders, intellectual
and in developing assessment tools geared toward
disability, ADHD, and hearing
impairment evaluating these dimensions. Many of these tools
Treatment Assess adaptive behavior assets and also appear to be useful in devising intervention
considerations deficits strategies for individuals with ASD.
Identify behavior problems that The available instruments can also differenti-
require intervention
Assess caregiver stress and ate ASD from other developmental disabilities
caregiver ability to provide adaptive (i.e., through assessing social reciprocity deficits
behavior instruction and behavioral specific to ASD), but this differentiation is par-
problem management ticularly difficult for very young children and
Evaluate response to previous
interventions those with more severe developmental delays.
Although new methods of assessment have made
strides in terms of differential validity, consider-
essential for determining the presence of social ably more progress is needed. The most promis-
reciprocity deficits and behavioral issues that are ing method for assessing and classifying autism
specific to ASD. at this time is a thorough and standard interview
As noted above, the most likely comorbid of the child’s primary caregiver along with either
condition for autism is intellectual disability. An a structured observation that systematically eval-
in-depth assessment of intelligence and adaptive uates symptoms of ASD or a functional analysis
behavior development is therefore required in that evaluates the communicative intent of a
every comprehensive assessment of a child sus- child’s behavioral problems. Complementing the
pected of autism or a related disorder. Sensory diagnostic process is the use of intelligence and
impairments must also be ruled out by other adaptive behavior scale such as the Vineland-3.
clinicians. In summary, current recommendations (e.g.,
The process of ruling out alternative causes de Bildt et al., 2004; Ozonoff et al., 2005) call for
can be a lengthy one requiring full participation a combination of a detailed developmental his-
by medical and other professionals in the assess- tory, interview focused on ASD criteria (e.g.,
ment process. Neurological problems, in particu- ADI-R), behavioral observation (e.g., ADOS-2),
Conclusions 385
Box 19.2 Case Example: A 5-year-old Girl with Autism Spectrum Disorder
Assessment Instruments
Parent interview
Teacher interview
Review of medical records
Stanford-Binet Scales for Early Childhood, 5th Edition (EARLY SB-5)
Developmental Test of Visual-Motor Integration (VMI)
Woodcock Johnson Psychoeducational Battery-4th Edition (WJ-IV) Tests of Achievement
Vineland Adaptive Behavior Scales-3 (Vineland-3)-Parent/Caretaker Form (Mother)
Vineland Adaptive Behavior Scales-2 (Vineland-3)-Classroom Edition (Teacher)
Autism Diagnostic Observation Schedule-2 (ADOS-2), Module 2
Achenbach System of Empirically Based Assessment (ASEBA) (CBCL and TRF)
Parenting Stress Index (PSI)
Referral Information
Josie was referred for a psychological evaluation by her parents when she was 5 years old.
Original referral questions included concern about delayed language development, short atten-
tion span, difficulty in her kindergarten class, and a possible learning disability. Subsequently,
however, Josie had two seizures and was diagnosed as having tuberous sclerosis and seizure
disorder. On learning that tuberous sclerosis often affects a child’s intellectual ability, her par-
ents requested a thorough evaluation to determine Josie’s current level of functioning.
Background Information
Josie lives with both parents, as well as her maternal grandmother, Nida. Her mother reported
that she had no serious difficulties when she was pregnant with Josie. Josie was reportedly born
full-term and weighed 7 pounds, 0 ounces. Both Josie and her mother were reportedly in good
condition at the time of birth. Josie did, however, have an open sore on her back that was sur-
rounded by a purple mark. The doctor reportedly remarked at the time that this type of wound
was very unusual.
According to her mother, Josie has also had frequent ear infections, beginning at 8 months of
age and continuing until 3 months ago. At the time of the last ear infection, she reportedly had
tubes placed in her ears. She has reportedly had no difficulties since then. She had a hearing
examination 2 months ago that indicated a slight hearing loss in the right ear. Josie has also had
frequent colds, according to her mother. Josie’s vision is reportedly normal.
Josie’s mother described Josie’s development as inconsistent. She seemed to learn gross
motor skills at a normal rate, but her fine motor and language development appeared to be
delayed. She reportedly spoke her first words at 9 months, but by 18 months, she could still
speak only four words. Her mother states that Josie did not speak in sentences until she was
2½ years old. Her speech articulation is reportedly clear, but her expressive language is still
delayed, as she generally speaks only in phrases. She currently receives speech therapy.
Her mother stated that Josie is sweet, has a good sense of humor, and enjoys singing. Josie,
however, can also be stubborn and sometimes does not seem to listen to or obey her mother. Her
mother noted that Josie likes to play by herself rather than with other children. She also reported
that Josie can be easily overstimulated in that she seems uncomfortable in crowded or noisy
places, is very energetic and active, has a short attention span, and is impulsive (e.g., not waiting
386 19 Assessment of Autism Spectrum Disorder
for her turn, grabbing objects at stores without permission). She added that Josie seems to
require a great deal of parental attention.
Given some of these concerns, more structured interview questions were administered to
determine the presence of symptoms consistent with autism spectrum disorder. Responses on
the ADI-R indicated that Josie lacks shared enjoyment of activities with both adults and other
children, engages in limited play with other children, uses her mother’s hand to reach for desired
objects, and exhibits facial expressions that do not seem to match the situation.
Josie has reportedly been in a preschool program since she was 2 years old. Her mother
stated that Josie’s teachers were concerned because Josie did not communicate in school, had
difficulty doing age-appropriate work, required much teacher attention, and seemed to lack self-
esteem and confidence. Josie is now enrolled in a kindergarten class. Her current teacher
reported that Josie is a sweet child and seems to want to please others. Josie reportedly memo-
rizes well and knows many songs. Josie’s teacher also stated, however, that Josie shows many
behaviors that make it difficult to teach her. She reportedly spends most of the day by herself,
talking only to herself or inanimate objects. She seems to repeat, over and over, phrases that she
has heard adults say. According to her teacher, Josie does not interact with the other children.
When called on in class, she seems embarrassed and then blurts out any answer. She reportedly
has a very short attention span and also sometimes leaves the room without teacher permission.
Her teacher stated that she is teased by the other children because of her behavior, which
includes chewing on her mat and her clothes. According to her teacher, Josie has not learned
many academic skills. She seems to have learned a few basic concepts, such as shapes. She can
also make a capital “A,” but it is frequently upside down. She reportedly writes and works with
other objects upside down frequently and does not seem to recognize that she is doing so. She
reportedly cannot complete worksheets, as she just draws lines up and down the page or colors
the whole page one color. According to her teacher, Josie sometimes does better with one-on-
one assistance, but at other times, she will not accept help from her teacher.
Josie was briefly observed at school during lunch. She was sitting at the end of the table by
herself and did not interact with the other children. Although she had talked with the observer
previously, she showed little response to him at this time. When asked if she remembered the
observer, she said that she did, but she still did not interact with him. She just continued
eating.
Behavioral Observations and ADOS-2
Josie was tested over a 2-day period. Two examiners participated in her evaluation. She was
brought to the first testing session by her mother. She seemed very uncomfortable meeting
strangers and clung to her mother. Her mother carried her to the testing room and stayed for a
while until Josie was more comfortable. Josie did not become relaxed with the examiners for
some time. Adequate rapport for testing purposes was finally established.
During testing, Josie was very active and distractible. The examiners had difficulty keeping
her attention on the test materials. She required much direction and attention from the examin-
ers to stay on task. She frequently wanted to play with the other toys. At one point, she sat on
the floor, facing the wall. When asked what she was doing, she said she was in time-out. On
further questioning, she responded that she was put there because she did not pay attention.
Later, she again sat on the floor, this time repeating “Josie, it’s time to go to school.” The exam-
iner had difficulty redirecting her from this behavior. This type of behavior seemed to occur
most when Josie was being asked to perform a task that was difficult for her. After this last time,
Conclusions 387
Josie could not be enticed to pay any further attention to the test materials. She got up and left
the room without the examiner’s permission.
Josie was more cooperative during the second testing session. She seemed much more com-
fortable with the examiners and accompanied them readily to the testing room. She was, how-
ever, still very distractible and active, requiring much examiner attention for her to stay on task.
During the second session, the ADOS-2 was administered. Throughout the session, Josie exhib-
ited positive affect but made no efforts at engaging the examiners in tasks or conversation.
During bubble play, she observed the examiners but did not blow bubbles herself. She repeat-
edly said the word “bubbles.” She was able to demonstrate brushing her teeth with her finger
and engaged well in turn taking with prompts. Josie covered her ears during the birthday party
task apparently because of the loud noise. Her mother remarked that this reaction was typical
for Josie. When asked to tell a story from a book, Josie was brief in her response, and it was
difficult to determine a clear sequence of events from her story. Her score of 9 on the ADOS-2
is consistent with a diagnosis of Autism Spectrum Disorder.
Tests Results and Interpretations
Intellectual Functioning
Josie’s intellectual functioning was evaluated with the Stanford-Binet Scales for Early
Childhood (EARLY SB-5). Her performance on this test indicates that her current level of intel-
lectual functioning is far, to well below, that of her same-aged peers. This level of functioning
is consistent with her teacher’s and mother’s reports of her ability and with her current school
performance.
More specifically, Josie’s performance on the EARLY SB-5 (Full Scale IQ = 69) suggests
that her level of intellectual functioning meets or exceeds that of 2% of her same-aged peers.
Her performance across subtests was consistent, showing no relative strengths or weaknesses.
Visual Motor Integration
Josie was also given the VMI to assess her ability to coordinate what she sees with what she
is doing. This test required Josie to copy designs in a booklet. Her performance on this task was
at or above that of 8% of children her age, suggesting that her visual motor integration and fine
motor skills are well below average. Her mother and teacher also reported difficulty with fine
motor tasks, such as drawing and writing.
Academic Achievement
Josie was administered the Woodcock-Johnson Tests of Achievement-Fourth Edition to
determine what basic reading and math skills Josie had learned. However, only two subtests
were able to be administered. Her performance on the subtests administered was far below aver-
age. Specifically, her score on the Letter-Word Identification subtest meets or exceeds that of
0.5–2% of other children her age, whereas her score on the Applied Problems (math) subtest
meets or exceeds that of 1–5%. On these subtests and through informal assessment, Josie
showed the ability to sing the alphabet and count objects with one-to-one correspondence.
Adaptive Behavior
The Vineland-3 is a measure of Josie’s ability to take care of herself, get along with others,
and live in the community appropriately for her age. It includes four domains: Communication,
Daily Living Skills, Socialization, and Motor Skills. On the Vineland-3, Josie’s mother indi-
cated that Josie’s overall ability in these domains is below average for her age. Most of her
scores were consistent with her overall Adaptive Composite score. Her score in the Socialization
domain was lower than the others, which is consistent with maternal report on the ADI-R. This
score, is also consistent with teacher and behavioral observations.
388 19 Assessment of Autism Spectrum Disorder
Josie’s teacher completed the Vineland-3, Classroom Edition. Her teacher indicated that
Josie’s adaptive behavior is variable. Josie’s score on the Daily Living Skills domain from her
teacher’s report is similar to that given by her mother and is in the below average range. However,
her other scores are in the well below average range. Her Socialization domain score is consis-
tent with that of her mother’s rating. This result indicates far below average socialization skills
for Josie at school relative to same-aged peers.
Overall, the report of Josie’s mother and teacher on the Vineland-3 indicates that Josie’s
adaptive behavior is below what would be expected for her age but is somewhat above her intel-
lectual ability as assessed in this evaluation. However, it does not appear that a diagnosis of
intellectual disability is appropriate at this point, as Josie’s adaptive functioning is not far
enough below what would be expected for her age to warrant such a diagnosis. In addition, her
overall intellectual ability is right at the level typically considered for such a diagnosis.
Behavioral, Emotional, and Social Functioning
The results of the ADOS-2 and the parent interview indicate that Josie meets criteria for a
diagnosis of Autism Spectrum Disorder (ASD). She has limited and delayed verbal communica-
tion, demonstrates little social reciprocity (e.g., joint attention, eye contact, initiating and main-
taining conversation), demonstrates some echolalia, and seems hypersensitive to loud noises.
Her symptoms do not appear to be better accounted for by an intellectual disability.
Josie’s mother and father completed the ASEBA CBCL, whereas her teacher completed the
TRF. This rating scale assesses behavioral and emotional problems of children. Although the
raters reported similar behavioral patterns, only Josie’s father indicated that Josie has any seri-
ous problems. He rated Josie as having problems in the areas of social withdrawal and hyperac-
tivity. Josie’s mother and teacher also rated her high in these areas but not significantly so. These
findings are consistent with teacher and parent report and behavioral observations in that Josie
has some incidents of apparent overactivity in the classroom and consistently seems uncomfort-
able interacting with others. Based on Josie’s current emotional, behavioral, social, intellectual,
and adaptive functioning, as well as her history of communication delays, a diagnosis of autism
spectrum disorder seems appropriate.
Summary and Conclusions
Josie is a 5-year-old girl referred by her parents to determine her current level of functioning.
She has been diagnosed as having seizure disorder and tuberous sclerosis. Her mother reported
that her language and fine motor development were delayed. Her teacher reported that she has
made little progress in adjusting to kindergarten or learning basic academic skills. Parent and
teacher reports indicate that she is active and inattentive and requires much adult attention and
supervision. These behaviors were also observed in the testing situation, in which examiners
had much difficulty getting Josie to cooperate and attend to the tasks. She was very active and
distractible. Parent and teacher reports, as well as observation during testing, also suggest that
Josie displays several unusual behaviors, such as talking to herself or inanimate objects, repeat-
ing the same phrases or motions, little spontaneous engagement in conversation with others,
sensitivity to loud noises, chewing on clothes, and preferring to stay to herself.
Most of Josie’s test results, including intellectual, achievement, visual motor, and some
adaptive behavior areas, are in the well below average range. Other adaptive behavior scores,
however, were only slightly below average for her age. Information from rating scales indicates
that Josie tends to be socially withdrawn, hyperactive, distractible, slow to learn or adapt to new
situations, and sometimes fussy.
Conclusions 389
Taking all of the assessment information together, the diagnosis of Autism Spectrum Disorder
(ASD) seems appropriate. The level of impairment appears to be “Level 1 – Requiring Support.”
At this time, a diagnosis of intellectual disability does not seem applicable. Although her intel-
ligence test scores are within the mild intellectual disability range, her adaptive behavior scores
lie outside this range. Both of these areas of functioning must be significantly below average to
warrant a diagnosis of intellectual disability. However, the specifier “With Accompanying
Intellectual Impairment” to her ASD diagnosis seems warranted.
Recommendations
Josie should be evaluated again at the end of this academic year to assess her progress. At this
time, the following recommendations are made for Josie:
1. With her diagnosis of ASD, Josie is eligible for an Individualized Education Plan (IEP) at
school. Her parents should consult school personnel to determine the most appropriate
placement and accommodations for her. She would benefit from a highly structured educa-
tional environment where she can get the attention and supervision she needs to learn.
2. Josie will need much individual instruction, gradually moving to small-group instruction.
3. Josie’s classroom assignments should be within the range appropriate for her intellectual
and academic level.
4. Instructional tasks should be organized into short, structured units.
5. Josie will need to be taught at a very concrete and practical level, using many manipulatives
and teaching aids.
6. Skills and concepts may need to be repeated and reviewed often for Josie to master them.
7. Josie may be helped by being tutored by someone who has been trained in tutoring children
with her set of difficulties.
8. Josie’s parents and teachers should continue to search for activities that she does well and
enjoys and encourage her in these activities.
9. Josie should be provided with increased opportunities to learn and practice age-appropriate
self-care and socialization skills. Some social skills can be practiced with positive, typically
developing peers.
10. Josie’s parents and teachers should continue to use behavioral management strategies, such
as positive reinforcement and redirection to help her learn appropriate behavior. Her par-
ents may wish to attend sessions with a mental health specialist to become more facile with
these techniques.
11. Inappropriate behavior should be ignored as much as possible, while appropriate behavior
is rewarded with praise and other appropriate reinforcers.
12. Communication between school and home should be maintained. A school-home note, in
which Josie is rewarded at home for appropriate behavior at school, may be helpful.
13. Josie’s parents should closely monitor her physical functioning and symptoms and main-
tain regular contact with her physicians regarding her seizures.
14. Josie should be evaluated in approximately 1 year to determine her level of functioning,
particularly after the above strategies have been implemented.
Psychometric Summary
Josie’s EARLY SB-5 scores
Scores Percentile
Verbal IQ 69 2
Nonverbal IQ 73 3
Full Scale IQ 69 2
390 19 Assessment of Autism Spectrum Disorder
Abdekhodaie, Z., Tabatabaei, S. M., & Gholizadeh, M. Achenbach, T. M., & Rescorla, L. A. (2010a). Multicultural
(2012). The investigation of ADHD prevalence in kin- supplement to the manual for the ASEBA preschool
dergarten children in northeast Iran and a determina- forms and profiles. Burlington, VT: Authors.
tion of the criterion validity of Conners’ questionnaire Achenbach, T. M., Rescorla, L. A., & Ivanova, M. (2012).
via clinical interview. Research in Developmental International epidemiology of child and adolescent
Disabilities, 33, 357–361. psychopathology I: Diagnoses, dimensions, and con-
Abidin, R., Flens, J. R., & Austin, W. G. (1995). The ceptual issues. Journal of the American Academy of
Parenting Stress Index. In R. P. Archer (Ed.), Forensic Child and Adolescent Psychiatry, 51, 1261–1272.
uses of clinical assessment instruments (pp. 297–328). AERA, APA, & NCME. (2014). Standards for educa-
Mahwah, NJ: Erlbaum. tional and psychological testing. Washington, DC:
Abidin, R. R. (2012). Parenting stress index, third edition AERA.
short form. Lutz, FL: PAR, Inc. Allen, B., & Tussey, C. (2012). Can projective draw-
Abidin, R. R. (2014). Parenting stress index, fourth edi- ings detect if a child experienced sexual or physical
tion. Lutz, FL: PAR, Inc. abuse? A systematic review of the controlled research.
Achenbach, T. M. (1982). Assessment and taxonomy of Trauma, Violence, & Abuse, 12, 97–111.
children’s behavior disorders. In B. B. Lahey & A.E. Allen, J. C., & Hollifield, J. (2003). Using the
Kazdin (Eds.), Advances in clinical child psychology Rorschach with children and adolescents: The Exner
(Vol. 5, pp. 1–38). New York: Plenum. Comprehensive System. In C. R. Reynolds & R. W.
Achenbach, T. M. (1986). Child behavior checklist-direct Kamphaus (Eds.), Handbook of psychological and
observation form (rev. ed.). Burlington, VT: University educational assessment of children: Personality,
of Vermont. behavior, and context (2nd ed., pp. 182–197).
Achenbach, T. M. (1991). Manual for the Child Behavior New York, NY: Guilford.
Checklist/4-18 and 1991 profile. Burlington, VT: Altepeter, T. S., & Breen, M. J. (1989). The home situ-
University of Vermont, Department of Psychiatry. ations questionnaire (HSQ) and the school situa-
Achenbach, T. M. (2001). Manual for the ASEBA school- tions questionnaire (SSQ): Normative data and an
age forms and profiles. Burlington, VT: Research evaluation of psychometric properties. Journal of
Center for Children, Youth, and Families. Psychoeducational Assessment, 7, 312–322.
Achenbach, T. M., Dumenci, L., & Rescorla, L. A. (2003). Amato, P. R., & Keith, B. (1991). Parental divorce and the
DSM-oriented and empirically based approaches to well-being of children: A meta-analysis. Psychological
constructing scales from the same item pools. Journal Bulletin, 110, 26–46.
of Clinical Child and Adolescent Psychology, 32, American Psychiatric Association. (1968). The diagnostic
328–340. and statistical manual of mental disorders (2nd ed.).
Achenbach, T. M., McConaughy, S. H., & Howell, C. T. Washington, DC: Author.
(1987). Child/adolescent behavioral and emotional American Psychiatric Association. (1980). The diagnostic
problems: Implications of cross-informant correla- and statistical manual of mental disorders (3rd ed.).
tions for situational specificity. Psychological Bulletin, Washington, DC: Author.
101, 213–232. American Psychiatric Association. (1987). The diagnostic
Achenbach, T. M., & Rescorla, L. A. (2000). Manual for and statistical manual of mental disorders (3rd ed.).
the ASEBA preschool forms lv profiles. Burlington, Washington, DC: Author.
VT: University of Vermont, Department of Psychiatry. American Psychiatric Association. (1994). The diagnostic
Achenbach, T. M., & Rescorla, L. A. (2001). Manual for the and statistical manual of mental disorders (4th ed.).
ASEBA school-age forms and profiles. Burlington, VT. Washington, DC: Author.
American Psychiatric Association. (2000). The diagnostic behaviors and school performance. Development and
and statistical manual of mental disorders (4th ed.). Psychopathology, 21, 319–341.
Washington, DC: Author. Aos, S., Lieb, R., Mayfield, J., Miller, M., & Pennucci,
American Psychiatric Association. (2013). The diagnostic A. (2004). Benefits and costs of prevention and early
and statistical manual of mental disorders (5th ed.). intervention programs for youth. Olympia, WA:
Washington, DC: Author. Washington State Institute for Public Policy.
American Psychological Association. (2016). Ethical Appelbaum, M. (2009). How to handle hard-to-handle
principles of psychologists and code of conduct. parents. Thousand Oaks, CA: Corwin Press.
Washington, DC: Author. Applegate, B., Lahey, B. B., Hart, E. L., Waldman, I.,
American Psychological Association. (2017). Biederman, J., Hynd, G. W., et al. (1997). The age of
Multicultural guidelines: An ecological approach to onset criterion for DSM-IV attention-deficit hyper-
context, identity, and intersectionality. Washington, activity disorder: A report of the DSM-IV field tri-
DC: Author. als. Journal of the American Academy of Child and
Ames, L. B., Metraux, R., Rodell, J. L., & Walker, Adolescent Psychiatry, 36, 1211–1221.
R. (1974). Child Rorschach responses (rev. ed.). Archer, R. P. (1997). MMPI-A: Assessing adolescent
New York, NY: Bruner/Mazel. psychopathology (2nd ed.). Mahwah, NJ: Lawrence
Anastasi, A. (1988). Psychological testing (16th ed.). Erlbaum.
New York, NY: Macmillan. Archer, R. P. (2005). MMPI-A: Assessing adolescent
Anastasi, A. (1992). Introductory remarks. In K. F. psychopathology (3rd ed.). Mahwah, NJ: Lawrence
Geisinger (Ed.), Psychological testing of Hispanics Erlbaum Associates.
(pp. 1–7). Washington, DC: American Psychological Archer, R. P. (2017). Assessing adolescent psychopathol-
Association. ogy: MMPI-A/MMPI-A-RF (4th ed.). New York, NY:
Anastasi, A., & Urbina, S. (1998). Psychological testing Routledge Press.
(7th ed.). New York, NY: Prentice Hall. Archer, R. P., Handel, R. W., Ben-Porath, Y. S., & Tellegen,
Andershed, H., Hodgins, S., & Tengström, A. (2007). A. (2016). Minnesota Multiphasic Personality
Convergent validity of the Youth Psychopathic Traits Inventory-Adolescent-Restructured Form (MMPI-
Inventory (YPI) association with the Psychopathy A-RF): Administration, scoring, interpretation, and
Checklist: Youth Version (PCL: YV). Assessment, 14, technical manual. Minneapolis, MN: University of
144–154. Minnesota Press.
Andershed, H. A., Kerr, M., Stattin, H., & Levander, S. Archer, R. P., & Krishnamurthy, R. (1997). MMPI-A and
(2002). Psychopathic traits in non-referred youths: A Rorschach indices related to depression and conduct dis-
new assessment tool. In Psychopathy: Current inter- order: An evaluation of the incremental validity hypoth-
national perspectives (pp. 131–158). Den Haag, The esis. Journal of Personality Assessment, 9, 517–533.
Netherlands: Elsevier. Arney, F., Rogers, H., Baghurst, P., Sawyer, M., & Prior,
Anderson, E., & Hope, D. A. (2008). A review of the M. (2008). The reliability and validity of the Parenting
tripartite model for understanding the link between Scale for Australian mothers of pre-school aged chil-
anxiety and depression in youth. Clinical Psychology dren. Australian Journal of Psychology, 60, 44–52.
Review, 28, 275–287. Arnold, D. S., O’Leary, S. G., Wolff, L. S., & Acker, M. M.
Andover, M. S., Izzo, G. N., & Kelly, C. A. (2011). (1993). The Parenting Scale: A measure of dysfunc-
Comorbid and secondary depression. In M. Crozier, tional parenting in discipline situations. Psychological
S. Gillihan, & M. Powers (Eds.), Handbook of child Assessment, 5, 137–144.
and adolescent anxiety disorders (pp. 135–153). Asher, S. R., & Hymel, S. (1981). Children’s social com-
New York, NY: Springer New York. petence in peer relations: Sociometric and behavioral
Ang, R. P., Lowe, P. A., & Yusof, N. (2011). An exami- assessment. In J. D. Wine & M. D. Smye (Eds.), Social
nation of the RCMAS-2 scores across gender, ethnic competence (pp. 125–157). New York, NY: Guilford
background, and age in a large Asian school sample. Press.
Psychological Assessment, 23, 899–910. Asher, S. R., Singleton, L. C., Tinsley, B. R., & Hymel, S.
Ang, R. P., Rescorla, L. A., Achenbach, T. M., Ooi, Y. P., (1979). A reliable sociometric measure for preschool
Fung, D. S. S., & Woo, B. (2012). Examining the cri- children. Developmental Psychology, 15, 443–444.
terion validity of CBCL and TRF problem scales and Babinski, D. E., Waxmonsky, J. G., & Pelham, W. E.
items in a large Singapore sample. Child Psychiatry (2014). Treating parents with attention-deficit/hyper-
and Human Development, 43, 70–86. activity disorder: The effects of behavioral parent
Angold, A., & Costello, J. (2000). The child and ado- training and acute stimulant medication treatment on
lescent psychiatric assessment (CAPA). Journal parent-child interactions. Journal of Abnormal Child
of the American Academy of Child and Adolescent Psychology, 42, 1129–1140.
Psychiatry, 39, 49–58. Badaly, D., Duong, M. T., Ross, A. C., & Schwartz, D.
Ansary, N. S., & Luthar, S. S. (2009). Distress and aca- (2015). A peer-nomination assessment of electronic
demic achievement among adolescents of affluence: forms of aggression and victimization. Journal of
A study of externalizing and internalizing problem Adolescence, 44, 77–87.
References 395
Bae, Y. (2012). Review of children’s depression inven- Barkley, R. A., Fischer, M., Smallish, L., & Fletcher, K.
tory (CDI-2) (2nd ed.). Journal of Psychoeducational (2002). The persistence of attention-deficit/hyperac-
Assessment, 30, 304–308. tivity disorder into young adulthood as a function of
Bagner, D. M., Boggs, S. R., & Eyberg, S. M. (2010). reporting source and definition of disorder. Journal of
Evidence-based school behavior assessment of exter- Abnormal Psychology, 111(2), 279–289.
nalizing behavior in young children. Education & Barkley, R. A., Guevremont, D. G., Anastopoulos, A. D.,
Treatment of Children, 33, 65–83. & Fletcher, K. F. (1993). Driving related risks and
Bagner, D. M., Harwood, M. D., & Eyberg, S. M. (2006). outcomes of attention-deficit hyperactivity disorder in
Psychometric considerations. In M. Hersen (Ed.), adolescents and young adults. Pediatrics, 92, 212–218.
Clinician’s handbook of child behavioral assessment Barkley, R. A., Murphy, K. R., & Fischer, M. (2008).
(pp. 63–79). New York, NY: Elsevier Academic Press. ADHD in adults: What the science says. New York,
Baio, J., Wiggins, L., Christensen, D. L., Maenner, M. J., NY: Guilford Press.
Daniels, J., Warren, Z., … Dowling, N. F. (2014). Barlow, D. H. (2002). Anxiety and its disorders: The
Prevalence of autism spectrum disorder among chil- nature and treatment of anxiety and panic (2nd ed.).
dren aged 8 years—Autism and developmental dis- New York, NY: Guilford.
abilities monitoring network. MMWR Surveillance Barrios, B. A. (1993). Direct observation. In F. H.
Summary, 67, 1–23. Ollendick & M. Hersen (Eds.), Handbook of child and
Balázs, J., & Keresztény, Á. (2014). Subthreshold atten- adolescent assessment (pp. 140–164). Boston, MA:
tion deficit hyperactivity in children and adolescents: Allyn & Bacon.
A systematic review. European Child & Adolescent Barrios, B. A., & Hartmann, D. P. (1988). Fears and anxi-
Psychiatry, 23, 393–408. ety. In E. J. Mash & L. G. Terdal (Eds.), Behavioral
Baldwin, J. S., & Dadds, M. R. (2008). Anxiety disorders. assessment of child-hood disorders (pp. 167–221).
In M. Hersen & D. Reitman (Eds.), Handbook of psy- New York, NY: Guilford Press.
chological assessment, case conceptualization, and Barroso, N. E., Hungerford, G. M., Garcia, D., Graziano,
treatment (pp. 231–263). Hoboken, NJ: John Wiley & P. A., & Bagner, D. M. (2016). Psychometric proper-
Sons. ties of the Parenting Stress Index-Short Form (PSI-SF)
Ball, J. D., Archer, R. P., Gordon, R. A., & French, J. in a high risk sample of mothers and their infants.
(1991). Rorschach depression indices with children Psychological Assessment, 28, 1331–1335.
and adolescents: Concurrent validity findings. Journal Barry, C. T., Frick, P. J., & Grafeman, S. J. (2008).
of Personality Assessment, 57, 465–476. Child versus parent report of parenting practices:
Bao-Yu, W., & Lin-Yan, S. (2004). Family environ- Implications for the conceptualization of child
ment and the development of behavior in children behavioral and emotional problems. Assessment, 15,
with ADHD. Chinese Journal of Psychology, 1(2), 294–303.
145–146. Barry, T. D., Pinard, F. A., Barry, C. T., Garland, B. H.,
Barker, P. (1990). Clinical interviews with children and & Lyman, R. D. (2011). The utility of home problem
adolescents. New York, NY: Norton. pervasiveness and severity in classifying children
Barkley, R. A. (1988). Attention-deficit disorder with diagnosed with attention-deficit/hyperactivity disor-
hyperactivity. In E. J. Mash & L. G. Terdal (Eds.), der. Child Psychiatry and Human Development, 42,
Behavioral assessment of childhood disorders (2nd 152–165.
ed., pp. 69–104). New York, NY: Guilford Press. Bauer, D. H. (1976). An exploratory study of develop-
Barkley, R. A. (1990). Attention-deficit hyperactivity dis- mental changes in children’s fears. Journal of Child
order: A handbook to diagnosis and treatment (2nd Psychology and Psychiatry, 17(1), 69–74.
ed.). New York, NY: Guilford Press. Baumrind, D. (1971). Harmonious parents and their
Barkley, R. A. (1991). The ecological validity of labo- preschool children. Developmental Psychology, 4,
ratory and analogue assessment methods of ADHD 99–102.
symptoms. Journal of Abnormal Child Psychology 1, Bean, T., Mooijaart, A., Eurelings-Bontekoe, E., &
9(2), 149–178. Sprinhoven, P. (2007). Validation of the teacher’s
Barkley, R. A. (1997a). Attention deficit hyperactivity report form for teachers of unaccompanied refugee
disorder. In E. J. Mash & L. G. Terdel (Eds.), Assess minors. Journal of Psychoeducational Assessment, 25,
ment of childhood disorders (3rd ed., pp. 71–129). 53–68.
New York: Guilford Press. Beck, S. T., Steer, R. A., & Garbin, M. G. (1988).
Barkley, R. A. (1997b). ADHD and the nature of self- con- Psychometric properties of the Beck Depression
trol. New York: Guilford Press. Inventory: Twenty-five years of evaluation. Clinical
Barkley, R. A., & Edelbrock, C. (1987). Assessing situ- Psychology Review, 8, 77–100.
ational variation in children’s problem behaviors: Becker, E. M., Jensen-Doss, A., Kendall, P., Birmaher,
The home and school situations questionnaires. In B., & Ginsburg, G. S. (2016). All anxiety is not cre-
R. J. Prinz (Ed.), Advances in behavioral assess- ated equal: Correlates of parent/youth agreement vary
ment of children and families (Vol. 3, pp. 157–176). across subtypes of anxiety. Journal of Psychopathology
New York, NY: JAI. and Behavioral Assessment, 38, 528–537.
396 References
Beesdo, K., Knappe, S., & Pine, D. S. (2009). Anxiety of the American Academy. of Child and Adolescent
and anxiety disorders in children and adolescents: Psychiatry, 32, 361–368.
Developmental issues and implications for DSM-V. Blader, J. C. (2004). Symptom, family, and service pre-
Psychiatric Clinics of North America, 32, 483–524. dictors of children’s psychiatric rehospitalization
Behar, L. B., & Stringfield, S. A. (1974). A behavior rat- within one year of discharge. Journal of the American
ing scale for the preschool child. Child Development, Academy of Child & Adolescent Psychiatry, 43,
10, 601–610. 440–451.
Beidel, D. C., Turner, S. M., & Morris, T. (1995). A new Blake, J. J., Kim, E. S., & Lease, A. M. (2011). Exploring
inventory to assess childhood social anxiety and pho- the incremental validity of nonverbal social aggres-
bia—The Social Phobia and Anxiety Inventory for sion: The utility of peer nominations. Merrill-Palmer
children. Psychological Assessment, 7, 73–79. Quarterly, 57, 293–318.
Belardinelli, C., Hatch, J. P., Olvera, R. L., Fonseca, M., Blashfield, R. K. (1984). The classification of psy-
Caetano, S. C., Nicoletti, M., … Soares, J. C. (2008). chopathology: Neo-Kraepelinian and quantitative
Family environment patterns in families with bipolar approaches. New York, NY: Plenum.
children. Journal of Affective Disorders, 107(1–3), Blum, G. S. (1950). The blacky pictures: Manual of
299–305. instructions. New York, NY: The Psychological
Bell, N. L., & Nangle, R. J. (1999). Interpretive issues Corporation.
with the Roberts apperception test for children: Bohnenkamp, J. H., Glascoe, T., Gracey, K. A., Epstein,
Limitations of the standardization group. Psychology R. A., & Benningfield, M. M. (2015). Implementing
in the Schools, 36, 277–283. clinical outcomes assessment in everyday school men-
Bellak, L., & Bellak, S. S. (1949). The children’s apper- tal health practice. Child and Adolescent Psychiatric
ception test. Larchmont, NY: CPS. Clinics of North America, 24, 399–413.
Benton, T. D., Blume, J., Crits-Cristoph, P., Dubé, P., Boothroyd, R. A., & Armstrong, M. (2010). An exami-
& Evans, D. L. (2013). Depression and medical ill- nation of the psychometric properties of the Pediatric
ness. In D. Charney, J. D. Buxbaum, P. Sklar, & E. J. Symptom Checklist with children enrolled in
Nestler (Eds.), Neurobiology of mental illness (4th Medicaid. Journal of Emotional and Behavioral
ed., pp. 496–507). New York, NY: Oxford University Disorders, 18(2), 113–126.
Press. Borden, L. A., Herman, K. C., Stormont, M., Goel, N.,
Bernal, P., Estroff, D. B., Aboudarham, J. F., Murphy, Darney, D., Reinke, W. M., & Webster-Stratton, C.
M., Keller, A., & Jellinek, M. S. (2000). Psychosocial (2014). Latent profile analysis of observed parenting
morbidity: The economic burden in a pediatric behavior in a clinical sample. Journal of Abnormal
health maintenance organization sample. Archives of Child Psychology, 42, 731–742.
Pediatrics & Adolescent Medicine, 154, 261–266. Bornstein, M. R., Bellack, A. S., & Hersen, M. (1977).
Bertha, E. A., & Balázs, J. (2013). Subthreshold depres- Social-skills training for unassertive children: A mul-
sion in adolescence: A systematic review. European tibaseline analysis. Journal of Applied Behavioral
Child & Adolescent Psychiatry, 22, 589–603. Analysis, 10, 183–195.
Biederman, J., Gao, H., Rogers, A., & Spencer, T. (2006). Bourgeron, T. (2016). Current knowledge of the genet-
Comparison of parent and teacher reports of attention- ics of autism and propositions for future research.
deficit/hyperactivity disorder symptoms from two Comptus Rendes Biologies, 339, 300–307.
placebo-controlled studies of atomoxetine in children. Bower, G. H. (1981). Mood and memory. American
Biological Psychiatry, 60, 1106–1110. Psychologist, 36, 129–148.
Biederman, J., Keenan, K., & Faraone, S. V. (1990). Bowlby, J. (1969). Attachment and loss: I. Attachment.
Parent-based diagnosis of attention deficit disorder New York, NY: Basic Books.
predicts a diagnosis based on teacher report. Journal Boyd, C. P., Gullone, E., Needleman, G. L., & Burt, T.
of the American Academy of Child and Adolescent (1997). The family environment scale: Reliability
Psychiatry, 29, 698–701. and normative data for an adolescent sample. Family
Bigras, M., LaFrenier, P. J., & Dumas, J. E. (1996). Process, 36, 369–373.
Discriminant validity of the parent-child scales of Braaten, E. (2007). The child clinician’s report writing
the Parenting Stress Index. Early Education and handbook. New York, NY: Guilford Press.
Development, 7, 167–178. Bradley, R., Doolittle, J., & Bartolotta, R. (2008). Building
Bird, H. R., Gould, M. S., & Stagheeza, B. (1992). on the data and adding to the discussion: The experi-
Aggregating data from multiple informants in child ences and outcomes of students with emotional dis-
psychiatry epidemiological research. Journal of turbance. Journal of Behavioral Education, 17, 4–23.
the American Academy of Child and Adolescent Bradshaw, C. P., Mitchell, M. M., O’Brennan, L. M.,
Psychiatry, 31, 78–85. & Leaf, P. J. (2010). Multilevel exploration of fac-
Bird, H. R., Gould, M. S., & Staghezza, B. M. (1993). tors contributing to the overrepresentation of black
Patterns of diagnostic comorbidity in a community students in office disciplinary referrals. Journal of
sample of children aged 9 through 16 years. Journal Educational Psychology, 102, 508–520.
References 397
Brown, M. M., & Grumet, J. G. (2009). School-based Butcher, J. N., Mineka, S., & Hooley, J. M. (2007).
suicide prevention with African American youth in an Abnormal psychology (13th ed.). Boston, MA:
urban setting. Professional Psychology: Research and Pearson Education.
Practice, 40, 111–117. Butcher, J. N., Williams, C. L., Graham, J. R., Archer,
Brown, T. E. (2001). Brown attention-deficit disorder R. P., Tellegen, A., Ben-Porath, Y. S., & Kaemmer, B.
scales for children and adolescents. San Antonio, TX: (1992). MIJPI-A, Minnesota Multiphasic Personality
Psychological Corporation. Inventory-Adolescent: Manual for administra-
Brubaker, R. G., & Szakowski, A. (2000). Parenting prac- tion scoring, and interpretation. Minneapolis, MN:
tices and behavior problems among deaf children. University of Minnesota Press.
Child & Family Behavior Therapy, 22, 13–28. Butcher, J. N., Williams, C. L., Graham, J. R., Tellegen,
Budd, K. S., Heilman, N. E., & Kane, D. (2000). A., Ben-Porath, Y. S., Archer, R. P., & Kaemmer, B.
Psychosocial correlates of child abuse potential in (1992). MMPI-A, Minnesota multiphasic personal-
multiply disadvantaged adolescent mothers. Child ity inventory- adolescent: Manual for administra-
Abuse & Neglect, 24, 611–625. tion, scoring, and interpretation. Minneapolis, MN:
Burgess, A. M., & Nakamura, B. J. (2014). An evalua- University of Minnesota Press.
tion of the two-factor model of emotion: Clinical Byrne, S. P., Lebowitz, E., Ollendick, T. H., & Silverman,
moderators within a large, multi-ethnic sample of W. K. (2018). Anxiety disorders and anxiety-related
youth. Journal of Psychopathology and Behavioral disorders in youth. In J. Hunsley & E. J. Mash (Eds.),
Assessment, 36, 124–135. A guide to assessments that work (2nd ed., pp. 217–
Burke, J., & Loeber, R. (2010). Oppositional defiant 241). New York, NY: Oxford University Press.
disorder and the explanation of the comorbidity Campaign for Mental Health Reform. (2005). A public
between behavioral disorders and depression. Clinical health crisis: Children and adolescents with mental
Psychology: Science and Practice, 17, 319–326. disorders. Congressional briefing. Retrieved from
Burke, J. D., Hipwell, A. E., & Loeber, R. (2010). www.mhreform.org/kids
Dimensions of oppositional defiant disorder as pre- Campbell, S. B. (1986). Developmental issues in childhood
dictors of depression and conduct disorder in girls. anxiety. In R. Gittelman (Ed.), Anxiety disorders of
Journal of the American Academy of Child and childhood (pp. 24–57). New York, NY: Guilford Press.
Adolescent Psychiatry, 49, 484–492. Canino, G., Polanczyk, G., Bauermeister, J. J., Rohde,
Burke, J. D., Loeber, R., Lahey, B. B., & Rathouz, P. J. L. A., & Frick, P. J. (2010). Does the prevalence of CD
(2005). Developmental transitions among affective and ODD vary across cultures? Social Psychiatry and
and behavioral disorders in adolescent boys. Journal Psychiatric Epidemiology, 45, 695–704.
of Child Psychology and Psychiatry, 46, 1200–1210. Cantwell, D. P. (1996). Classification of child and ado-
Burlingame, G. M., Wells, M. G., Cox, J. C., Lambert, lescent psychopathology. Journal of Child Psychology
M. J., Latkowski, M., & Ferre, R. (2005). and Psychiatry, 37(1), 3–12.
Administration and scoring manual for the Y-OQ Cantwell, D. P., Lewinsohn, P. M., Rohde, P., & Seeley,
(Youth Outcome Questionnaire). Stevenson, MD: J. R. (1997). Correspondence between adolescent
American Professional Credentialing Services. report and parent report of psychiatric diagnostic
Burns, R. C., & Kaufman, S. H. (1970). Kinetic family data. Journal of the American Academy of Child and
drawings (KF-D): An introduction to understanding Adolescent Psychiatry, 36, 610–619.
children through kinetic drawings. New York, NY: Capaldi, D. M. (1992). Co-occurrence of conduct prob-
Bruner/ Mazel. lems and depressive symptoms in early adolescent
Burt, S. A. (2012). How do we optimally conceptualize boys: II. A 2-year follow-up at Grade 8. Development
the heterogeneity within antisocial behavior? An argu- and Psychopathology, 4(1), 125–144.
ment for aggressive versus non-aggressive behavioral Cardinale, E. M., & Marsh, A. A. (2017). The reliability and
dimensions. Clinical Psychology Review, 32, 263–279. validity of the Inventory of Callous Unemotional Traits:
Burt, S. A., Rescorla, L. A., Achenbach, T. M., Ivanova, A meta-analytic review. Assessment, 27(1), 57–71.
M. Y., Almqvist, F., Begovac, I., … Verhulst, F. C. Carlson, C. L., & Lahey, B. B. (1983). Factor structure of
(2015). The association between aggressive and non- teacher rating scales for children. School Psychology
aggressive antisocial problems as measured with the Review, 12, 285–292.
Achenbach System of Empirically Based Assessment: Carlson, C. L., Lahey, B. B., & Neeper, R. (1984).
A study of 27,861 parent–adolescent dyads from 25 Peer assessment of the social behavior of accepted,
societies. Personality and Individual Differences, 85, rejected, and neglected children. Journal of Abnormal
86–92. Child Psychology, 12, 189–198.
Buss, A. H. (1995). Personality: Temperament, social Carr, L., Henderson, J., & Nigg, J. T. (2010). Cognitive
behavior, and the self. Boston, MA: Allyn and Bacon. control and attentional selection in adolescents with
Butcher, J. N., Cabiyo, J., Lucio, E., & Garrido, M. (2007). ADHD versus ADD. Journal of Clinical Child and
Assessing Hispanic clients using the MMPI-2 and Adolescent Psychology, 39, 726–740.
MMPI-A. Washington, DC: American Psychological Carrion, V. G., Weems, C. F., Ray, R., & Reiss, A. L.
Association. (2002). Toward an empirical definition of pediatric
398 References
PTSD: The phenomenology of PTSD symptoms in Christensen, A., Margolin, G., & Sullaway, M. (1992).
youth. Journal of the American Academy of Child & Inter-parental agreement on child behavior problems.
Adolescent Psychiatry, 41, 166–173. Psychological Assessment, 4, 419–425.
Carroll, A., Houghton, S., Taylor, M., West, J., & List- Cicchetti, D., Lord, C., Koenig, K., Klin, A., & Volkmar,
Kerz, M. L. (2006). Responses to interpersonal and F. (2008). Reliability of the ADI-R: Multiple exam-
physically provoking situations: The utility and appli- iners evaluate a single case. Journal of Autism and
cation of an observation schedule for school-aged stu- Developmental Disorders, 38, 764–770.
dents with and without attention deficit/hyperactivity Cicchetti, D., & Rogosch, F. A. (2002). A developmental
disorder. Educational Psychology, 26, 483–498. psychopathology perspective on adolescence. Journal
Carter, C. L., & Dacey, C. M. (1996). Validity of the Beck’s of Consulting and Clinical Psychology, 70, 6–20.
depression inventory, MMPI, and Rorschach in assess- Cicchetti, D., & Toth, S. L. (1998). The development of
ing adolescent depression. Journal of Adolescence, 19, depression in children and adolescents. American
223–231. Psychologist, 53, 221–241.
Castellanos, F. X., Lee, P. P., Sharp, W., Jeffries, N. O., Cicchetti, D., & Walker, E. F. (2001). Stress and devel-
Greenstein, D. K., Clasen, L. S., et al. (2002). opment: Biological and psychological consequences.
Developmental trajectories of brain volume abnor- Development and Psychopathology, 13, 413–418.
malities in children and adolescents with attention Ciucci, E., Baroncelli, A., Franchi, M., Golmaryami,
deficit hyperactivity disorder. Journal of the American F. N., & Frick, P. J. (2014). The association between
Medical Association, 288, 1740–1748. callous-unemotional traits and behavioral and aca-
Center for Behavioral Health Statistics and Quality. demic adjustment in children: Further validation of
(2017). 2016 National Survey on Drug Use and the Inventory of Callous-Unemotional Traits. Journal
Health: Methodological summary and definitions. of Psychopathology and Behavioral Assessment, 36,
Rockville, MD: Substance Abuse and Mental Health 189–200.
Services Administration. Clark, C., Haines, M. M., Head, J., Klineberg, E.,
Centers for Disease Control and Prevention (CDC). Arephin, M., Vilner, R., … Stansfeld, S. A. (2007).
(2013). Mental health surveillance among children – Psychological symptoms and physical health and
United States, 2005—2011. (Supplements: 62, 02; health behaviours in adolescents: A prospective 2-year
1–35). study in East London. Addiction, 102, 126–135.
Chafouleas, S. M., Riley-Tillman, T. C., & Christ, T. J. Clark, J. E., & Frick, P. J. (2018). Positive parenting
(2009). Direct Behavior Rating (DBR): An emerg- and callous-unemotional traits: Their association
ing method for assessing social behavior within a with school behavior problems in young children.
tiered intervention system. Assessment for Effective Journal of Clinical Child and Adolescent Psychology,
Intervention, 34, 195–200. 47(Suppl 1), S242–S254.
Chandler, L. A. (1990). The projective hypothesis and the Clerkin, S. M., Marks, D. J., Policaro, K., & Halperin,
development of projective techniques for children. In J. M. (2007). Psychometric properties of the Alabama
C. R. Reynolds & R. W. Kamphaus (Eds.), Handbook Parenting Questionnaire-Preschool Revision. Journal
of psychological and educational assessment of chil- of Clinical Child and Adolescent Psychology, 36,
dren: Personality, behavior, & context (pp. 55–69). 19–28.
New York, NY: Guilford Press. Coie, J. D., & Dodge, K. A. (1983). Continuities and
Chartier, M., Vander Stoep, A. V., McCauley, E., Herting, changes in children’s social status: A five year longi-
J. R., Tracy, M., & Lymp, J. (2008). Passive versus tudinal study. Merrill-Palmer Quarterly, 29, 261–282.
active parental permission: Implications for the ability Coie, J. D., Dodge, K. A., & Coppotelli, H. (1982).
of school-based depression screening to reach youth at Dimensions and types of social status: A cross-age
risk. Journal of School Health, 78, 157–164. perspective. Developmental Psychology, 18, 557–570.
Chia, A., Assan, B., Finch, E., Stargatt, R., Burchell, P., Coie, J. D., & Kupersmidt, J. B. (1983). A behavioral
Jones, H., & Heywood-Smith, J. (2013). Innovations analysis of emerging social status in boys’ groups.
in practice: Effectiveness of specialist adolescent Child Development, 54, 1400–1416.
outreach service for at-risk adolescents. Child and Colins, O. F., Andershed, H., Frogner, L., Lopez-Romero,
Adolescent Mental Health, 18, 116–119. L., Veen, V., & Andershed, A. K. (2014). A new mea-
Chowdhury, M., Aman, M. G., Lecavalier, L., Smith, T., sure to assess psychopathic personality in children:
Johnson, C., Swiezy, N., … Scahill, L. (2016). Factor The Child Problematic Traits Inventory. Journal of
structure and psychometric properties of the revised Psychopathology and Behavioral Assessment, 36(1),
Home Situations Questionnaire for autism spectrum 4–21.
disorder: The Home Situations Questionnaire-Autism Conners, C. K. (1970). Symptom patterns in hyperkinetic,
Spectrum Disorder. Autism, 20, 528–537. neurotic, and normal children. Child Development, 41,
Christ, T. J., Riley-Tillman, T. C., & Chafouleas, S. M. 667–682.
(2009). Foundation for the development and use Conners, C. K. (1995). Conners continuous performance
of Direct Behavior Rating (DBR) to assess and test. Toronto: Multi-Health Systems.
evaluate student behavior. Assessment for Effective Conners, C. K. (2008a). Conners, 3rd edition. Toronto,
Intervention, 34(4), 201–213. Canada: Multi-Health Systems.
References 399
Conners, C. K. (2008b). Conners Comprehensive association between parenting and child conduct prob-
Behavior Rating Scales. Toronto, Canada: Multi- lems. Child Psychiatry and Human Development, 46,
Health Systems. 967–980.
Conners, K. (2014). Conners (3rd ed.). Toronto, Canada: Csorba, J., Rozsa, S., vetro, A., Gadoros, J., Makra, J.,
Multi-Health Systems. Somogyi, E., … Kapornay, K. (2001). Family and
Connor, K., Davidson, J., Chirchill, E., Sherwood, A., school related stresses in depressed Hungarian chil-
Foa, E., & Weisler, R. (2000). Psychometric proper- dren. European Psychiatry, 16, 18–26.
ties of the Social Phobia Inventory (SPIN)—A new Cummings, E. M., Keller, P. S., & Davies, P. T. (2005).
self rating scale. British Journal of Psychiatry, 176, Towards a family process model of maternal and
379–386. paternal depressive symptoms: Exploring multiple
Conoley, J. C., & Conoley, C. W. (1991). Collaboration relations with child and family functioning. Journal of
for child adjustment: Issues for school and clinic- Child Psychology and Psychiatry, 46, 479–489.
based child psychologists. Journal of Consulting and Cummings, J. A. (1986). Projective drawings. In H. M.
Clinical Psychology, S9, 821–829. Knoll (Ed.), The assessment of child and adolescent
Constantino, G., & Malgady, R. G. (1983). Verbal fluency personality (pp. 199–244). New York, NY: Guilford
of Hispanic, Black, & White children on TAT and Press.
TEMAS, a new thematic apperception test. Hispanic Cunningham, C. E., & Boyle, M. H. (2002). Preschoolers
Journal of Behavioral Sciences, 5, 199–206. at risk for attention-deficit hyperactivity disorder and
Constantino, G., Malgady, R. G., & Rogler, L. H. (1988). oppositional defiant disorder: Family, parenting, and
Technical manual: TEMAS thematic apperception test. behavioral correlates. Journal of Abnormal Child
Los Angeles, CA: Western Psychological Services. Psychology, 30, 555–569.
Costa, N. M., & Weems, C. W. (2005). Maternal and child Cunningham, N. R., & Ollendick, T. H. (2010).
anxiety: Do attachment beliefs or children’s percep- Comorbidity of anxiety and conduct problems in chil-
tions of maternal control mediate their association? dren: Implications for clinical research and practice.
Social Development, 14, 574–590. Clinical Child and Family Psychology Review, 13,
Costa, P. T., & McCrae, R. R. (2005). NEO Five-Factor 333–347.
Inventory-3. New York, NY: PAR. Curry, J. F., & Craighead, W. E. (1993). Depression.
Costello, E. J., Egger, H., & Angold, A. (2005). 10-year In T. H. Ollendick & M. Hersen (Eds.), Handbook
research update review: The epidemiology of child and of child and adolescent assessment (pp. 251–268).
adolescent psychiatric disorders: I. Methods and pub- Boston, MA: Allyn and Bacon.
lic health burden. Journal of the American Academy of Cuthbert, B., & Insel, T. (2010). The data of diagno-
Child and Adolescent Psychiatry, 44, 972–986. sis: New approaches to psychiatric classification.
Costello, E. J., Mustillo, S., Erkanli, A., Keeler, G., & Psychiatry: Interpersonal and Biological Processes,
Angold, A. (2003). Prevalence and development of 73, 311–314.
psychiatric disorders in childhood and adolescence. Cytryn, L., & McKnew, D. H. (1974). Factors influenc-
Archives of General Psychiatry, 60, 837–844. ing the changing clinical expression of the depressive
Courchesne, E., Karns, C. M., Davis, B. S., Ziccardi, process in children. American Journal of Psychiatry,
R., Carper, R. A., Tigue, Z. D., … Courchesne, R. Y. 131, 879–881.
(2001). Unusual brain growth patterns in early life D’Angelo, E. J., & Augenstein, T. M. (2012).
in patients with autistic disorder: An MRI study. Developmentally informed evaluation of depression:
Neurology, 57, 245–254. Evidence-based instruments. Child and Adolescent
Cowen, E. L., Dorr, D., Clarfield, S. P., Kreling, B., Psychiatric Clinics of North America, 21, 279–298.
McWilliams, S. A., Pokracki, F., … Wilson, A. B. Dadds, M. R., Barrett, P. M., Rapee, R. M., & Ryan, S.
(1973). The AML: A quick screening device for (1996). Family process and child anxiety and aggres-
early identification of school maladaption. American sion: An observational analysis. Journal of Abnormal
Journal of Community Psychology, 1, 12–35. Child Psychology, 24, 715–734.
Crandal, B. R., Foster, S. L., Chapman, J. E., Cunningham, Dadds, M. R., Maujean, A., & Fraser, J. A. (2003).
P. B., Brennan, P. A., & Whitmore, E. A. (2015). Parenting and conduct problems in children: Australian
Therapist perception of treatment outcome: Evaluating data and psychometric properties of the Alabama
treatment outcomes among youth with antisocial Parenting Questionnaire. Australian Psychologist, 38,
behavior problems. Psychological Assessment, 27, 238–241.
710–725. Dana, J., Dawes, R., & Peterson, N. (2013). Belief in the
Cronbach, L. J. (1960). Essentials of psychological testing unstructured interview: The persistence of an illusion.
(2nd ed.). New York, NY: Harper & Row. Judgment and Decision Making, 8, 512–520.
Crooks, C. V., & Wolfe, D. A. (2007). Child abuse Dandreaux, D. M., & Frick, P. J. (2009). Developmental
and neglect. In E. J. Mash & R. A. Barkley (Eds.), pathways to conduct problems: A further test of the
Assessment of childhood disorders (4th ed., pp. 639– childhood and adolescent-onset distinction. Journal of
684). New York, NY: Guilford Press. Abnormal Child Psychology, 37, 375–385.
Crum, K. I., Waschbusch, D. A., Bagner, D. M., & Coxe, Danforth, J. S., Connor, D. F., & Doerfler, L. A. (2016).
S. (2015). Effects of callous-unemotional traits on the The development of comorbid conduct problems in
400 References
children with ADHD: An example of an integrative A neurobiologically and behaviorally distinct disor-
developmental psychopathology perspective. Journal der from attention-deficit/hyperactivity disorder (with
of Attention Disorders, 20, 214–229. hyperactivity). Development and Psychopathology,
Danforth, J. S., & DuPaul, G. J. (1996). Intcrrater reli- 17, 807–825.
ability of teacher rating scales for children with Dickey, W. C., & Blumberg, S. J. (2004). Revisiting the
Attention-Deficit Hyperactivity Disorder. Journal factor structure of the strengths and difficulties ques-
of Psychopathology and Behavioral Assessment, 18, tionnaire: United States, 2001. Journal of the American
227–238. Academy of Child and Adolescent Psychiatry, 43,
Darling, N., & Steinberg, L. (1993). Parenting style as 1159–1167.
context: An integrative model. Psychological Bulletin, Ding, C. S., & Hershberger, S. L. (2002). Assessing con-
113, 487–496. tent validity and content equivalence using structural
David-Ferdon, C., & Kaslow, N. J. (2008). Evidence- equation modeling. Structural Equation Modeling, 9,
based psychosocial treatments for child and adolescent 283–297.
depression. Journal of Clinical Child and Adolescent Dingle, K., Alati, R., Williams, G. M., Najman, J. M.,
Psychology, 37, 62–104. & Bor, W. (2010). The ability of YSR DSM-oriented
Davies, P. T., Martin, M. J., & Cummings, E. M. (2018). depression scales to predict DSM-IV depression
Interparental conflict and children’s social problems: in young adults: A longitudinal study. Journal of
Insecurity and friendship affiliation as cascading Affective Disorders, 121, 45–51.
mediators. Developmental Psychology, 54, 83–97. DiStefano, C., Greer, F. W., & Dowdy, E. (2019).
Dawkins, T., Meyer, A. T., & Van Bourgondien, M. E. Examining the BASC-3 BESS Parent Form–
(2016). The relationship between the Childhood Preschool using Rasch methodology. Assessment, 26,
Autism Rating Scale: Second Edition and clinical 1162–1175.
diagnosis utilizing the DSM-IV-TR and the DSM-5. DiStefano, C., & Morgan, G. (2011). Examining classifi-
Journal of Autism and Developmental Disorders, 46, cation criteria: A comparison of three cut score meth-
1–8. ods. Psychological Assessment, 23, 354–363.
de Bildt, A., Sytema, S., Ketelaars, C., Kraijer, D., DiStefano, C. A., & Kamphaus, R. W. (2007).
Mulder, E., Volkmar, F., & Minderaa, R. (2004). Development and validation of a behavioral screening
Interrelationships between the ADOS, ADI–R, and for preschool-age children. Journal of Emotional and
DSM–IV–TR classification in children and adoles- Behavioral Disorders, 15, 93–102.
cents with mental retardation. Journal of Autism and Dodge, K. A., Coie, J. D., & Brakke, N. P. (1982).
Developmental Disorders, 34, 129–137. Behavior patterns of socially rejected and neglected
De Los Reyes, A. (2011). Introduction to the special preadolescents: The roles of social approach and
section: More than measurement error: Discovering aggression. Journal of Abnormal Child Psychology,
meaning behind informant discrepancies in clinical 10(3), 389–409.
assessments of children and adolescents. Journal of Dodge, K. A., McClaskey, C. L., & Feldman, E. (1985).
Clinical Child & Adolescent Psychology, 40(1), 1–9. Situational approach to the assessment of social
De Los Reyes, A., & Kazdin, A. E. (2005). Informant competence in children. Journal of Consulting and
discrepancies in the assessment of childhood psycho- Clinical Psychology, 53, 344–353.
pathology: A critical review, theoretical framework, Dodge, K. A., & Pettit, G. S. (2003). A biopsychosocial
and recommendations for further study. Psychological model of the development of chronic conduct prob-
Bulletin, 131, 483–509. lems in adolescence. Developmental Psychology, 32,
De Los Reyes, A., & Kazdin, A. E. (2006). Informant 1065–1072.
discrepancies in assessing child dysfunction relate to Dodge, K. A., & Pettit, G. S. (2003). A biopsychosocial
dysfunction within mother-child interactions. Journal model of the development of chronic conduct prob-
of Child and Family Studies, 15, 645–663. lems in adolescence. Developmental Psychology, 39,
Derefinko, K. J., Adams, Z. W., Milich, R., Fillmore, 349–371.
M. T., Lorch, E. P., & Lynam, D. R. (2008). Response Dodge, K. A. (1983). Behavioral antecedents of peer
style differences in the inattentive and combined social status. Child Development, 54(6), 1386.
subtypes of attention-deficit/hyperactivity disorder. Doleys, D. M. (1977). Behavioral treatments for nocturnal
Journal of Abnormal Child Psychology, 36, 745–758. enuresis in children: A review of the recent literature.
Derogatis, L. R. (1993). Brief symptom inventory (BSI). Psychological Bulletin, 84(1), 30–54.
San Antonio, TX: Pearson Publishing. Dopke, C. A., Lundahl, B. W., Dunsterville, E., &
Dever, B. V., Mays, K. L., Dowdy, E., & Kamphaus, R. W. Lovejoy, M. C. (2003). Interpretations of child com-
(2012). The factor structure of the BASC-2 behavioral pliance in individuals at high- and low-risk for child
and emotional screening system teacher form, child/ physical abuse. Child Abuse & Neglect, 27, 285–302.
adolescent. Journal of Psychoeducational Assessment, Dowdy, E., Doane, K., Eklund, K., & Dever, B. V. (2013).
30, 488–495. A comparison of teacher nomination and screening to
Diamond, A. (2005). Attention-deficit disorder (attention- identify behavioral and emotional risk within a sample
deficit/ hyperactivity disorder without hyperactivity):
References 401
of underrepresented students. Journal of Emotional ture of the child behavior checklist for clinician report.
and Behavioral Disorders, 21(2), 127–137. Journal of Clinical Psychology, 60, 65–85.
Dowdy, E., Furlong, M. J., & Sharkey, J. D. (2013). Using Eaton, D. K., Lowry, R., Brener, N. D., Grunbaum, J. A.,
surveillance of mental health to increase understand- & Kann, L. (2004). Passive versus active parental per-
ing of youth involvement in high risk behaviors: mission in school-based survey research: Does the
A value added analysis. Journal of Emotional and type of permission affect prevalence estimates of risk
Behavioral Disorders, 21(1), 33–44. behaviors? Evaluation Review, 28, 564–577.
Dowdy, E., Harrell-Williams, L., Dever, B. V., Furlong, Edelbrock, C. (1979). Empirical classification of chil-
M. J., Moore, S., Raines, T., & Kamphaus, R. W. dren’s behavior disorders: Progress based on parent
(2016). Predictive validity of a student self-report and teacher ratings. School Psychology Digest, 8,
screener of behavioral and emotional risk in an urban 355–369.
high school. School Psychology Review, 45, 458–476. Edelbrock, C., & Bohnert, A. (2009). Structured inter-
Dowdy, E., Twyford, J. M., Chin, J. K., DiStefano, C. A., views for children and adolescents. In G. Goldstein &
Kamphaus, R. W., & Mays, K. L. (2011). Factor M. Hersen (Eds.), Handbook of psychological assess-
structure of the BASC-2 Behavioral and Emotional ment (3rd edition). Oxford, UK: Elsevier.
Screening System student form. Psychological Edelbrock, C., Costello, A. J., Dulcan, M. K., Conover,
Assessment, 23, 379–387. N. C., & Kalas, R. (1986). Parent child agreement
Drabick, D. A. G., & Kendall, P. C. (2010). Developmental on child psychiatric symptoms assessed via struc-
psychopathology and the diagnosis of mental health tured interview. Journal of Child Psychology and
problems among youth. Clinical Psychology: Science Psychiatry, 27, 181–190.
and Practice, 17, 272–280. Edelbrock, C., Costello, A. J., Dulcan, M. K., Kalas, R.,
Drummond, T. (1994). The student risk screening scale & Conover, N. C. (1985). Age differences in the reli-
(SRSS). Grants Pass, OR: Josephine County Mental ability of the psychiatric interview of the child. Child
Health Program. Development, 56, 265–275.
DuBois, P. H. (1970). A history of psychological testing. Edlavitch, S. A., & Byrns, P. J. (2014). Primary preven-
Boston, MA: Allyn and Bacon. tion research in suicide. Crisis: The Journal of Crisis
Duncan, L., Georgiades, K., Wang, L., Van Lieshout, Intervention and Suicide Prevention, 35, 69–73.
R. J., MacMillan, H. L., Ferro, M. A., … Boyle, Ehrlich, K. B., Cassidy, J., & Dykas, M. J. (2011).
M. H. (2018). Psychometric evaluation of the Reporter discrepancies among parents, adolescents,
Mini International Neuropsychiatric Interview for and peers: Adolescent attachment and informant
Children and Adolescents (MINI-KID). Psychological depressive symptoms as explanatory factors. Child
Assessment, 30, 916–928. Development, 82, 999–1012.
DuPaul, G. J., Anastopolous, A. D., Shelton, T. L., Eiraldi, R. B., Power, T. J., Karustis, J. L., & Goldstein,
Guevremont, I. C., & Metevia, L. (1992). Multimethod S. G. (2000). Assessing ADHD and comorbid disor-
assessment of attention-deficit hyperactivity disorder: ders in children: The Child Behavior Checklist and
The diagnostic utility of clinic-based tests. Journal of the Devereux Scales of Mental Disorders. Journal of
Clinical Child Psychology, 21(394), 402. Clinical Child Psychology, 29, 3–16.
DuPaul, G. J., & Barkley, R. A. (1992). Situational vari- Eisen, A. R., Sussman, J. M., Schmidt, T., Mason, L.,
ability of attention problems: Psychometric properties Hausler, L. A., & Hashim, R. (2011). Separation anxi-
of the revised home and school situations question- ety disorder. In M. Crozier, S. Gillihan, & M. Powers
naires. Journal of Clinical Child Psychology, 21, (Eds.), Handbook of child and adolescent anxiety
178–188. disorders (pp. 245–259). New York, NY: Springer
DuPaul, G. J., Eckert, T. L., & Vilardo, B. (2012). The New York.
effects of school-based interventions for attention Eisenstadt, T. H., Eyberg, S., McNeil, C. B., Newcomb,
deficit hyperactivity disorder: A meta-analysis. School K., & Funderbunk, B. (1993). Parent-child interac-
Psychology Review, 41, 387–412. tion therapy with behavior problem children: Relative
DuPaul, G. J., Power, T. J., Anastopolous, A. D., & Reid, effectiveness of two stages and overall treatment
R. (1998). ADHD-IV rating scale: Checklists, norms, outcome. Journal of Clinical Child Psychology, 22,
and clinical interpretation. New York: Guilford. 42–51.
DuPaul, G. J., Reid, R., Anastopoulos, A. D., & Power, Eklund, K., Renshaw, T. L., Dowdy, E., Jimerson, S. R.,
T. J. (2014). Assessing ADHD symptomatic behav- Hart, S. R., Jones, C. N., & Earhart, J. (2009). Early
iors and functional impairment in school settings: identification of behavioral and emotional problems
Impact of student and teacher characteristics. School in youth: Universal screening versus teacher-referral
Psychology Quarterly, 29, 409–421. identification. California School Psychologist, 14,
DuPaul, G. J., & Stoner, G. (1994). ADHD in the schools: 89–95.
Assessment and intervention strategies. New York: Elgar, F. J., Waschbusch, D. A., Dadds, M. R., &
Guilford. Sivaldason, N. (2007). Development and validation of
Dutra, L., Campbell, L., & Westen, D. (2004). Quantifying a short form of the Alabama Parenting Questionnaire.
clinical judgment in the assessment of adolescent psy- Journal of Child and Family Studies, 16, 243–259.
chopathology: Reliability, validity, and factor struc-
402 References
Elkins, R. K., Kassenboehmer, S. C., & Schurer, S. (2017). dren and adolescents (2nd ed.). New York, NY: John
The stability of personality traits in adolescence and Wiley and Sons.
young adulthood. Journal of Economic Psychology, 60, Eyberg, S., & Pincus, D. (1999). Eyberg child behav-
37–52. ior inventory & Sutter-Eyberg student behavior
Elliott, D. S., Huizinga, D., & Ageton, S. S. (1985). inventory-revised: Professional manual. Odessa, FL:
Explaining delinquency and drug use. Newbury Park, Psychological Assessment Resources.
CA: Sage. Eyberg, S. M., Nelson, M. M., Duke, M., & Boggs, S. R.
Erford, B. T., Bardhoshi, G., Haecker, P., Shingari, B., & (2005). Manual for the dyadic parent-child interaction
Schleichter, J. (2018). Selecting assessment instru- coding system (3rd ed.). Available on-line at www.
ments for problem behavior outcome research with PCIT.org
youth. Measurement and Evaluation in Counseling Eyberg, S. M., & Robinson, E. A. (1982). Conduct prob-
and Development, 52(1), 52–68. lem behavior: Standardization of a behavioral rat-
Erickson, M. T. (1998). Etiological factors. In T. H. ing scale with adolescents. Journal of Clinical Child
Ollendick & M. Hersen (Eds.), Handbook of child Psychology, 12, 347–354.
psychopathology (3rd ed., pp. 37–62). New York, NY: Eyberg, S. M., & Ross, A. W. (1978). Assessment of child
Plenum Press. behavior problems: The validation of a new inventory.
Erikson, E. H. (1968). Identity, youth, and crisis. Journal of Clinical Child Psychology, 7, 113–116.
New York, NY: Norton. Eyde, L. D., Robertson, G. J., Krug, S. E., Moreland,
Eriksson, M. A., Westerlund, J., Hedvall, A., Amark, P., K. L., Robertson, A. G., Shewan, C. M., … Primoff,
Gillberg, C., & Fernell, E. (2013). Medical conditions E. S. (1993). Responsible test use: Case studies
affect the outcome of early interventions in preschool for assessing human behavior. Washington, DC:
children with autism spectrum disorders. European American Psychological Association.
Child & Adolescent Psychiatry, 22, 23–33. Eysenck, H. J. (1965). Review of the Wittenborn
Esbensen, F. A., Deschenes, E. P., Vogel, R. E., West, J., Psychiatric Rating Scales. In O. K. Buros (Ed.),
Arboit, K., & Harris, L. (1996). Active parental con- Mental measurements yearbook (5th ed.). Highland
sent in school-based research: An examination of ethi- Park, NJ: Gryphon Press.
cal and methodological issues. Evaluation Review, 20, Ezpeleta, L., & Granero, R. (2015). Executive functions
737–753. in preschoolers with ADHD, ODD, and co-morbid
Essau, C. A., Sasagawa, S., & Frick, P. J. (2006). ADHD-ODD: Evidence from ecological and perfor-
Psychometric properties of the Alabama Parenting mance-based measures. Journal of Neuropsychology,
Questionnaire. Journal of Child and Family Studies, 9, 258–270.
15, 597–616. Ezpeleta, L., Osa, N. D. L., Granero, R., Penelo, E., &
Evans, S. W., Owens, J. S., & Bunford, N. (2014). Domènech, J. M. (2013). Inventory of callous-unemo-
Evidence-based psychosocial treatments for children tional traits in a community sample of preschoolers.
and adolescents with Attention-Deficit/Hyperactivity Journal of Clinical Child & Adolescent Psychology,
Disorder. Journal of Clinical Child and Adolescent 42(1), 91–105.
Psychology, 43, 527–551. Falkmer, T., Anderson, K., Falkmer, M., & Horlin, C.
Exner, J. E. (1974). The Rorschach: A comprehensive sys- (2013). Diagnostic procedures in autism spectrum
tem. New York, NY: Wiley. disorders: A systematic literature review. European
Exner, J. E. (1978). The Rorschach: A comprehensive sys- Journal of Child and Adolescent Psychiatry, 22,
tem, II: Recent research and advanced interpretation. 329–340.
New York, NY: Wiley. Fantuzzo, J., Bulotsky, R., McDermott, P., Mosca, S., &
Exner, J. E. (1983, June). The depression index. 1983 Lutz, M. (2003). A multivariate analysis of emotional
Alumni Newsletter (p. 7). Bayville, NY: Rorschach and behavioral adjustment and preschool educational
Workshops. outcomes. School Psychology Review, 32, 185–203.
Exner, J. E. (1990, April). The depression index. 1990 Feeney-Kettler, K. A., Kratochwill, T. R., Kaiser, A. P.,
Alumni Newsletter (pp. 12–13). Asheville, NC: Henneter, M. L., & Kettler, R. J. (2010). Screening
Rorschach Workshops. young children’s risk for mental health problems: A
Exner, J. E., & Martin, L. S. (1983). The Rorschach: A review of four measures. Assessment for Effective
history and description of the comprehensive system. Intervention, 35, 218–230.
School Psychology Review, 12, 407–413. Ferdinand, R. (2008). Validity of the CBCL/YSR
Exner, J. E., Thomas, E. A., & Mason, B. (1985). DSM-IV scales anxiety problems and affective prob-
Children’s Rorschachs: Description and prediction. lems. Journal of Anxiety Disorders, 22, 126–134.
Journal of Personality Assessment, 49, 13–20. Finch, A. J., & Belter, R. W. (1993). Projective techniques.
Exner, J. E., & Weiner, I. B. (1982). The Rorschach: A In T. H. Ollendick & M. Hersen (Eds.), Handbook
comprehensive system, Vol. 3: Assessment of children of child and adolescent assessment (pp. 229–238).
and adolescents. New York, NY: Wiley. Boston, MA: Allyn and Bacon.
Exner, J. E., & Weiner, I. B. (1994). The Rorschach: First, M. B., Williams, J. B. W., Karg, R. S., & Spitzer,
A comprehensive system, Vol. 3: Assessment of chil- R. L. (2016). The structured clinical interview
References 403
for DSM-5 disorders. Washington, DC: American Freeman, J., Flessner, C. A., & Garcia, A. (2011). The
Psychiatric Association. Children’s Yale-Brown Obsessive Compulsive Scale:
Fischer, M., Barkley, R. A., Fletcher, K., & Smallish, L. Reliability and validity for use among 5–8 year olds
(1993). The adolescent outcome of hyperactive chil- with obsessive-compulsive disorder. Journal of
dren diagnosed by research criteria: V. Predictors of Abnormal Child Psychology, 39, 877–883.
outcome. Journal of the American Academy of Child French, J. L., & Hale, R. L. (1990). A history of the devel-
and Adolescent Psychiatry, 32, 324–332. opment of psychological and educational testing. In
Fisher, P., Wicks, J., Shaffer, D., Piacentini, J., & Lapkin, C. R. Reynolds & R. W. Kamphaus (Eds.), Handbook
J. (1992). NIMH diagnostic interview schedule for of psychological and educational assessment of
children: Users manual. New York, NY: New York children: Intelligence and achievement (pp. 3–28).
State Psychiatric Institute. New York, NY: Guilford Press.
Flanagan, K. S., Bierman, K. L., & Kam, C. (2003). Freud, S. (1936). The problem of anxiety. New York, NY:
Identifying at-risk children at school entry: The use- W. W. Norton & Co.
fulness of multibehavioral problem profiles. Journal Frick, P. J. (1993). Childhood conduct problems in family
of Clinical Child and Adolescent Psychology, 32(3), context. School Psychology Review, 22, 376–385.
396–407. Frick, P. J. (1994). Family dysfunction and the disruptive
Flanagan, R., & DiGiuseppe, R. (1999). Critical review behavior disorders: A review of recent empirical find-
of the TEMAS: A step within the development of the- ings. In T. H. Ollendick & R. J. Prinz (Eds.), Advances
matic apperception instruments. Psychology in the in clinical child psychology (Vol. 16, pp. 213–226).
Schools, 36, 21–30. New York, NY: Plenum.
Fletcher-Watson, S., & McConachie, H. (2017). The Frick, P. J. (2000). Laboratory and performance-based
search for an early intervention outcome measure- measures of childhood disorders: Introduction to the
ment tool in autism. Focus on Autism and Other special section. Journal of Clinical Child Psychology,
Developmental Disabilities, 32, 71–80. 29(4), 475–478.
Foley, D. L., Rutter, M., Angold, A., Pickles, A., Maes, Frick, P. J. (2004). Integrating research on tempera-
H. M., Silberg, J. L., & Eaves, L. J. (2005). Making ment and childhood psychopathology: Its pitfalls and
sense of informant disagreement for overanxious dis- promise. Journal of Clinical Child and Adolescent
order. Journal of Anxiety Disorders, 19(2), 193–210. Psychology, 33(1), 2–7.
Forehand, R., Long, N., & Hedrick, M. (1987). Family Frick, P. J. (2012). Developmental pathways to conduct
characteristics of adolescents who display overt disorder: Implications for future directions in research,
and covert behavior problems. Journal of Behavior assessment, and treatment. Journal of Clinical Child
Therapy and Experimental Psychiatry, 18, 325–328. and Adolescent Psychology, 41, 378–389.
Forness, S. R., Cluett, S. E., Ramey, C. T., Ramey, S. L., Frick, P. J. (2016). Current research on conduct disorder
Zima, B. T., Hsu, C., … MacMillan, D. L. (1998). in children and adolescents. South Africa Journal of
Special education identification of Head Start chil- Psychology, 46, 160–174.
dren with emotional and behavioral disorders in Frick, P. J., Bodin, S. D., & Barry, C. T. (2000).
second grade. Journal of Emotional and Behavioral Psychopathic traits and conduct problems in commu-
Disorders, 6(4), 194–204. nity and clinic-referred samples of children: Further
Forness, S. R., & Kritzer, J. (1992). A new proposed defi- development of the Psychopathy Screening Device.
nition and terminology to replace “serious emotional Psychological Assessment, 12, 382–393.
disturbance” in the individuals with disabilities act. Frick, P. J., Christian, R. C., & Wootton, J. M. (1999).
School Psychology Review, 21, 12–20. Age trends in the association between parenting prac-
Forth, A. E., Kosson, D. S., & Hare, R. D. (2003). Hare tices and conduct problems. Behavior Modification,
psychopathy checklist: Youth version. Toronto: Multi- 23, 106–128.
Health Systems. Frick, P. J., & Hare, R. D. (2001). The Antisocial Process
Fowler, P. C. (1982). Factor structure of the family envi- Screening Device. Toronto: Multi-Health Systems.
ronment scale: Effects of social desirability. Journal of Frick, P. J., & Jackson, Y. K. (1993). Family functioning and
Clinical Psychology, 38(2), 285–292. childhood antisocial behavior. Yet another reinterpreta-
Fraser, A., Cooper, M., Agha, S. S., Collishaw, S., Rice, tion. Journal of Clinical Child Psychology, 22, 410–419.
F., Thapar, A., & Eyre, O. (2017). The presentation Frick, P. J., Kamphaus, R. W., Lahey, B. B., Loeber, R.,
of depression symptoms in attention-deficit/hyperac- Christ, M. A. G., Hart, E. L., & Tannenbaum, L. E.
tivity disorder: Comparing child and parent reports. (1991). Academic underachievement and the disrup-
Child and Adolescent Mental Health, 23, 243–250. tive behavior disorders. Journal of Consulting and
Frazier, T. W., Youngstrom, E. A., Kubu, C. S., Sinclair, Clinical Psychology, 59(2), 289–294.
L., & Rezai, A. (2008). Exploratory and confirmatory Frick, P. J., & Kimonis, E. (2008). Externalizing dis-
factor analysis of the Autism Diagnostic Interview- orders. In J. E. Maddux & B. A. Winstead (Eds.),
Revised. Journal of Autism and Developmental Psychopathology: Contemporary issues, theory, and
Disorders, 38, 474–480. research (pp. 349–374). Mahwah, NJ: Erlbaum.
404 References
Frick, P. J., Kimonis, E. R., Dandreaux, D. M., & Farell, and association with maternal anxiety. Psychological
J. M. (2003). The 4 year stability of psychopathic traits Assessment, 6, 372–379.
in non-referred youth. Behavioral Sciences & the Law, Frick, P. J., & Thornton, L. C. (2017). A brief history of
21(6), 713–736. the diagnostic classification of childhood externaliz-
Frick, P. J., & Lahey, B. B. (1991). The nature and char- ing disorders. In L. Centifanti & D. Williams (Eds.),
acteristics of Attention-deficit Hyperactivity Disorder. The Wiley Blackwell handbook of developmental
School Psychology Review, 20, 163–173. psychopathology (pp. 475–495). Oxford, UK: Wiley
Frick, P. J., Lahey, B. B., Applegate, B., Kerdyck, L., Blackwell.
Ollendick, T., Hynd, G. W., … Waldman, I. (1994). Frick, P. J., & Viding, E. M. (2009). Antisocial behavior
DSM-IV field trials for the disruptive behavior dis- from a developmental psychopathology perspective.
orders: Symptom utility estimates. Journal of the Development and Psychopathology, 21, 1111–1131.
American Academy of Child & Adolescent Psychiatry, Friedman, R. M., Katz-Leavy, J., Manderscheid, R., &
33(4), 529–539. Sondheimer, D. (1996). Prevalence of serious emo-
Frick, P. J., Lahey, B. B., Loeber, R., Tannenbaum, tional disturbance in children and adolescents. In R. W.
L. E., Van Horn, Y., Christ, M. A. C., et al. (1993). Manderscheid & M. A. Sonnenschein (Eds.), Mental
Oppositional defiant disorder and conduct disorder: A health, United States, 1996 (pp. 71–88). Rockville,
meta-analytic review of factor analyses and cross-vali- MD: Center for Mental Health Services.
dation in a clinic sample. Clinical Psychology Review, Fuchs, D., & Fuchs, L. S. (1986). Test procedure bias: A
13, 319–340. meta-analysis of examiner familiarity effects. Review
Frick, P. J., Lilienfeld, S. O., Ellis, M. L., Loney, B. R., & of Educational Research, 56, 243–262.
Silverthorn, P. (1999). The association between anxi- Furlong, M., Dowdy, E., Carnazzo, K., Bovery, B. L.,
ety and psychopathy dimensions in children. Journal & Kim, E. (2014). Covitality: Fostering the building
of Abnormal Child Psychology, 27, 381–390. blocks of complete mental health. Communiqué, 8, 1.
Frick, P. J., & Loney, B. R. (1999). Outcomes of children Furr, R. M. (2018). Psychometrics: An introduction (3rd
and adolescents with conduct disorder and opposi- ed.). Los Angeles, CA: Sage.
tional defiant disorder. In H. C. Quay & A. Hogan Gallucci, N. T. (1989). Personality assessment with chil-
(Eds.), Handbook of disruptive behavior disorders dren of superior intelligence: Divergence versus psy-
(pp. 507–524). New York, NY: Plenum Press. chopathology. Journal of Personality Assessment,
Frick, P. J., & Loney, B. R. (2000). The use of laboratory 53(4), 749–760.
and performance-based measures in the assessment Galton, F. (1884). Measurement of character. Fortnightly
of children and adolescents with conduct disorders. Review, 42, 179–185. Reprinted in L. D. Goodstein
Journal of Clinical Child Psychology, 29, 540–554. & R. I. Lanyon (Eds.), Readings in Personality
Frick, P. J., & Loney, B. R. (2002). Understanding the Assessment, New York, NY: Wiley.
association between parent and child antisocial behav- Garb, H. N., Wood, J. M., & Nezworski, M. T. (2000).
ior. In R. J. McMahon & R. D. Peters (Eds.), The Projective techniques and the detection of child sexual
effects of parental dysfunction on children (pp. 105– abuse. Child Maltreatment, 5, 161–168.
126). New York, NY: Plenum Press. Garber, J. (1984). Classification of childhood psycho-
Frick, P. J., & Morris, A. S. (2004). Temperament and pathology: A developmental perspective. Child
developmental pathways to conduct problems. Development, 55(1), 30.
Journal of Clinical Child and Adolescent Psychology, Gartstein, M. A., Bridgett, D. J., Dishion, T. J., &
33, 54–68. Kaufman, N. K. (2009). Depressed mood and mater-
Frick, P. J., & Nigg, J. T. (2012). Current issues in the nal report of child behavior problems: Another look
diagnosis of attention-deficit hyperactivity disorder, at the depression-distortion hypothesis. Journal of
oppositional defiant disorder, and conduct disorder. Applied Developmental Psychology, 30, 149–160.
Annual Review of Clinical Psychology, 8, 77–107. Geller, B., Fox, L. W., & Fletcher, M. (1993). Effect of
Frick, P. J., & O’Brien, B. S. (1994). Conduct disorders. tricyclic antidepressants on switching to mania and
In R. T. Ammerman & M. Hersen (Eds.), Handbook on the onset of bipolarity in depressed 6- to 12-year-
of child behaviour therapy in the psychiatric setting olds. Journal of the American Academy of Child and
(pp. 199–216). New York, NY: Wiley. Adolescent Psychiatry, 32, 43–50.
Frick, P. J., & Ray, J. V. (2015). Evaluating callous- Georgiades, S., Bishop, S. L., & Frazier, T. (2017).
unemotional traits as a personality construct. Journal Editorial perspective: Longitudinal research in autism:
of Personality, 83(6), 710–722. Introducing the concept of ‘chronogeneity’. Journal of
Frick, P. J., Ray, J. V., Thornton, L. C., & Kahn, R. E. (2014). Child Psychology and Psychiatry, 58, 534–536.
Can callous-unemotional traits enhance the understand- Ghiselli, E. E., Campbell, J. P., & Zedeck, S. (1981).
ing, diagnosis, and treatment of serious conduct prob- Measurement theory for the behavioral sciences. San
lems in children and adolescents? A comprehensive Francisco, CA: W. H. Freeman.
review. Psychological Bulletin, 140, 1–57. Gilliam, J. E. (2014). Gilliam autism rating scale-third
Frick, P. J., Silverthorn, P., & Evans, C. S. (1994). edition (GARS-3). Austin, TX: Pro-Ed.
Assessment of childhood anxiety using structured Gittelman-Klein, R. (1986). Questioning the clinical
interviews: Patterns of agreement among informants usefulness of projective psychological tests for chil-
References 405
dren. Developmental and Behavioral Pediatrics, 7, Gorsuch, R. L. (2015). Factor analysis: Classic edition.
378–382. New York, NY: Routledge.
Glover, T., & Albers, C. (2007). Considerations for evalu- Gottesman, I. I., McGuffin, P., & Farmer, A. E. (1987).
ating universal screening assessments. Journal of Clinical genetics as clues to the “real” genetics of
School Psychology, 45, 117–135. schizophrenia. Schizophrenia Bulletin, 13, 23–47.
Glueck, S., & Glueck, E. (1968). Delinquents and non- Greaves, J. R., & Salloum, A. (2015). Evaluation of a
delinquents in perspective. Cambridge, MA: Harvard youth with sexual behavior problems (YSBP) outpa-
University Press. tient treatment program. Child & Adolescent Social
Goldberg, L. R. (1992). The development of mark- Work Journal, 32, 177–185.
ers for the Big-Five factor structure. Psychological Green, K., Vosk, B., Forehand, R., & Beck, S. (1981).
Assessment, 4(1), 26–42. An examination of differences among sociometrically
Goldenson, R. M. (1984). Longman dictionary of psychol- identified accepted, rejected, and neglected children.
ogy and psychiatry. New York, NY: Longman. Child Study Journal, 11, 117–124.
Goldsmith, H. H., Buss, A. H., Plomin, R., Rothbart, Green, M. (1992). Pediatric diagnosis: Interpretation of
M. K., Thomas, A., Chess, S., … McCall, R. B. (1987). symptoms and signs in infants, children, and adoles-
Roundtable: What is temperament? Four approaches. cents (5th ed.). Philadelphia, PA: Saunders.
Child Development, 58, 509–525. Green, S. M., Loeber, R., & Lahey, B. B. (1991). Stability
Goldsmith, H. H., & Rieser-Danner, L. A. (1990). of mothers’ recall of the age of onset of their child’s
Assessing early temperament. In C. R. Reynolds & attention and hyperactivity problems. Journal of
R. W. Kamphaus (Eds.), Handbook of psychological the American Academy of Child and Adolescent
and educational assessment of children: Personality, Psychiatry, 30, 135–137.
behavior, and context (pp. 245–278). New York, NY: Gregory, A., Skiba, R. J., & Noguera, P. A. (2010). The
The Guilford Press. achievement gap and the discipline gap: Two sides of
Gomez, R., Vance, A., & Gomez, R. M. (2014). Analysis of the same coin? Educational Researcher, 39, 59–68.
the convergent and discriminant validity of the CBCL, Gresham, F. M., & Little, S. G. (1993). Peer-referenced
TRF, and YSR in a clinic-referred sample. Journal of assessment strategies. In T. H. Ollendick & M. Hersen
Abnormal Child Psychology, 42, 1413–1425. (Eds.), Handbook of child and adolescent assessment
Gommans, R., & Cillessen, A. H. N. (2015). Nominating (pp. 165–179). Boston, MA: Allyn and Bacon.
under constraints: A systematic comparison of unlim- Grieten, H., De Haene, L., & Uyttebroek, K. (2007).
ited and limited peer nomination methodologies in ele- Cross-cultural validation of the Child Abuse Potential
mentary school. International Journal of Behavioral Inventory in Belgium (Flanders): Relations with
Development, 39, 77–86. demographic characteristics and parenting problems.
Goodman, A., & Goodman, R. (2009). Strengths and Journal of Family Violence, 22, 223–229.
Difficulties Questionnaire as a dimensional measure of Grills, A. E., & Ollendick, T. H. (2003). Multiple infor-
child mental health. Journal of the American Academy mant agreement and the Anxiety Disorders Interview
of Child & Adolescent Psychiatry, 48, 400–403. Schedule for Parents and Children. Journal of the
Goodman, R. (1997). The Strengths and Difficulties American Academy of Child & Adolescent Psychiatry,
Questionnaire: A research note. Journal of Child 42, 30–40.
Psychology and Psychiatry, 38, 581–586. Groth-Marnat, G. (2009). The five assessment issues you
Goodman, R., & Scott, S. (1999). Comparing the strengths meet when you go to heaven. Journal of Personality
and difficulties questionnaire and the child behavior Assessment, 91, 303–310.
checklist: Is small beautiful? Journal of Abnormal Gutierrez, P. M., & Osman, A. (2009). Getting the best
Child Psychology, 27, 17–24. return on your screening investment: An analysis of
Goodman, S. H., & Gotlib, I. H. (1999). Risk for psy- the suicidal ideation questionnaire and Reynolds ado-
chopathology in the children of depressed mothers: A lescent depression scale. School Psychology Review,
developmental model for understanding mechanisms 38, 200–217.
of transmission. Psychological Review, 106, 458–490. Gutman, L., Sameroff, A., & Cole, R. (2003). Academic
Goodman, S. H., Rouse, M. H., Connell, A. M., Broth, growth curve trajectories from 1st grade to 12th
M. R., Hall, C. M., & Heyward, D. (2011). Maternal grade: Effects of multiple social risk factors and pre-
depression and child psychopathology: A meta-ana- school child factors. Developmental Psychology, 39,
lytic review. Clinical Child and Family Psychology 777–790.
Review, 14, 1–27. Haack, L. M., Villodas, M., McBurnett, K., Hinshaw,
Goodman, W. K., Price, L. H., Rasmussen, S. A., Mazure, S., & Pfiffner, L. J. (2017). Parenting as a mecha-
C., Fleischmann, R. L., Hill, C. L., … Charney, D. S. nism of change in psychosocial treatment for youth
(1989). The Yale–Brown Obsessive–Compulsive with ADHD, Predominantly Inattentive Presentation.
Scale: I. Development, use, and reliability. Archives of Journal of Abnormal Child Psychology, 45, 841–855.
General Psychiatry, 46, 1006–1011. Haak, R. A. (1990). Using the sentence completion to
Gordon, M. (1983). The Gordon diagnostic system. assess emotional disturbance. In C. R. Reynolds &
DeWitt, NY: Gordon Systems. R. W. Kamphaus (Eds.), Handbook of psychological
406 References
and educational assessment of children: Personality, Hart, D. H. (1986). The sentence completion techniques.
behavior, and context (pp. 147–167). New York, NY: In H. M. Knoff (Ed.), The assessment of child and
Guilford Press. adolescent personality (pp. 245–272). New York, NY:
Haak, R. A. (2003). Using the sentence completion to Guilford Press.
assess emotional disturbance. In C. R. Reynolds & Hartmann, D. P., Roper, B. L., & Bradford, D. C. (1979).
R. W. Kamphaus (Eds.), Handbook of psychological Some relationships between behavioral and traditional
and educational assessment of children: Personality, assessment. Journal of Behavioral Assessment, 1, 3–21.
behavior, and context (2nd ed., pp. 159–181). Harvey, E. A., Fischer, C., Weieneth, J. L., Hurwitz,
New York, NY: Guilford. S. D., & Sayer, A. G. (2013). Predictors of discrep-
Hall, C. L., Valentine, A. Z., Groom, M. J., Walker, G. M., ancies between informants’ ratings of preschool-aged
Sayal, K., Daley, D., & Hollis, C. (2016). The clinical children’s behavior: An examination of ethnicity,
utility of the continuous performance test and objec- child characteristics, and family functioning. Early
tive measures of activity for diagnosing and monitor- Childhood Research Quarterly, 28, 668–682.
ing ADHD in children. A systematic review. European Harvey, V. S. (2006). Variables affecting the clarity of
Child & Adolescent Psychiatry, 25, 677–699. reports. Journal of Clinical Psychology, 62, 5–18.
Hamilton, C., Fuchs, D., Fuchs, L. S., & Roberts, H. Haskett, M. E., Ahern, L. S., Ward, C. S., & Allaire, J. C.
(2000). Rates of classroom participation and the valid- (2006). Factor structure and validity of the parenting
ity of sociometry. School Psychology Review, 29, stress index-short form. Journal of Clinical Child and
251–266. Adolescent Psychology, 35, 302–312.
Hamilton, E., & Carr, A. (2016). Systematic review Haskett, M. E., Scott, S. S., & Fann, K. D. (1995). Child
of self-report family assessment measures. Family Abuse Potential Inventory and parenting behavior:
Process, 55, 16–30. Relationships with high-risk correlates. Child Abuse
Hammen, C., Brennan, P. A., Keenan-Miller, D., & Herr, & Neglect, 19, 1483–1495.
N. R. (2008). Early onset recurrent subtype of ado- Hathaway, S. R., & McKinley, J. C. (1940). A multiphasic
lescent depression: Clinical and psychosocial corre- personali ty schedule (Minnesota): I. Construction of
lates. Journal of Child Psychology & Psychiatry, 49, the schedule. Journal of Psychology, 10, 249–254.
433–440. Hathaway, S. R., & McKinley, J. C. (1942). Minnesota
Hammer, E. F. (1958). The clinical application of projec- Multiphasic Personality Inventory. Minneapolis, MN:
tive drawings. Springfield, IL: Thomas. University of Minnesota Press.
Handel, R. W. (2016). An introduction to the Minnesota Hawes, D. J., & Dadds, M. R. (2006). Assessing parent-
Multiphasic Personality Inventory-Adolescent- ing practices through parent-report and direct observa-
Restructured Form (MMPI-A-RF). Journal of Clinical tion during parenting-training. Journal of Child and
Psychology in Medical Settings, 23(4), 361–373. Family Studies, 15, 555–568.
Hankin, B. L., Mermelstein, R., & Roesch, L. (2007). Sex Hayvren, M., & Hymnel, S. (1984). Ethical issues in
differences in adolescent depression: Stress exposure sociometric testing: Impact of sociometric measures
and reactivity. Child Development, 78, 279–295. on interaction behavior. Developmental Psychology,
Hankin, B. L., Wetter, E., & Cheely, C. (2008). Sex differ- 20, 849–884.
ences in child and adolescent depression: A develop- Haz, A. M., & Ramirez, V. (1998). Preliminary validation
mental psychopathological approach. In J. R. Z. Abela of the Child Abuse Potential Inventory in Chile. Child
& B. L. Hankin (Eds.), Handbook of depression in Abuse & Neglect, 22, 869–879.
children and adolescents (pp. 377–414). New York, Helzer, J. E., & Robins, L. N. (1988). The Diagnostic
NY: Guilford Press. Interview Schedule: Its development, evolution, and
Hare, R. D., & Neumann, C. S. (2008). Psychopathy as a use. Social Psychiatry and Psychiatric Epidemiology,
clinical and empirical construct. The Annual Review of 23, 6–16.
Clinical Psychology, 4, 217–246. Hembree-Kigin, T. L., & McNeil, C. B. (1995). Parent-
Harris, F. C., & Lahey, B. B. (1982a). Recording system child interaction therapy. New York, NY: Plenum
bias in direct observational methodology: A review Press.
and critical analysis of factors causing inaccurate cod- Hendren, R. L., DeBacker, I., & Pandina, G. J. (2000).
ing behavior. Clinical Psychology Review, 2, 539–556. Review of neuroimaging studies of child and ado-
Harris, F. C., & Lahey, B. B. (1982b). Subject reactivity lescent psychiatric disorders from the past 10 years.
in direct observational assessment: A review and criti- Journal of the American Academy of Child and
cal analysis. Clinical Psychology Review, 2, 523–538. Adolescent Psychiatry, 39, 815–828.
Harris, K., Collinson, C., & das Nair, R. (2012). Service- Hill, L. G., Lochman, J. E., Coie, J. D., Greenberg, M. T.,
users’ experiences of an early intervention in psy- & The Conduct Problems Prevention Research Group.
chosis service: An interpretative phenomenological (2004). Effectiveness of early screening for external-
analysis. Psychology and Psychotherapy: Theory, izing problems: Issues of screening accuracy and util-
Research and Practice, 85, 456–489. ity. Journal of Consulting and Clinical Psychology,
Hart, D. H. (1972). The Hart sentence completion test 72, 809–820.
for children. Salt Lake City, UT: Educational Support Hiller, J. B., Rosenthal, R., Bornstein, R. F., Berry,
Systems. D. T. R., & Brunell-Neulieb, S. (1999). A compara-
References 407
tive meta-analysis of Rorschach and MMPI validity. Hoza, B., Gerdes, A. C., Mrug, S., Hinshaw, S. P.,
Psychological Assessment, 11, 278–296. Bukowski, W. M., Gold, W. A., … Wigal, T. (2005).
Hinshaw, S. P. (1987). On the distinction between atten- Peer-assessed outcomes in the multimodal treatment
tional deficits/hyperactivity and conduct problems/ study of children with attention deficit hyperactiv-
aggression in child psychopathology. Psychological ity disorder. Journal of Clinical Child & Adolescent
Bulletin, 101(3), 443–463. Psychology, 34(1), 74–86.
Hinshaw, S. P., Owens, E. B., Zalecki, C., Huggins, Huang, C. Y., Costeines, J., Kaufman, J. S., & Ayala, C.
S. P., Montenegro-Nevado, A. J., Schrodek, E., … (2014). Parenting stress, social support, and depres-
Swanson, E. N. (2012). Prospective follow-up of girls sion for ethnic minority adolescent mothers: Impact
with Attention-Deficit/Hyperactivity Disorder into on child development. Journal of Child and Family
early adulthood: Continuing impairment includes Studies, 23, 255–262.
elevated risk for suicide attempts and self-injury. Huesmann, L. R., Eron, L. D., Lefkowitz, M. M., &
Journal of Consulting and Clinical Psychology, 80, Walder, L. O. (1984). Stability of aggression over
1041–1051. time and generations. Developmental Psychology, 20,
Hipwell, A. E., Stepp, S., Feng, X., Burke, J., Battista, 1120–1134.
D. R., Loeber, R., & Keenan, K. (2011). Impact of Hughes, J. (1990). Assessment of social skills: Sociometric
oppositional defiant disorder dimensions on the tem- and behavioral approaches. In C. R. Reynolds &
poral ordering of conduct problems and depression R. W. Kamphaus (Eds.), Handbook of psychological
across childhood and adolescence in girls. Journal of and educational assessment of children: Personality,
Child Psychology and Psychiatry, 52, 1099–1108. behavior, and context (pp. 423–444). New York, NY:
Hoeve, M., Dubas, J. S., Gerris, J. R. M., van der Laan, Guilford Press.
P. H., & Smeenk, W. (2011). Maternal and pater- Hughes, T. L., Gacono, C. B., & Owen, P. F. (2007).
nal parenting styles: Unique and combined links to Current status of Rorschach assessment: Implications
adolescent and early adult delinquency. Journal of for the school psychologist. Psychology in the Schools,
Adolescence, 34, 813–827. 44, 281–291.
Hogendoorn, S. M., Wolters, L. H., de Haan, E., Lindauer, Hunsley, J., & Bailey, J. M. (2001). Whither the
R. J. L., Tillema, A., Vervoot, L., … Prins, P. J. M. Rorschach? An analysis of the evidence. Psychological
(2014). Advancing an understanding of the Anxiety Assessment, 13, 472–485.
Control Questionnaire for Children (AQC-C) in clini- Hunsley, J., & DiGiulio, G. (2001). Norms, norming, and
cally anxious and non-anxious youth: Psychometric clinical assessment. Clinical Psychology: Science and
properties, incremental prediction, and develop- Practice, 8, 378–382.
mental differences. Journal of Psychopathology and Hunsley, J., & Mash, E. J. (2007). Evidence-based assess-
Behavioral Assessment, 36, 288–289. ment. The Annual Review of Clinical Psychology, 3,
Hogendoorn, S. M., Wolters, L. H., Vervoot, L., Prins, P. J., 29–51.
Boer, F., Kooj, E., & de Haan, E. (2010). Measuring neg- Hunsley, J., & Mash, E. J. (2018). A guide to assessments
ative and positive thoughts in children: An adaptation that work (2nd ed.). New York, NY: Oxford Press.
of the Children’s Automatic Thoughts Scale (CATS). Hurley, K. D., Huscroft-D’Angelo, J., Trout, A., Griffith, A.,
Cognitive Therapy and Research, 34, 467–478. & Epstein, M. (2014). Assessing parenting skills and atti-
Hojnoski, R. L., Morrison, R., Brown, M., & Matthews, tudes: A review of the psychometrics of parenting mea-
W. J. (2006). Projective test use among school sures. Journal of Child and Family Studies, 23, 812–823.
psychologists: A survey and critique. Journal of Husky, M. M., Kanter, D. A., McGuire, L., & Offson, M.
Psychoeducational Assessment, 24, 145–159. (2012). Mental health screening of African American
Holaday, M. (2000). Rorschach protocols from children adolescents and facilitated access to care. Community
and adolescents diagnosed with Posttraumatic Stress Mental Health Journal, 48, 71–78.
Disorder. Journal of Personality Assessment, 75(1), Husky, M. M., Sheridan, M., McGuire, L., & Offson, M.
143–157. (2011). Mental health screening and follow-up care in
Holaday, M., Smith, D. A., & Sherry, A. (2000). Sentence public high schools. Journal of the American Academy
completion tests: A survey of the literature and results of Child & Adolescent Psychiatry, 50, 881–891.
of a survey of members of the Society for Personality Hutcheson, J. J., & Black, M. M. (1996). Psychometric
Assessment. Journal of Personality Assessment, 74, properties of the Parenting Stress Index in a sample
371–383. of low-income African-American mother of infants
Holden, E. W., & Banez, G. A. (1996). Child abuse poten- and toddlers. Early Education and Development, 7,
tial and parenting stress within maltreating families. 381–400.
Journal of Family Violence, 11, 1–12. Hutt, M. L. (1980). The Michigan picture test – revised.
Holtzman, W. H., & Swartz, J. D. (2003). Use of the New York, NY: Grune & Stratton.
Holtzman Inkblot Technique (HIT) with children. In Hysing, M., Eigen, I., Gillberg, C., Lie, S. A., &
C. R. Reynolds & R. W. Kamphaus (Eds.), Handbook Lundervold, A. J. (2007). Chronic physical illness and
of psychological and educational assessment in chil- mental health in children: Results from a large-scale
dren: Personality, behavior, and context (2nd ed., population study. Journal of Child Psychology and
pp. 198–2218). New York, NY: Guilford. Psychiatry and Allied Disciplines, 48, 785–792.
408 References
Ikeda, M. J., Neesen, E., & Witt, J. C. (2008). Best Jimerson, S., Egeland, B., & Teo, A. (1999). A longitudi-
practices in universal screening. In A. Thomas & nal study of achievement trajectories: Factors associ-
J. Grimes (Eds.), Best practices in school psychology ated with change. Journal of Educational Psychology,
V (pp. 721–734). Bethesda, MD: National Association 91, 116–126.
of School Psychologists. Johnson, A. O. (2015). Review of Parenting Stress Index,
Ivanova, M. Y., Achenbach, T. M., Rescorla, L. A., Fourth Edition (PSI-4). Journal of Psychoeducational
Dumenci, L., Almqvist, F., Bilenberg, N., … Verhulst, Assessment, 33, 698–702.
F. C. (2007). The generalizability of the Youth Self- Johnston, C. (1996). Parent characteristics and parent-
Report syndrome structure in 23 societies. Journal of child interactions in families of nonproblem children
Consulting and Clinical Psychology, 75(5), 729–738. and ADHD children with higher and lower levels of
Jackson, M. F., Barth, J. M., Powell, N., & Lochman, J. E. oppositional-defiant behavior. Journal of Abnormal
(2006). Classroom contextual effects of race on chil- Child Psychology, 24(1), 85–104.
dren’s peer cominations. Child Development 7, 7(5), Johnston, C., & Mash, E. J. (2001). Families of children
1325–1337. with attention-deficit/hyperactivity disorder: Review
Jacobs, J. R., Boggs, S. R., Eyberg, S. M., Edwards, D., and recommendations for future research. Clinical
Durning, P., Querido, J. G., … Funderburk, B. W. Child and Family Review, 4, 183–207.
(2000). Psychometric properties and reference point Johnston, C., & Murray, C. (2003). Incremental validity in
data for the Revised Edition of the School Observation the psychological assessment of children and adoles-
Coding System. Behavior Therapy, 31, 695–712. cents. Psychological Assessment, 15, 496–507.
Jamieson, K. H., & Romer, D. (2005). A call to action Joiner, T. E., Schmidt, K. L., & Barnett, J. (1996). Size,
on adolescent mental health. In D. L. Evans, E. B. detail, and line heaviness in children’s drawings as
Foa, R. E. Gur, H. Hendin, C. P. O’Brien, M. E. correlates of emotional distress: (More) negative
P. Seligman, T. Walsh (Eds.) & Annenberg Foundation evidence. Journal of Personality Assessment, 67,
Trust at Sunnylands, Treating and preventing adoles- 127–141.
cent mental health disorders: What we know and what Joint Committee on Testing Practices. (2004). Code of
we don’t know: A research agenda for improving the fair testing practices in education. Washington, DC:
mental health of our youth (p. 617–623). Oxford, UK: American Psychological Association.
Oxford University Press. Jones, D., Dodge, K. A., Foster, E. M., & Nix, R.
Jamison, J., & Freeman, B. (2015). How children (2002). Early identification of children at risk for
understand family conflict and divorce. Minerva costly mental health service use. Prevention Science,
Psichiatrica, 56, 117–133. 3, 247–256.
Jarrett, M. A., Van Meter, A., Youngstrom, E. A., Hilton, Jones, L. M., Mitchell, K. J., & Finkelhor, D. (2012).
D. C., & Ollendick, T. H. (2018). Evidence-based Trends in youth internet victimization: Findings
assessment of ADHD in youth using a receiver operat- from three youth internet safety surveys 2000–2010.
ing characteristic approach. Journal of Clinical Child Journal of Adolescent Health, 50, 179–186.
and Adolescent Psychology, 47, 808–820. Jordan, P., Rescorla, L. A., Althoff, R. R., & Achenbach,
Jellinek, M. S., Murphy, J. M., Robinson, J., Feins, A., T. M. (2016). International comparisons of the Youth
Lamb, S., & Fenton, T. (1988). Pediatric Symptom Self-Report Dysregulation Profile: Latent class
Checklist: Screening school-age children for psy- analyses in 34 societies. Journal of the American
chosocial dysfunction. Journal of Pediatrics, 112, Academy of Child & Adolescent Psychiatry, 55,
201–209. 1046–1053.
Jensen, P. S. (2003). Comorbidity and child psychopathol- Jung, C. J. (1910). The association method. American
ogy: Recommendations for the next decade. Journal Journal of Psychology, 21, 219–235. Reprinted in
of Abnormal Child Psychology, 31, 293–300. L. D. Goodstein & R. I. Lanyon (Eds.), R eadings in
Jensen, P. S., Rubio-Stipec, M., Canino, G., Bird, H. R., personality assessment. New York, NY: Wiley.
Dulcan, M. K., Schwab-Stone, M. E., & Lahey, B. B. Kagan, J., & Snidman, N. (1991). Temperamental factors
(1999). Parent and child contributions to diagnosis of in human development. American Psychologist, 46,
mental disorder: Are both informants always neces- 856–862.
sary? Journal of the American Academy of Child and Kahn, R. E., Frick, P. J., Youngstrom, E., Findling, R. L.,
Adolescent Psychiatry, 3(8), 1569–1579. & Youngstrom, J. K. (2012). The effects of including
Jensen, P. S., Watanabe, H. E., & Richters, J. E. (1999). a callous-unemotional specifier for the diagnosis of
Who’s up first? Testing for order effects in structured conduct disorder. Journal of Child Psychology and
interviews using a counterbalanced experimental Psychiatry, 53, 271–282.
design. Journal of Abnormal Child Psychology, 27, Kamp-Becker, I., Albertowski, K., Becker, J., Ghareman,
439–446. M., Langmann, A., Mingebach, T., … Stroth, S.
Jiang, X. L., & Cillessen, A. H. N. (2005). Stability of (2018). Diagnostic accuracy of the ADOS and
continuous measures of sociometric status: A meta- ASDO-2 in clinical practice. European Child &
analysis. Developmental Review, 25, 1–25. Adolescent Psychiatry, 27(9), 1193–1207.
References 409
Kamphaus, R. W. (2001). Clinical assessment of chil- Kazdin, A. E. (1988). Childhood depression. In E. J. Mash
dren’s intelligence (2nd ed.). Boston, MA: Allyn and & L. G. Terdal (Eds.), Behavioral assessment of child-
Bacon. hood disorders (2nd ed., pp. 157–195). New York,
Kamphaus, R. W. (2012). Screening for behavioral and NY: Guilford Press.
emotional risk: Constructs and practicalities. School Keenan, K. (2000). Emotional dysregulation as a risk fac-
Psychology Forum, 6(4), 89–97. tor for child psychopathology. Clinical Psychology:
Kamphaus, R. W., Dever, B. D., Raines, T. C., DiStefano, Science and Practice, 7, 418–434.
C., & Dowdy, E. (2011, August). Implementation of Keller, H. R. (1986). Behavioral observation approaches
an adolescent behavioral and emotional risk sur- to personality assessment. In H. M. Knoff (Ed.),
veillance protocol in schools. Poster presented at the The assessment of child and adolescent personality
annual convention of the American Psychological (pp. 353–390). New York, NY: Guilford Press.
Association, Washington, DC. Kendall, P. C., & Clarkin, J. F. (1992). Introduction to
Kamphaus, R. W., Morgan, A. W., Cox, M. R., & Powell, special section: Comorbidity and treatment implica-
R. M. (1995). Personality and intelligence in the tions. Journal of Consulting and Clinical Psychology,
psychodiagnostic process. In D. H. Saklofske & 60, 833–834.
M. Zeidner (Eds.), International handbook of person- Kendall, P. C., Puliafico, A. C., Barmish, A. J., Choudhury,
ality and intelligence (pp. 525–544). New York, NY: M. S., Henin, A., & Treadwell, K. S. (2007). Assessing
Plenum Press. anxiety with the Child Behavior Checklist and the
Kamphaus, R. W., Petoskey, M. D., Cody, A. H., Row, Teacher Report Form. Journal of Anxiety Disorders,
E. W., Huberty, C. J., & Reynolds, C. R. (1999). A 21, 1004–1015.
typology of parent-rated child behavior for a national Kendler, K. S., Prescott, C. A., Myers, J., & Neale, M. C.
U.S. sample. Journal of Child Psychology and (2003). The structure of genetic and environmental
Psychiatry and Allied Disciplines, 40, 607–616. risk factors for common psychiatric and substance
Kamphaus, R. W., & Reynolds, C. R. (2007). BASC-2 use disorders in men and women. Archives of General
behavioral and emotional screening system. Psychiatry, 60, 929–937.
Minneapolis, MN: Pearson Assessment. Kerr, D., Lunkenheimer, E., & Olson, S. (2007).
Kaplan, R. M., & Saccuzzo, D. P. (2018). Psychological Assessment of child problem behaviors by multiple
testing: Principles, applications, and issues (9th ed.). informants: A longitudinal study from preschool
Belmont, CA: Wadsworth Cengage Learning. to school entry. Journal of Child Psychology and
Karver, M. S. (2006). Determinants of multiple informant Psychiatry, 48, 967–975.
agreement on child and adolescent behavior. Journal Kerwin, M. E., Kirby, K. C., Speziali, D., Duggan, M.,
of Abnormal Child Psychology, 34, 251–261. Mellitz, C., Versek, B., & McNamara, A. (2015).
Karver, M. S., Handelsman, J. B., Fields, S., & Bickman,L. What can parents do? A review of state laws regarding
(2006). Meta-analysis of therapeutic relationship vari- decision making for adolescent drug abuse and men-
ables in youth and family therapy: The evidence for tal health treatment. Journal of Child & Adolescent
different relationship variables in the child and adoles- Substance Abuse, 24, 166–176.
cent treatment outcome literature. Clinical Psy chol- Kieling, C., Kieling, R., Rohde, L. A., Frick, P. J., Moffitt,
ogy Review, 26, 50–65. T., Nigg, J. T., … Castellanos, F. X. (2010). The age
Kashani, J. H., Orvaschel, H., Burk, J. P., & Reid, J. C. of onset of attention-deficit hyperactivity disorder.
(1985). Informant variance: The issue of parent-child American Journal of Psychiatry, 167, 14–16.
disagreement. Journal of the American Academy of Kilgus, S. P., Bowman, N. A., Christ, T. J., & Taylor,
Child and Adolescent Psychiatry, 24, 437–441. C. N. (2017). Predicting academics via behavior
Kashner, T. M., Rush, A. J., Surís, A., Biggs, M. M., within an elementary sample: An evaluation of the
Gajewski, V. L., Hooker, D. J., … Altshuler, K. Z. social, academic, and emotional behavior risk screener
(2003). Impact of structured clinical interviews on (SAEBRS). Psychology in the Schools, 54, 246–260.
physicians’ practices in community mental health set- Kilgus, S. P., Eklund, K., von der Embse, N. P., Taylor,
tings. Psychiatric Services, 54, 712–718. C. N., & Sims, W. A. (2016). Psychometric defensibility
Kauffman, J. M. (1999). How we prevent the prevention of the Social, Academic, and Emotional Behavior Risk
of emotional and behavioral disorders? Exceptional Screener (SAEBRS) Teacher Rating Scale and multiple
Children, 65, 448–468. gating procedure within elementary and middle school
Kaufman, J., Birmaher, B., Axelson, D., Perepletchikova, samples. Journal of School Psychology, 58, 21–39.
F., Brent, D., & Ryan, N. (2016). Schedule for affec- Kilgus, S. P., Riley-Tillman, T. C., Chafouleas, S. M.,
tive and disorders and schizophrenia for school aged Christ, T. J., & Welsh, M. E. (2014). Direct ehavior
children (6–18 years): Kiddie-SADS-lifetime ver- rating as a school-based behavior universal screener:
sion (K-SADS-PL DSM 5). Pittsburgh, PA: Western Replication across sites. Journal of School Psychology,
Psychiatric Institute and Clinic. 52, 63–82.
Kazdin, A. E. (1981). Behavioral observation. In Kilgus, S. P., & von der Embse, N. P. (2015). Social,
M. Hersen & A. S. Bellack (Eds.), Behavioral assess- academic, and emotional behavior risk screener
ment: A practical handbook (pp. 59–100). New York, (SAEBRS). Minneapolis, MN: Theodore J. Christ &
NY: Pergamon Press. Colleagues.
410 References
Kilgus, S. P., von der Embse, N. P., Allen, A. N., Taylor, lescent psychopathy (pp. 79–109). New York, NY:
C. N., & Eklund, K. (2018). Examining SAEBRS Guilford Press.
technical adequacy and the moderating influence of Kovacs, M. (1992). The Children’s Depression Inventory
criterion type on cut score performance. Remedial and manual. New York, NY: Multi-Health Systems.
Special Education, 39, 377–388. Kovacs, M. (2011). Children’s depression inventory (2nd
Kimonis, E. R., Fanti, K. A., Frick, P. J., Moffitt, T. E., ed.). North Tonawanda, NY: Multi-Health Systems.
Essau, C., Bijjtebier, P., & Marsee, M. A. (2015). Kowalski, R. M., & Limber, S. P. (2013). Psychological,
Using self-reported callous-unemotional traits to physical, and academic correlates of cyberbullying
cross-nationally assess the DSM-5 “With Limited and traditional bullying. Journal of Adolescent Health,
Prosocial Emotions” specifier. Journal of Child 53, S13–S20.
Psychology and Psychiatry, 56, 1249–1261. Kroon, N., Goudena, P. P., & Rispens, J. (1998). Thematic
Kimonis, E. R., Fanti, K. A., & Singh, J. P. (2014). apperception tests for child and adolescent assess-
Establishing cut-off scores for the parent-reported ment: A practitioner’s consumer guide. Journal of
inventory of callous-unemotional traits. Archives of Psychoeducational Assessment, 16, 99–117.
Forensic Psychology, 1(1), 27–48. Kugler, B. B., Bloom, M., Kaercher, L. B., Nagy, S.,
Kimonis, E. R., Frick, P. J., & McMahon, R. J. (2014). Truax, T. V., Kugler, K. M., … Storch, E. A. (2013).
Conduct and oppositional defiant disorders. In E. J. Predictors of differential responding on a sentence
Mash & R. A. Barkley (Eds.), Child psychopathology completion task in traumatized children. Journal of
(3rd ed., pp. 145–179). New York, NY: Guilford. Child and Family Studies, 22, 244–252.
Kimonis, E. R., Frick, P. J., Skeem, J., Marsee, M. A., Kuijpers, R. C., Kleinjan, M., Engels, R. C., Stone, L. L.,
Cruise, K., Munoz, L. C., … Morris, A. S. (2008). & Otten, R. (2015). Child self-report to identify inter-
Assessing callous-unemotional traits in adolescent nalizing and externalizing problems and the influ-
offenders: Validation of the Inventory of Callous- ence of maternal mental health. Journal of Child and
Unemotional Traits. International Journal of Law and Family Studies, 24, 1605–1614.
Psychiatry, 31, 241–251. Kujala, S., Walsh, T., Nurkka, P., & Crisan, M. (2014).
Klein, D. N., Dougherty, L. R., & Olino, T. M. (2005). Sentence completion for understanding users and eval-
Toward guidelines for evidence-based assessment uating user experience. Interacting with Computers,
of depression in children and adolescents. Journal 26, 238–255.
of Clinical Child and Adolescent Psychology, 34, Lachar, D. (1990). Objective assessment of child and
412–432. adolescent personality: The Personality Inventory
Kleinmuntz, B. (1967). Personality measurement: An for Children. In C. R. Reynolds & R. W. Kamphaus
introductio n. Homewood, IL: Dorsey Press. (Eds.), Handbook of psychological and educa tional
Kline, R. B. (1995). New objective rating scales for child assessment of children: Personality, behavior, and
assessment, II. Self-report scales for children and ado- context (pp. 298–323). New York, NY: Guilford Press.
lescents: Self-report of Personality of the Behavior Lachar, D., & Gruber, C. (2001). Personality Inventory
Assessment System for Children, the Youth Self-Report, for Children, second edition (PIC-2): Standard form
and the Personality Inventory for Youth. Journal of and behavioral summary manual. Los Angeles, CA:
Psychoeducational Assessment, 13, 169–193. Western Psychological Services.
Knoff, H. M. (1983). Justifying projective/personal- Lachar, D., Wingenfeld, S. A., Kline, R. B., & Gruber,
ity assessment in school psychology: A response to C. P. (2000). Student behavior survey. Los Angeles,
Batsch and Peterson. School Psychology Review, 12, CA: Western Psychological Services.
446–451. LaGreca, A. M., & Lopez, N. (1998). Social anxiety
Konold, T. R., Walthall, J. C., & Pianta, R. C. (2004). The among adolescents: Linkages with peer relation-
behavior of child ratings: Measurement structure of ships and friendships. Journal of Abnormal Child
the child behavior checklist across time, informants, Psychology, 26, 83–94.
and child gender. Behavioral Disorders, 29, 372–383. Lahey, B. B., Applegate, B., McBurnett, K., Biederman,
Koot, H. M., & Verhulst, F. C. (1992). Prediction of chil- J., Greenhill, L., Hynd, G. W., et al. (1994). DSM-IV
dren’s referral to mental health and special education field trials for attention-deficit hyperactivity disorder
services from earlier adjustment. Journal of Child in children and adolescents. American Journal of
Psychology and Psychiatry, 33, 717–729. Psychiatry, 151, 1673–1685.
Koppitz, E. M. (1968). Psychological evaluation of chil- Lahey, B. B., Applegate, B., Waldman, I. D., Loft, J. D.,
dren’s human figure drawings. New York, NY: Grune Hankin, B. L., & Rick, J. (2004). The structure of child
at Stratton. and adolescent psychopathology: Generating new
Koppitz, E. M. (1983). Projective drawings with chil- hypotheses. Journal of Abnormal Psychology, 113,
dren, and adolescents. S chool Psychology Review, 12, 358–385.
421–427. Lahey, B. B., Carlson, C. L., & Frick, P. J. (1997).
Kotler, J. S., & McMahon, R. J. (2010). Assessment of Attention-deficit disorders: A review of research rel-
child and adolescent psychopathy. In D. R. Lynam evant to diagnostic classification. In T. A. Widiger,
& R. T. Salekin (Eds.), Handbook of child and ado- A. J. Frances, H. A. Pincus, R. Ross, M. B. First,
References 411
& W. Davis (Eds.), DSM-IV sourcebook (Vol. 3, Last, C. G., Hersen, M., Kazdin, A. E., Francis, G., &
pp. 163–188). Washington, DC: American Psychiatric Grubb, H. J. (1987). Psychiatric illness in the mothers
Association. of anxious children. American Journal of Psychiatry,
Lahey, B. B., Goodman, S. H., Waldman, I. D., Bird, 144, 1580–1583.
H., Canino, G., Jenson, P., … Applegate, B. (1999). Le Couteur, A., Rutter, M., Lord, C., Rios, P., Robertson, S.,
Relation of age of onset to the type of severity of Holdgrafer, M., & McLennan, J. (1989). Autism diag-
child and adolescent conduct problems. Journal of nostic interview: A standardized investigator-based
Abnormal Child Psychology, 27, 247–260. instrument. Journal of Autism and Developmental
Lahey, B. B., Krueger, R. F., Rathouz, P. J., Waldman, Disorders, 19, 363–387.
I. D., & Zald, D. H. (2017). A hierarchical causal LeBlanc, L. A., Sautter, R. A., & Dore, D. J. (2006). Peer
taxonomy of psychopathology across the life span. relationship problems. In M. Hersen (Ed.), Clinician’s
Psychological Bulletin, 143, 142–186. handbook of child behavioral assessment; clinician’s
Lahey, B. B., & Loeber, R. (1994). Framework for a handbook of child behavioral assessment (pp. 377–
developmental model of oppositional defiant disorder 399). San Diego, CA: Elsevier Academic Press.
and conduct disorder. In D. K. Routh (Ed.), Disruptive Lee, S. S., Humphreys, K. L., Flory, K., Liu, R., & Glass,
behavior disorders in childhood (pp. 139–180). K. (2011). Prospective association of childhood
New York: Plenum. attention-deficit/hyperactivity disorder (ADHD) and
Lahey, B. B., Loeber, R., Stouthamer-Loeber, M., Christ, substance use and abuse/dependence: A meta-analytic
M. A. G., Green, S., Russo, M. F., … Dulcan, M. review. Clinical Psychology Review, 31, 328–341.
(1990). Comparison of DSM-III and DSM-III-R diag- Leeuwen, K. V., Meerschaert, T., Bosmans, G., De Medts,
noses for prepubertal children: Changes in prevalence L., & Braet, C. (2006). The strengths and difficulties
and validity. Journal of the American Academy of questionnaire in a community sample of young chil-
Child & Adolescent Psychiatry, 29(4), 620–626. dren in Flanders. European Journal of Psychological
Laird, R. D., & De Los Reyes, A. (2013). Testing infor- Assessment, 22, 189–197.
mant discrepancies as predictors of early adolescent Leffler, J. M., Riebel, J., & Hughes, H. M. (2014). A
psychopathology: Why difference scores cannot review of child and adolescent diagnostic interviews
tell you what you may want to know and how poly- for clinical practitioners. Assessment, 22, 690–703.
nomial regression may. Journal of Abnormal Child Lefkowitz, M. M., & Tesiny, E. P. (1980). Assessment
Psychology, 41, 1–14. of childhood depression. Journal of Consulting and
Lambert, M., Whipple, J., Hawkins, E., Vermeersch, D., Clinical Psychology, 48, 43–50.
Nielsen, S., & Smart, D. (2003). Is it time for clini- Leiner, M. A., Balcazar, H., Straus, D. C., Shirsat, P.,
cians to routinely track patient outcome? A meta-anal- & Handal, G. (2007). Screening Mexicans for psy-
ysis. Clinical Psychology: Science and Practice, 10, chosocial behavioral p roblems during pediatric
288–301. consultation. Revista de Investigación Clínica, 59,
Lamers, F., van Oppen, P., Comijs, H. C., Smit, J. H., 116–123.
Spinhoven, P., van Balkom, A. J. L. M., … Penninx, Lemery, K., Essex, M., & Smider, N. (2002). Revealing
B. W. J. H. (2011). Comorbidity patterns of anxiety the relation between temperament and behavior prob-
and depressive disorders in a large cohort study: The lem symptoms by eliminating measurement con-
Netherlands study of depression and anxiety (NESDA). founding: Expert ratings and factor analyses. Child
The Journal of Clinical Psychiatry, 72, 341–348. Development, 73, 867–882.
Lane, K. L., Carter, E. W., Pierson, M. R., & Glaeser, Lett, N. J., & Kamphaus, R. W. (1997). Differential valid-
B. C. (2006). Academic, social, and behavioral charac- ity of the BASC student observation system and the
teristics of high school students with emotional distur- BASC teacher rating scales. Canadian Journal of
bances or learning disabilities. Journal of Emotional School Psychology, 13, 1–14.
and Behavioral Disorders, 14, 108–117. Levitt, J. M., Saka, N., Romanelli, L. H., & Hoagwood, K.
Lane, K. L., Kalberg, J. R., Parks, R. J., & Carter, E. W. (2007). Early identification of mental health problems
(2008). Student Risk Screening Scale: Initial evidence in schools: The status of instrumentation. Journal of
for score reliability and validity at the high school School Psychology, 45, 163–191.
level. Journal of Emotional and Behavioral Disorders, Lichtenberger, E. O., Mather, N., Kaufman, N. L., &
16, 178–190. Kaufman, A. S. (2004). Essentials of assessment
Lane, K. L., Little, M. A., Casey, A. M., Lambert, W., report writing. Hoboken, NJ: John Wiley & Sons.
Wehby, J., & Weisenbach, J. L. (2009). A comparison Lilienfeld, S. O. (2003). Comorbidity between and within
of systematic screening tools for emotional and behav- childhood externalizing and internalizing disorders:
ior disorders. Journal of Emotional and Behavioral Reflections and directions. Journal of Abnormal Child
Disorders, 17, 93–105. Psychology, 31, 285–291.
Lane, K. L., Parks, R. J., Kalberg, J. R., & Carter, E. W. Lilienfeld, S. O., Wood, J. M., & Garb, H. N. (2000).
(2007). Systematic screening at the middle school The scientific status of projective techniques.
level: Score reliability and validity of the Student Risk Psychological Science in the Public Interest, 1, 27–66.
Screening Scale. Journal of Emotional and Behavioral Lloyd, J. W., Kauffman, J. M., Landrum, T. J., & Roe,
Disorders, 15, 209–222. D. L. (1991). Why do teachers refer pupils for spe-
412 References
cial education? An analysis of referral records. view schedule for DSM-IV: Child and parent ver-
Exceptionality, 2, 115–126. sion. Journal of the American Academy of Child &
Lochman, J. E., & The Conduct Problems Prevention Adolescent Psychiatry, 46, 731–736.
Research Group. (1995). Screening of child behav- Lynskey, M. T., & Fergusson, D. M. (1995). Childhood
ior problems for prevention programs at school entry. conduct problems, attention deficit behaviors, and
Journal of Consulting and Clinical Psychology, 63, adolescent alcohol, tobacco, and illicit drug use. J
549–559. ournal of Abnormal Child Psychology, 23, 281–302.
Locke, H. J., & Wallace, K. M. (1959). Short mari- Maccoby, E. E., & Martin, A. (1983). Socialization in
tal- adjustment and prediction tests: Their reliabil- the context of the family: Parent-child interaction. In
ity and validity. Marriage and Family Living, 21, P. H. Mussen & E. M. Hetherington (Eds.), Handbook
251–255. of child psychology: IV; Socialization, personality,
Loeber, R., Green, S. M., & Lahey, B. B. (1990). Mental and social development (pp. 1–101). New York, NY:
health professionals’ perception of the utility of chil- Wiley.
dren, mothers, and teachers as informants on child- Madsen, C. J. (1990, September). Attention deficit dis-
hood psychopathology. Journal of Clinical Child orders and application of the EFL4 and Section
Psychology, 19, 136–143. 504. Paper presented to the Seventh Annual Pacific
Loeber, R., Green, S. M., Lahey, B. B., & Stouthamer- Northwest Institute on Special Education and Law.
Loeber, M. (1991). Differences and similarities Maguin, E., Nochajski, T. H., DeWit, D. J., & Safyer, A.
between children, mothers, and teachers as informants (2016). Examining the validity of the adapted Alabama
on disruptive child behavior. Journal of Abnormal Parenting Questionnaire-Parent Global Report
Child Psychology, 19, 75–95. Version. Psychological Assessment, 28, 613–625.
Lorber, M. F., & Slep, A. M. S. (2015). Are persistent Mann, B. J., & MacKenzie, E. P. (1996). Pathways among
early onset conduct problems predicted by the tra- marital functioning, parental behaviors, and child
jectories and initial levels of discipline practices. behavior problems in school-age boys. Journal of
Developmental Psychology, 51, 1048–1061. Clinical Child Psychology, 25, 183–191.
Lord, C., Risi, S., DiLavore, P. S., Shulman, C., Thurm, Mannuzza, S., Castellanos, F. X., Roizen, E. R., Huchison,
A., & Pickles, A. (2006). Autism from 2 to 9 years J. A., Lashua, E. C., & Klein, R. G. (2011). Impact
of age. Archives of General Psychiatry, 63, 694–701. of the impairment criterion in the diagnosis of adult
Lord, C., Rutter, M., DiLavore, P. C., & Risi, S. (1999). ADHD: 33-year follow-up study of boys with ADHD.
Autism diagnostic observation schedule-WPS (ADOS- Journal of Attention Disorders, 15, 122–129.
WPS). Los Angeles, CA: Western Psychological Mannuzza, S., Gittelman-Klein, R., Konig, P. H., &
Services. Giampino, T. L. (1989). Hyperactive boys almost
Lord, C., Rutter, M., DiLavore, P. C., Risi, S., Gotham, grown up: IV. Criminality and its relationship to psy-
K., & Bishop, S. (2012). Autism diagnostic obser- chiatric status. Archives of General Psychiatry, 46,
vation schedule (2nd ed.). Torrance, CA: Western 1073–1079.
Psychological Services. Martin, R. P. (1983). The ethical issues in the use and inter-
Lord, C., Rutter, M., & Le Couteur, A. (1994). Autism pretation of the Draw-a-Person Test and other similar
diagnostic interview-revised: A revised version of projective procedures. The School Psychologist, 3
a diagnostic inter-view for caregivers of individu- 8(6), 8.
als with possible pervasive developmental disorders. Martin, R. P. (1988). Assessment of personality and
Journal of Autism and Developmental Disorders, 24, behavior problems: Infancy through adolescence.
659–685. New York, NY: Guilford Press.
Lorr, M. (1965). Review of the Wittenborn Psychiatric Mash, E., & Hunsley, J. (2005). Evidence-based assess-
Rating Scales. In O. K. Buros (Ed.), Mental measure- ment of child and adolescent disorders: Issues and
ments yearbook (5th ed.). Highland Park, NJ: Gryphon challenges. Journal of Clinical Child and Adolescent
Press. Psychology, 34, 362–379.
Lowe, P. A. (2014). A closer look at the psychometric Mash, E. J., & Barkley, R. (2014). Child psychopathology.
properties of the Revised Children’s Manifest Anxiety New York, NY: The Guilford Press.
Scale-Second Edition among U.S. elementary and sec- Mash, E. J., & Terdal, L. G. (1988). Behavioral assess-
ondary school students. Journal of Psychoeducational ment of child and family disturbance. In E. J. Mash &
Assessment, 32, 495–508. L. G. Terdal (Eds.), Behavioral assessment of child-
Lubman, D. I., Hides, L., & Elkins, K. (2008). Developing hood disorders (2nd ed., pp. 3–68). New York, NY:
integrated models of care within the youth alcohol Guilford Press.
and other drug sector. Australasian Psychiatry, 16, Masi, A., DeMayo, M. M., Glozier, N., & Guastella, A. J.
363–366. (2017). An overview of autism spectrum disorder,
Lucia, V. C., & Breslau, N. (2006). Family cohesion and heterogeneity and treatment options. Neuroscience
children’s behavior problems: A longitudinal investi- Bulletin, 33, 183–193.
gation. Psychiatry Research, 141, 141–149. Masia-Warner, C., Storch, E., Pincus, D., Klein, R.,
Lyneham, H. J., Abbott, M. J., & Rapee, R. M. (2007). Heinberg, R., & Leibowitz, M. R. (2003). The
Interrater reliability of the anxiety disorders inter- Leibowitz social anxiety scale for children and adoles-
References 413
cents: An initial psychometric investigation. Journal social competence. In P. S. Strain, M. J. Guralnick,
of the American Academy of Child and Adolescent & H. M. Walker (Eds.), Children’s social behavior:
Psychiatry, 42, 1076–1084. Development, assessment, and modification (pp. 215–
Massetti, G. M., Lahey, B. B., Pelham, W. E., Loney, 286). New York, NY: Academic Press.
J., Ehrhardt, A., Lee, S. S., et al. (2008). Academic McCrae, R. R., Stone, S. V., Fagan, P. J., & Costa, P. T.
achievement over 8years among children who met (2008). Identifying causes of disagreement between
modified criteria for attention-deficit/hyperactivity self-reports and spouse ratings of personality. Journal
disorder at 4–6 years of age. Journal of Abnormal of Personality, 66(3), 285–313.
Child Psychology, 36, 399–410. McDonald, K. L., Putallaz, M., Grimes, C. L.,
Maughan, B., Collishaw, S., & Stringaris, A. (2013). Kupersmidt, J. B., & Coie, J. D. (2007). Girl talk:
Depression in childhood and adolescence. Journal Gossip, friendship, and sociometric status. Merrill-
of the Canadian Academy of Child and Adolescent Palmer Quarterly, 53, 381–411.
Psychiatry, 22, 35–40. McEvoy, A., & Welker, R. (2000). Antisocial behavior,
Mayeux, L., Bellmore, A. D., & Cillessen, A. H. N. academic failure, and school climate: A critical review.
(2007a). Predicting changes in adjustment using Journal of Emotional and Behavioral Disorders, 8,
repeated measures of sociometric status. The Journal 130–140.
of Genetic Psychology, 168, 401–424. McFall, R. M. (2005). Theory and utility-Key themes in
Mayeux, L., Underwood, M. K., & Risser, S. D. (2007b). evidence-based assessment: Comment on the special
Perspectives on the ethics of sociometric research with section. Psychological Assessment, 17, 312–323.
children. Merrill-Palmer Quarterly, 53, 53–78. McGoldrick, M. (2016). The genogram casebook: A clini-
McArthur, D. S., & Roberts, G. E. (1982). Roberts cal companion to genograms: Assessment and inter-
Apperception Test for children. Los Angeles, CA: vention. New York, NY: W. W. Norton & Company.
Western Psychological Services. McMahon, R. J., & Frick, P. J. (2005). Evidence-based
McCarney, S. B. (2004). The attention deficit disorders assessment of conduct problems in children and ado-
evaluation scale, home and school versions: Technical lescents. Journal of Clinical Child and Adolescent
manual. Columbia, MO: Hawthorne. Psychology, 34, 477–505.
McClendon, D. T., Warren, J. S., Green, K. M., McMahon, R. J., & Frick, P. J. (2007). Conduct and
Burlingame, G. M., Eggett, D. L., & McClendon, R. J. oppositional disorders. In E. J. Mash & R. A. Barkley
(2011). Sensitivity to change of youth treatment out- (Eds.), Assessment of childhood disorders (4th ed.,
come measures: A comparison of the CBCL, BASC- pp. 132–183). New York, NY: Guilford.
2, and Y-OQ. Journal of Clinical Psychology, 67, McMahon, R. J., & Peters, R. D. V. (2002). The effects
111–125. of parental dysfunction on children. New York, NY:
McConaughy, S. H. (2005). Direct observational assess- Plenum Press.
ment during test sessions. School Psychology Review, McReynolds, P., Ballachey, E. L., & Ferguson, J. T.
34, 490–506. (1952). Development and evaluation of a behavioral
McConaughy, S. H. (2013). Clinical interviews for chil- scale for appraising the adjustment of hospitalized
dren and adolescents, second edition: Assessment to patients. American Psychologist, 7, 340.
intervention. New York, NY: Guilford Press. Meehl, P. E., & Rosen, A. (1955). Antecedent probability
McConaughy, S. H., & Achenbach, T. M. (1989). and the efficiency of psychometric signs, patterns, or
Empirically based assessment of serious emotional cutting scores. Psychological Bulletin, 52, 194–216.
disturbance. Journal of School Psychology, 27, Mellor, D., & Stokes, M. (2007). The factor structure of
91–117. the Strengths and Difficulties Questionnaire. European
McConaughy, S. H., & Achenbach, T. M. (2004). Journal of Psychological Assessment, 23, 105–112.
Manual for the test observation form for ages 2–18. Melton, T. H., Croarkin, P. E., Strawn, J. R., & Mcclintock,
Burlington, VT: Research Center for Children, Youth, S. M. (2016). Comorbid anxiety and depressive symp-
and Families. toms in children and adolescents: A systematic review
McConaughy, S. H., Achenbach, T. M., & Gent, C. L. and analysis. Journal of Psychiatric Practice, 22,
(1988). Multiaxial empirically based assessment: 84–98.
Parent, teacher, observational, cognitive, and per- Mendenhall, T. J., & Trudeau-Hern, S. (2014). Using
sonality correlates of child behavior profile types for medical genograms in clinical supervision. In R. A.
6- to 11-year-old boys. Journal of Abnormal Child Bean, S. D. Davis, & M. P. Davey (Eds.), Clinical
Psychology, 16, 485–509. supervision activities for increasing competence and
McConaughy, S. H., Stanger, C., & Achenbach, T. M. self-awareness (pp. 141–147). Hoboken, NJ: John
(1992). Three year course of behavioral emotional Wiley & Sons.
problems in a national sample of 4- to 16-year-olds: I. Meyer, G. J., Erdberg, P., & Shaffer, T. W. (2007).
Agreement among informants. Journal of the American Toward international normative reference data for
Academy of Child and Adolescent Psychiatry, 31, the Comprehensive System. Journal of Personality
932–940. Assessment, 89(Suppl. 1), S201–S216.
McConnell, S. R., & Odom, S. L. (1986). Sociometrics: Meyer, J. K., Hong, S., & Morey, L. C. (2015). Evaluating
Peer-referenced measures and the assessment of the validity indices of the Personality Assessment
414 References
Inventory-Adolescent Version. Assessment, 22, Milner, J. S., & Crouch, J. L. (2012). Psychometric
490–496. characteristics of translated versions of the Child
Midgley, N., Parkinson, S., Holmes, J., Stapley, E., Abuse Potential Inventory. Psychology of Violence, 2,
Eatough, V., & Target, M. (2017). “Did I bring it on 239–259.
myself?” an exploratory study of the beliefs that ado- Mischel, W. (1968). Personality and assessment.
lescents referred to mental health services have about New York, NY: Wiley.
the causes of their depression. European Child & Moens, M. A., Weeland, J., van der Glessen, D.,
Adolescent Psychiatry, 26, 25–34. Chhangur, R. R., & Overbeek, G. (2018). In the eye of
Mihura, J. L., Roy, M., & Graceffo, R. A. (2017). the beholder? Parent-observer discrepancies in parent-
Psychological assessment training in clinical psy- ing and child disruptive behavior assessments. Journal
chology doctoral programs. Journal of Personality of Abnormal Child Psychology, 46, 1147–1159.
Assessment, 99, 153–164. Moffett, T. E. (2018). Male antisocial behavior in ado-
Mikami, A. Y., & Hinshaw, S. P. (2003). Buffers of peer lescence and beyond. Nature Human Behavior, 2(3),
rejection among girls with and without ADHD: The 177–186.
role of popularity with adults and goal-directed soli- Mohr, C., & Schneider, S. (2013). Anxiety disorders.
tary play. Journal of Abnormal Child Psychology, 31, European Child & Adolescent Psychiatry, 22, 17–22.
381–397. Mondimore, F. M., Zandi, P. P., MacKinnon, D. F.,
Mildenberger, K., Sitter, S., Noterdaeme, M., & Amorosa, McInnis, M. G., Miller, E. B., Crowe, R. P., …
H. (2001). The use of the ADI–R as a diagnostic tool Potash, J. B. (2006). Familial aggregation of illness
in the differential diagnosis of children with infantile chronicity in recurrent, early-onset major depression
autism and children with receptive language disor- pedigrees. The American Journal of Psychiatry, 163,
der. European Child and Adolescent Psychiatry, 10, 1554–1560.
248–255. Monuteaux, M. C., Faraone, S. V., Gross, L. M., &
Milette-Winfree, M., & Mueller, C. W. (2018). Treatment- Biederman, J. (2007). Predictors, clinical character-
as-usual therapy targets for comorbid youth dis- istics, and outcome of conduct disorder in girls with
proportionately focus on externalizing problems. attention deficit hyperactivity disorder: A longitudinal
Psychological Services, 15, 65–77. study. Psychological Medicine, 37, 1731–1741.
Milich, R. (1984). Cross-sectional and longitudinal Moos, R. H., & Moos, B. S. (1986). Family environment
observations of activity level and sustained attention scale manual (2nd ed.). Palo Alto, CA: Consulting
in a normative sample. Journal of Abnormal Child Psychologists Press.
Psychology, 12(2), 261–275. Morales-Hidalgo, P., Hernández-Martínez, C., Vera, M.,
Milich, R., Balentine, A. C., & Lynam, D. R. (2001). Voltas, N., & Canals, J. (2017). Psychometric prop-
ADHD combined type and ADHD predominantly erties of the Conners-3 and Conners early childhood
inattentive type are distinct and unrelated disor- indexes in a Spanish school population. International
ders. Clinical Psychology: Science and Practice, 8, Journal of Clinical and Health Psychology, 17, 85–96.
463–488. Moretz, M. W., & McKay, D. (2011). Dimensional diag-
Milich, R., Loney, J., & Landau, S. (1982). The indepen- nosis of anxiety in youth. In M. Crozier, S. Gillihan,
dent dimensions of hyperactivity and aggression: A & M. Powers (Eds.), Handbook of child and adoles-
validation with play-room observations. Journal of cent anxiety disorders (pp. 105–117). New York, NY:
Abnormal Psychology, 91, 183–198. Springer.
Miller, D. N., & Nickerson, A. B. (2007). Projective tech- Morey, L. C. (2007). Personality assessment inven-
niques and the school-based assessment childhood tory—Adolescent professional manual. Lutz, FL:
internalizing disorders: A critical analysis. Journal of Psychological Assessment Resources.
Projective Psychology and Mental Health, 14, 48–58. Morgenthaler, W. (1954). Hermann Rorschach, deceased,
Miller, L. C. (1967). Louisville Behavior Checklist for at his 70th birthday. Schweizerische Zeitschrift fuer
males 6–12 years of age. Psychological Reports, 21, Psychologie and ihre Anwendungen, 1 3, 315
885–896. Morris, R. J., & Kratochwill, T. R. (1983). Treating chil-
Milne, B. J., Caspi, A., Crump, R., Poulton, R., Rutter, dren’s fears and phobias: A behavioral approach.
M., Sears, M. R., & Moffitt, T. E. (2009). The validity New York, NY: Pergamon Press.
of the family history screen for assessing family his- Morris, T. L., & Ale, C. M. (2011). Social anxiety.
tory of mental disorders. American Journal of Medical In M. Crozier, S. Gillihan, & M. Powers (Eds.),
Genetics Part B: Neuropsychiatric Genetics, 180, Handbook of child and adolescent anxiety disorders
41–49. (pp. 289–301). New York, NY: Springer New York.
Milner, J. S. (1986). The Child Abuse Potential inventory: Morrison, J., & Flegel, K. (2016). Interviewing child and
Manual (2nd ed.). De Kalb, IL: Psytec. adolescents, second edition: Skills and strategies for
Milner, J. S., & Crouch, J. L. (1997). Impact and detec- effective DSM-5 diagnoses. New York, NY: Guilford
tion of response distortions on parenting measures Press.
used to assess risk for child physical abuse. Journal of Morsbach, S. K., & Prinz, R. J. (2006). Understanding
Personality Assessment, 69, 633–650. and improving the validity of self-report of parenting.
Clinical Child and Family Review, 9, 1–21.
References 415
Moss, J., Magiati, I., Charman, T., & Howlin, P. (2008). genetic studies: Rationale, unique features, and train-
Stability of the autism diagnostic interview-revised ing. Archives of General Psychiatry, 51, 849–859.
from pre-school to elementary school age in children O’Connell, M. E., Boat, T., & Warner, K. E. (Eds.).
with autism spectrum disorders. Journal of Autism and (2009). Preventing mental, emotional, and behavioral
Developmental Disorders, 38, 1081–1091. disorders among young people: Progress and possi-
Mucka, L. E., Hinrichs, J., Upton, F., Hetterscheidt, L., bilities. Washington, DC: National Academy Press.
Kuentzel, J., Bartoi, M., & Barnett, D. (2017). Barriers Odgers, D. L., Caspi, A., Broadbent, J. M., Dickson, N.,
to adherence to child assessment recommendations. Hancox, R. J., Harrington, H., … Moffitt, T. E. (2007).
Journal of Child and Family Studies, 26, 1029–1039. Prediction of differential adult health burden by con-
Muñoz, L. C., & Frick, P. J. (2007). The reliability, sta- duct problem subtypes in males. Archives of General
bility, and predictive utility of the self-report ver- Psychiatry, 64, 476–484.
sion of the Antisocial Process Screening Device. Odgers, D. L., Moffitt, T. E., Broadbent, J. M., Dickson,
Scandinavian Journal of Psychology, 48, 299–312. N., Hancox, R. J., Harrington, H., … Caspi, A. (2008).
Muratori, P., Milone, A., Nocentini, A., Manfredi, A., Female and male antisocial trajectories: From child-
Polidori, L., Ruglioni, L., … Lochman, J. E. (2015). hood origins to adult outcomes. Development and
Maternal depression and parenting practices predict Psychopathology, 20, 673–716.
treatment outcome in Italian children with disrup- Office of Strategic Services Assessment Staff. (1948).
tive behavior disorder. Journal of Child and Family Assessment of men: Selection of personnel for
Studies, 24, 2805–2816. the Office of Strategic Services. New York, NY:
Murray, H. A. (1943). Thematic apperception test. Rinehart.
Cambridge, MA: Harvard University Press. Okasha, A. (2009). Would the use of dimensions instead
Musser, E. D., Hawkey, E., Kachan-Liu, S. S., Lees, P., of categories remove problems related to subthresh-
Roullet, J. B., Goddard, K., … Nigg, J. T. (2014). old disorders? European Archives of Psychiatry and
Shared familial transmission of autism spectrum and Clinical Neuroscience, 259(2), S129–S133.
attention-deficit/hyperactivity disorders. Journal of Okiishi, J., Lambert, M., Eggett, D., Nielsen, L., Dayton,
Child Psychology and Psychiatry, 55, 819–827. D., & Vermeersch, D. (2006). An analysis of thera-
Mychailyszyn, M. P., & Whitehead, M. R. (2016). pist treatment effects: Toward providing feedback
Cognitive strategies. In D. W. Nangle, D. J. Hansen, to individual therapists on their clients’ psychother-
R. L. Grover, J. N. Kingery, & C. Suveg (Eds.), apy outcome. Journal of Clinical Psychology, 62,
Treating internalizing disorders in children and ado- 1157–1172.
lescents: Core techniques and strategies (pp. 39–67). Ollendick, T. H., & King, N. J. (1994). Diagnosis, assess-
New York, NY: Guilford Press. ment, and treatment of internalizing problems in
Nadeau, K. G. (1995). Attention-deficit disorder in adults: children: The role of longitudinal data. Journal of
Research, diagnosis, and treatment. New York: Consulting and Clinical Psychology, 62, 918–927.
Brunner & Mazel. Ondersma, S. J. (2016). Introduction to the second special
National Center for Education Statistics (NCES), Institute section on substance abuse and child maltreatment.
of Education Sciences, U.S. Department of Education. Child Maltreatment, 12(2), 111–113.
(2019). The condition of education 2019. Washington, Ondersma, S. J., Chaffin, M. J., Mullins, S. M., &
DC: Authors. LeBreton, J. M. (2005). A brief form of the Child
Nezu, A. M., & Nezu, C. M. (1993). Identifying and Abuse Potential Inventory: Development and vali-
selecting target problems for clinical interventions: A dation. Journal of Clinical Child and Adolescent
problem-solving model. Psychological Assessment, 5, Psychology, 34, 301–311.
254–263. Orchard, F., Cooper, P. J., & Creswell, C. (2015).
Nigg, J. T. (2006). What causes ADHD? Understanding Interpretation and expectations among mothers of
what goes wrong and why. New York: Guilford. children with anxiety disorders: Associations with
Nixon, R. D. V., Sweeney, L., Erickson, D. B., & Touyz, maternal anxiety disorder. Depression and Anxiety,
S. W. (2003). Parent-child interaction therapy: A 32, 99–107.
comparison of standard and abbreviated treatments Orvaschel, H., Ambrosini, P., & Rabinovich, H. (1993).
for oppositional defiant preschoolers. Journal of Diagnostic issues in child assessment. In T. H.
Consulting and Clinical Psychology, 71, 251–260. Ollendick & M. Hersen (Eds.), Handbook of child
Nock, M. K., Kazdin, A. E., Hiripi, A. E., & Kessler, and adolescent assessment (pp. 26–40). Boston, MA:
R. C. (2006). Prevalence, subtypes, and correlates of Allyn and Bacon.
DSM-IV conduct disorder in the National Comorbidity Osa, N., Granero, R., Penelo, E., Domenech, J. M., &
Survey Replication. Psychological Medicine, 36, Ezpeleta, L. (2014). Psychometric properties of the
699–710. Alabama Parenting Questionnaire-Preschool Revision
Nurnberger, J. I., Jr., Blehar, A. S. C., Kaufmann, C. A., (APQ-PR) in 3 year old Spanish preschoolers. Journal
York-Cooler, C., Simpson, S. G., Harkavy-Friedman, of Child and Family Studies, 23, 776–784.
J., … Reich, T. (1994). Diagnostic interview for Osman, A., Gutierrez, P. M., Bagge, C. L., Fang, Q., &
Emmerich, A. (2010). Reynolds adolescent depression
416 References
scale-second edition: A reliable and useful instrument. Payne, A. F. (1928). Sentence completions. New York,
Journal of Clinical Psychology, 66, 1324–1345. NY: New York Guidance Clinic.
Ostrov, J. M., & Keating, C. F. (2004). Gender differ- Pearson, K. (1901). On lines and planes of closest fit to
ences in preschool aggression during free play and systems of points in space. Philosophical Magazine
structured interactions: An observational study. Social (Series 6), 2, 559–572.
Development, 13, 255–277. Pedersen, A. L., Pettygrove, S., Lu, Z., Andrews, J.,
Owens, M., Stevenson, J., Hadwin, J. A., & Norgate, Meaney, F. J., Kurzius-Spencer, M., … Cunniff, C.
R. (2012). Anxiety and depression in academic per- (2017). DSM criteria that best differentiate intellec-
formance: An exploration of the mediating factors tual disability from autism spectrum disorder. Child
of worry and working memory. School Psychology Psychiatry and Human Development, 48, 537–545.
International, 33, 433–449. Pelco, L. E., Ward, S. B., Coleman, L., & Young, J. (2009).
Ozonoff, S., Goodlin-Jones, B., & Solomon, M. (2005). Teacher ratings of three psychological report styles.
Evidence-based assessment of autism spectrum disor- Training and Education in Professional Psychology,
ders in children and adolescents. Journal of Clinical 3, 19–27.
Child and Adolescent Psychology, 34, 523–540. Pelham, W. E., Fabiano, G. A., & Massetti, G. M. (2005).
Paikoff, R. L., & Brooks-Gunn, J. (1991). Do parent-child Evidence-based assessment of Attention-Deficit
relationships change during puberty? Psychological Hyperactivity Disorder in children and adolescents.
Bulletin, 110, 47–66. Journal of Clinical Child and Adolescent Psychology,
Papadopoulos, L., Bor, R., & Stanion, P. (1997). 34, 477–505.
Genograms in counselling practice: A review (part 1). Pelham, W. E., Fabiano, G. A., & Massetti, G. M. (2005).
Counseling Psychology Quarterly, 10, 17–28. Evidence-based assessment of attention deficit hyper-
Pardini, D. A. (2008). Novel insights into longstand- activity disorder in children and adolescents. Journal
ing theories of bidirectional parent–child influ- of Clinical Child and Adolescent Psychology, 34,
ences: Introduction to the special section. Journal of 449–476.
Abnormal Child Psychology, 36(5), 627–631. Penny, A. M., Waschbusch, D. A., Klein, R. M., Corkum,
Pardini, D. A., Lochman, J. E., & Powell, N. (2007). The P., & Eskes, G. (2009). Developing a measure of
development of callous-unemotional traits and antiso- sluggish cognitive tempo for children: Content valid-
cial behavior in children: Are there shared or unique ity, factor structure, and reliability. Psychological
predictors. Journal of Clinical Child & Adolescent Assessment, 21, 380–389.
Psychology, 36, 319–333. Penny, S. R., & Skilling, T. A. (2012). Moderators of
Parent, J., Forehand, R., Dunbar, J. P., Watson, K. H., informant agreement in the assessment of adolescent
Reising, M. M., Seehuus, M., & Compas, B. E. (2014). psychopathology: Extension to a forensic sample.
Parent and adolescent reports of parenting when a Psychological Assessment, 24, 386–401.
parent has a history of depression: Associations with Percy, A., McCrystal, P., & Higgins, K. (2008).
observations of parenting. Journal of Abnormal Child Confirmatory factor analysis of the Adolescent Self-
Psychology, 42, 173–183. Report Strengths and Difficulties Questionnaire.
Park, E., & Kim, J. (2016). Factor structure of the European Journal of Psychological Assessment, 24,
Childhood Autism Rating Scale as per DSM-5. 43–48.
Pediatrics International, 58, 139–145. Perry, M. A. (1990). The interview in developmental
Park, J. L., Hudec, K. L., & Johnson, C. (2017). Parental assessment. In J. H. Johnson & J. Goldman (Eds.),
ADHD symptoms and parenting behaviors: A meta- Developmental assessment in clinical child psy-
analytic review. Clinical Psychology Review, 56, 25–39. chology: A handbook (pp. 58–77). New York, NY:
Parker, J. G., & Asher, S. R. (1987). Peer relations and Pergamon Press.
later personal adjustment: Are low-accepted children Petersen, N. S., Kolen, M. J., & Hoover, H. D. (1989).
at risk? Psychological Bulletin, 102, 357–389. Scaling, norming, and equating. In R. L. Linn (Ed.),
Pas, E. T., Bradshaw, C. P., & Mitchell, M. M. (2011). Educational measurement (3rd ed., pp. 221–262).
Examining the validity of office discipline referrals as New York, NY: Macmillan.
an indicator of student behavior problems. Psychology Peterson, D. R. (1961). Behavior problems of middle
in the Schools, 48, 541–555. childhood. Journal of Consulting Psychology, 25,
Patel, J., Lukkes, J. L., & Shekhar, A. (2018). Overview of 205–209.
genetic models of autism spectrum disorders. Progress Pfiffner, L. J., & McBurnett, K. (2006). Family corre-
in Brain Research, 241, 1–36. lates of comorbid anxiety disorders in children with
Patterson, G. R. (1982). Coercive family process. Eugene, attention deficit hyperactivity disorder. Journal of
OR: Castalia. Abnormal Child Psychology, 34, 725–736.
Patterson, G. R., & Yoerger, K. (1993). Developmental Phares, E. J. (1984). Clinical psychology: Concepts, meth-
models for delinquent behavior. In S. Hodgins (Ed.), ods, and profession. Homewood, IL: Dorsey Press.
Mental disorder and crime (pp. 140–172). Newbury Phelan, J. C., Lucas, J. W., Ridgeway, C. L., & Taylor,
Park: Sage. C. J. (2014). Stigma, status, and population health.
Social Science & Medicine, 103, 15–23.
References 417
Piacentini, J., Robper, M., Jensen, P., Lucas, C., Fisher, Antisocial Process Screening Device (APSD) with
P., Bird, H., … Dulcan, M. (1999). Informant-based justice-involved adolescents. Criminal Justice and
determinants of symptom attenuation in structured Behavior, 33, 26–55.
child psychiatric interviews. Journal of Abnormal Pressman, L. J., Loo, S. K., Carpenter, E. M., Asarnow,
Child Psychology, 27, 417–428. J. R., Lynn, D., McCracken, J. T., … Smalley, S. L.
Piacentini, J. C., Cohen, P., & Cohen, J. (1992). (2006). Relationship of family environment and
Combining discrepant diagnostic information from parental psychiatric diagnosis to impairment in
multiple sources: Are complex algorithms better than ADHD. Journal of the American Academy of Child
simple ones? Journal of Abnormal Child Psychology, and Adolescent Psychiatry, 45, 346–354.
20, 51–63. Prinstein, M. J. (2007). Assessment of adolescents’ prefer-
Piasecki, T. M., Hufford, M. R., Sohan, M., & Trull, T. J. ence- and reputation-based peer status using sociomet-
(2007). Assessing clients in their natural envrioen- ric experts. Merrill-Palmer Quarterly, 53(2), 243–261.
ments with electronic diaries: Rationale, benefits, Prinstein, M. J., & Cillessen, A. H. N. (2003). Forms
limitations, and barriers. Psychological Assessment, and functions of adolescent peer aggression associ-
19, 25–43. ated with high levels of peer status. Merrill-Palmer
Pina, A. A., Silverman, W. K., Saavedra, L. M., & Weems, Quarterly, 49, 310–342.
C. F. (2001). An analysis of the RCMAS lie scale in a Prinzie, P., Onghena, P., & Hellinckx, W. (2007).
clinic sample of anxious children. Journal of Anxiety Reexamining the Parenting Scale: Reliability, factor
Disorders, 15, 443–457. structure, concurrent validity of a scale for assess-
Pinderhughes, E. E., Dodge, K. A., Bates, J. E., Pettit, ing the discipline practices of mothers and fathers of
G. S., & Zelli, A. (2000). Discipline responses: elementaryschool-aged children. European Journal of
Influences of parents’ socioeconomic status, ethnicity, Psychological Assessment, 23, 24–31.
beliefs about parenting, stress, and cognitive- emo- Puig-Antich, J., Blau, S., Marx, N., Greenhill, L. L., &
tional processes. Journal of Family Psychology, 14, Chambers, W. (1978). Prepubertal major depressive
280–400. disorders. Journal of the American Academy of Child
Pine, D. S., Costello, E. J., Dahl, R., James, R., Leckman, and Adolescent Psychiatry, 17, 695–707.
J. F., Klein, R., … Zeanah, C. H. (2011). Increasing Quay, H. C. (1986). Classification. In H. C. Quay & J. S.
the developmental focus in DSM-5: Broad issues and Werry (Eds.), Psychopathological disorders of child-
specific potential applications in anxiety. In D. A. hood (3rd ed., pp. 1–34). New York, NY: Wiley.
Regier, W. E. Narrow, E. A. Kuhl, & D. J. Kupfer Rabin, A. I. (1986). Concerning projective techniques.
(Eds.), The conceptual evolution of DSM-5 (pp. 305– In A. I. Rabin (Ed.), P rojective techniques for ado-
321). Arlington, VA: American Psychiatric Publishing. lescents and children (pp. 3–13). New York, NY:
Pinquart, M. (2017). Associations of parenting dimensions Springer.
and styles with externalizing problems of children and Racz, S. J., King, K. M., Wu, J., Witkiewitz, K., McMahon,
adolescents: An update meta-analysis. Developmental R. J., & The Conduct Problems Prevention Research
Psychology, 53, 873–932. Group. (2013). The predictive utility of a brief kinder-
Piotrowski, C. (2015). On the decline of projective tech- garten screening measure of child behavior problems.
niques in professional psychology training. North Journal of Consulting and Clinical Psychology, 81,
American Journal of Psychology, 17, 259–266. 588–599.
Piotrowski, C. (2018). Sentence completion methods: A Raines, T. C., Dever, B. V., Kamphaus, R. W., & Roach,
summary review of 70 survey-based studies of train- A. T. (2012). Universal screening for behavioral and
ing and preofession settings. Journal of Projective emotional risk: A promising method for reducing dis-
Psychology and Mental Health, 25, 60–75. proportionate placement in special education. Journal
Porcu, M., & Giambona, F. (2017). Introduction to latent of Negro Education, 81, 283–296.
class analysis with applications. Journal of Early Randolph, K. A., & Radey, M. (2011). Measuring par-
Adolescence, 37, 129–158. enting practices among parents of elementary school-
Porter, B., & O’Leary, K. D. (1980). Marital discord and age youth. Research on Social Work Practice, 21,
childhood behavior problems. Journal of Abnormal 88–97.
Child Psychology, 8, 287–295. Rapport, M. D., Chung, K. M., Shore, G., Denney, C. B.,
Power, T. J., Andrews, T. J., Eiraldi, R. B., Doherty, B. J., & Isaacs, P. (2000). Upgrading the science and tech-
Ikeda, M. J., DuPaul, G. J., & Landau, S. (1998). nology of assessment and diagnosis: Laboratory and
Evaluating attention deficit hyperactivity disorder clinic-based assessment of children with ADHD.
using multiple informants: The incremental utility of Journal of Clinical Child Psychology, 29, 555–568.
combining teacher with parent reports. Psychological Rapport, M. D., Denney, C. B., Chung, K., & Hustace, K.
Assessment, 10, 250–260. (2001). Internalizing behavior problems and scholas-
Poythress, N. G., Dembo, R., Wareham, J., & tic achievement in children: Cognitive and behavioral
Greenbaum, P. E. (2006). Construct validity of the pathways as mediators of outcome. Journal of Clinical
Youth Psychopathic Traits Inventory (YPI) and the Child & Adolescent Psychology, 30, 536–551.
418 References
Reed, M. L., & Edelbrock, C. (1983). Reliability and Reynolds, W. M. (2002). Reynolds adolescent depression
validity of the direct observational form of the Child scale, 2nd edition (RADS-2). Lutz, FL: Psychological
Behavior Checklist. Journal of Abnormal Child Assessment Resources.
Psychology, 11, 521–530. Reynolds, W. M. (2010). Reynolds child depression
Reese, J. B., Viglione, D. J., & Giromini, L. (2014). A scale, 2nd edition (RCDS-2). Lutz, FL: Psychological
comparison between comprehensive system and an Assessment Resources.
early version of the rorschach performance assess- Rhoades, K. A., & O’Leary, S. G. (2007). Factor structure
ment system administration with outpatient children and validity of the Parenting Scale. Journal of Clinical
and adolescents. Journal of Personality Assessment, Child and Adolescent Psychology, 36, 137–146.
96, 515–522. Riccio, C., & Reynolds, C. (2003). The assessment
Reich, W., & Earls, F. (1987). Rules for making psy- of attention via continuous performance tests. In
chiatric diagnoses in children on the basis of mul- Handbook of psychological and educational assess-
tiple sources of information: Preliminary strategies. ment of children: Personality, behavior, and context
Journal of Abnormal Child Psychology, 1S, 601–616. (2nd ed., pp. 291–319). New York, NY: Guilford
Reitman, D., Currier, R. O., Hupp, S. D. A., Rhode, Press.
P. C., Murphy, M. A., & O’Callahan, P. M. (2001). Rice, F., Riglin, L., Lomax, T., Souter, E., Potter, R.,
Psychometric characteristics of the Parenting Scale Smith, D. J., … Thapar, A. (2019). Adolescent and
in a Head Start population. Journal of Clinical Child adult differences in major depression symptom pro-
Psychology, 30, 514–524. files. Journal of Affective Disorders, 243, 175–181.
Rende, R., & Weissman, M. M. (1999). Assessment of Richters, J. E. (1992). Depressed mothers as informants
family history of psychiatric disorder. In D. Shaffer, about their children: A critical review of the evi-
C. P. Lucas, & J. E. Richters (Eds.), Diagnostic dence for distortion. Psychological Bulletin, 112(3),
assessment in child and adolescent psychopathology 485–499.
(pp. 230–255). New York, NY: Guilford Press. Ringel, J., & Sturm, R. (2001). National estimates of
Renk, K., & Phares, V. (2004). Cross-informant ratings mental health utilization and expenditure for children
of social competence in children and adolescents. in 1998. Journal of Behavioral Health Services and
Clinical Psychology Review, 24, 239–254. Research, 28, 319–333.
Renzaho, A., Mellor, D., McCabe, M., & Powell, M. Ringoot, A. P., van der Ende, J., Jansen, P. W., Measelle,
(2013). Family functioning, parental psychologi- J. R., Busten, M., So, P., … Tiemeier, H. (2015).
cal distress and child behaviours: Evidence from Why mothers and young children agree or disagree
the Victorian Child Health and Wellbeing Study. in their reports of the child’s problem behavior. Child
Australian Psychologist, 48, 217–225. Psychiatry and Human Development, 46, 913–927.
Rescorla, L. A., Ginzburg, S., Achenbach, T. M., Ivanova, Rios, J., & Morey, L. C. (2013). Detecting feigned ADHD
M. Y., Almquvist, F., Begovac, I., … Verhulst, F. C. in later adolescence: An examination of three PAI-A
(2013). Cross-informant agreement between parent- negative distortion indicators. Journal of Personality
reported and adolescent self-reported problems in 25 Assessment, 95, 594–599.
societies. Journal of Clinical Child and Adolescent Roberts, B. W., & DelVecchio, W. F. (2000). The rank-
Psychology, 42, 262–273. order consistency of personality traits from childhood
Reynolds, C. R., & Kaiser, S. M. (1990). Test bias in to old age: A quantitative review of longitudinal stud-
psychological assessment. In T. B. Gutkin & C. R. ies. Psychological Bulletin, 126, 3–25.
Reynolds (Eds.), The handbook of school psychology Roberts, M. A., Ray, R. S., & Roberts, R. J. (1984). A
(2nd ed., pp. 487–525). New York, NY: John Wiley playroom observational procedure for assessing
& Sons. hyperactive boys. Journal of Pediatric Psychology, 9,
Reynolds, C. R., & Kamphaus, R. W. (1992). Behavior 177–191.
assessment system for children (BASC). Circle Pines, Robertson, D., & Hyde, J. (1982). The factorial validity
MN: American Guidance Services. of the Family Environment Scale. Educational and
Reynolds, C. R., & Kamphaus, R. W. (2004). Behavior Psychological Measurement, 42, 1233–1241.
assessment system for children, 2nd edition (BASC-2). Robins, L. N. (1966). Deviant children grown up.
Circle Pines, MN: American Guidance Services. Baltimore, MD: Williams and Wilkins.
Reynolds, C. R., & Kamphaus, R. W. (2015). Behavior Robins, L. N., Cottler, L. B., Bucholz, K. K., Compton,
assessment system for children, 3rd edition (BASC-3). W., North, C., & Rourke, K. (2000). The diagnostic
Toronto, ON: Pearson. interview schedule for DSM-IV (DIS-IV). St. Louis,
Reynolds, C. R., & Richmond, B. O. (2008). Revised chil- MO: Washington University School of Medicine.
dren’s manifest anxiety scale (2nd ed.). Torrance, CA: Robinson, E. A., & Eyberg, S. M. (1981). The
Western Psychological Services. Dyadic Parent Child Interaction Coding System:
Reynolds, W. M. (1986). Reynolds adolescent depres- Standardization and validation. Journal of Consulting
sion scale. Odessa, FL: Psychological Assessment and Clinical Psychology, 49, 245–250.
Resources. Rockett, J. L., Murrie, D. C., & Boccacini, M. T. (2007).
Diagnostic labeling in juvenile justice settings: Do
References 419
psychopathy and conduct disorder findings influence Sanders, M. R., Markie-Dadds, C., Tully, L. A., & Bor,
clinicians? Psychological Services, 4, 107–122. W. (2000). The Triple P-Positive Parenting Program:
Roeser, R. W. (2001). To cultivate the positive … A comparison of enhanced, standard, and self-directed
Introduction to the special issue on schooling and behavioral family intervention for parents of chil-
mental health issues. Journal of School Psychology, dren with early onset conduct problems. Journal of
39, 99–110. Consulting and Clinical Psychology, 68, 624–640.
Roid, G. (2003). Stanford-binet intelligence scale: Fifth Sands, A., Thompson, E. J., & Gaysina, D. (2017). Long-
edition. Itasca, IL: Riverside. term influences of parental divorce on offspring affec-
Ronning, J. A., Handegaard, B. H., Sourander, A., & tive disorders: A systematic review and meta-analysis.
Morch, W. T. (2004). The strengths and difficulties Journal of Affective Disorders, 218, 105–114.
self-report questionnaire as a screening instrument in Sandstrom, M. J., & Cillessen, A. H. N. (2006). Likeable
Norwegian community samples. European Child and versus popular: Distinct implications for adolescent
Adolescent Psychiatry, 13, 73–82. adjustment. International Journal of Behavioral
Rorschach, H. (1951). P sychodiagnostics: A diagnostic Development, 30, 305–314.
test based on perception. (P. Lemkau & B. Kronenberg, Sanford, K., Bingham, C. R., & Zucker, R. A. (1999).
Trans.). New York, NY: Grune & Stratton. Validity issues with the Family Environment Scale:
Ross, A. O. (1981). Child behavior therapy: Principles, Psychometric resolution and research application with
procedures, and empirical basis. New York, NY: alcoholic families. Psychological Assessment, 11,
Wiley. 315–325.
Rotter, J. B., & Rafferty, J. E. (1950). Manual for the Sanislow, C. A., Pine, D. S., Quinn, K. J., Kozak, M. J.,
Rotter Incomplete Sentences Blank, College Form. Garvey, M. A., Heinssen, R. K., … Cuthber, B. N.
New York, NY: The Psychological Corporation. (2010). Developing constructs for psychopathol-
Rowe, R., Costello, E. J., Angold, A., Copeland, W. E., ogy research: Research domain criteria. Journal of
& Maughan, B. (2010). Developmental pathways in Abnormal Psychology, 119, 631–639.
oppositional defiant disorder and conduct disorder. Sarason, I. G., Johnson, J. H., & Siegel, J. M. (1978).
Journal of Abnormal Psychology, 119, 726–738. Assessing the impact of life changes: Development
Rowland, A. S., Skipper, B. J., Rabiner, D. L., Qeadan, of the Life Experiences Survey. Journal of Consulting
F., Campbell, A. J. N., & Umbach, D. M. (2018). and Clinical Psychology, 46, 932–946.
Attention-deficit/Hyperactivity Disorder (ADHD): Sargent, J., Liebman, R., & Silber, M. (1985). Family
Interaction between socioeconomic status and paren- therapy for anorexia nervosa. In D. M. Garner &
tal history of ADHD determines prevalence. Journal P. E. Garfinkel (Eds.), Handbook of psychotherapy
of Child Psychology and Psychiatry, 59, 213–222. for anorexia nervosa and bulimia (pp. 257–279).
Ruchkin, V., Jones, S., Vermeiren, R., & Schwab-Stone, New York, NY: Guilford Press.
M. (2008). The strengths and difficulties question- Sattler, J. M. (1988). Assessment of children (3rd ed.). San
naire: The self-report version in American urban Diego, CA.
and suburban youth. Psychological Assessment, 20, Schanding, G. T., Jr., & Nowell, K. P. (2013). Universal
175–182. screening for emotional and behavioral problems:
Ruchkin, V., Koposov, R., & Schwab-Stone, M. (2007). Fitting a population-based model. Journal of Applied
The strength and difficulties questionnaire: Scale vali- School Psychology, 29, 104–119.
dation with Russian adolescents. Journal of Clinical Schaughency, E. A., & Rothlind, J. (1991). Assessment
Psychology, 63, 861–869. and classification of attention-deficit hyperactivity dis-
Rudolph, K. D., Hammen, C., & Burge, D. (1994). order. School Psychology Review, 20, 187–202.
Interpersonal functioning and depressive symptoms Schneider, W. J., Lichtenberger, E. O., Mather, N., &
in childhood: Addressing the issues of specificity and Kaufman, N. L. (2018). Essentials of report writing
comorbidity. Journal of Abnormal Child Psychology, (2nd ed.). Hoboken, NJ: Wiley.
22, 355–373. Schniering, C. A., & Lyneham, H. J. (2007). The chil-
Rusby, J. C., Taylor, T. K., & Foster, E. M. (2007). A dren’s automatic thoughts scale in a clinical sample:
descriptive study of school discipline referrals in first Psychometric properties and clinical utility. Behaviour
grade. Psychology in the Schools, 44, 333–350. Research and Therapy, 45, 1931–1940.
Ruscio, J. (2000). The role of complex thought in clinical Schniering, C. A., & Rapee, R. M. (2002). Development
prediction: Social accountability and the need for cog- and validation of a measure of children’s automatic
nition. Journal of Consulting and Clinical Psychology, thoughts: The Children’s Automatic Thoughts Scale.
68, 145–154. Behaviour Research and Therapy, 40, 1091–1109.
Rutter, M., & Garmezy, N. (1983). Developmental psy- Scholte, E., Van Berckelaer-Onnes, I., & Van der Ploeg, J.
chopathology. In E. M. Hetherington (Ed.), Manual of (2008). Rating scale to screen symptoms of psychiat-
child psychology, IV: Social and personality develop- ric disorders in children. European Journal of Special
ment (pp. 775–912). New York, NY: Wiley. Needs Education, 23, 47–62.
Ryan, R. M., & Kuczkowski, R. (1994). The imaginary Schopler, E., Van Bourgondien, M. E., Wellman, G. J.,
audience, self-consciousness, and public individuation & Love, S. R. (2010). Childhood autism rating scale
in adolescence. Journal of Personality, 62, 219–238.
420 References
(2nd ed.). Los Angeles, CA: Western Psychological Kamphaus (Eds.), Handbook of psychological and
Services. educational assessment of children: Personality,
Schorre, B. E. H., & Vandvik, I. H. (2004). Global assess- behavior, and context (pp. 343–363). New York, NY:
ment of psychosocial functioning in child and ado- Guilford Press.
lescent psychiatry: A review of three unidimensional Sharma-Patel, K., Felton, B., Brown, E. J., Zlotnik, D.,
scales (CGAS, GAF, GAPD). European Child and Campbell, C., & Yedlin, J. (2011). Pediatric post-
Adolescent Psychiatry, 13, 273–286. traumatic stress disorder. In M. Crozier, S. Gillihan,
Schwean, V. L., Oakland, T., Weiss, L. G., Saklofske, & M. Powers (Eds.), Handbook of child and adoles-
D. H., Holdnack, J. A., & Prifitera, A. (2006). Report cent anxiety disorders (pp. 303–321). New York, NY:
writing: A child-centered approach. In L. G. Weiss, Springer New York.
J. A. Holdnack, A. Prifitera, & D. H. Saklofske Sheehan, D. V., Sheehan, K. H., Shytle, R. D., Janavs, J.,
(Eds.), WISC-IV: Advanced clinical interpretation Bannon, Y., Rogers, J. E., … Wilkinson, B. (2010).
(pp. 371–419). Ann Arbor, Michigan: ProQuest Reliability and validity of the Mini International
Ebook Central. Neuropsychiatric Interview for Children and
Scott, T. M., Anderson, C., Mancil, R., & Alter, P. (2009). Adolescents (MINI-KID). The Journal of Clinical
Function-based supports for individual students in Psychiatry, 71, 313–326.
school settings. In Handbook of positive behavior Shellenberger, S., Dent, M. M., Davis-Smith, M., Seale,
support (pp. 421–441). Boston, MA: Springer. J. P., Weintraut, R., & Wright, T. (2007). Cultural
Semrud-Clikeman, M. (1990). Assessment of childhood genogram: A tool for teaching and practice. Families,
depression. In C. R. Reynolds & R. W. Kamphaus Systems, & Health, 25, 367–381.
(Eds.), Handbook of psychological and educational Shelton, K. K., Frick, P. J., & Wootton, J. (1996). The
assessment of children: Personality, behavior, and assessment of parenting practices in families of ele-
context (pp. 279–297). New York, NY: Guilford mentary school-aged children. Journal of Clinical
Press. Child Psychology, 25, 317–327.
Severson, H. H., Walker, H. M., Hope-Doolittle, J., Shiner, R., & Caspi, A. (2003). Personality differences in
Kratochwill, T. R., & Gresham, F. M. (2007). childhood and adolescence: Measurement, develop-
Proactive, early screening to detect behaviorally at- ment, and consequences. Journal of Child Psychology
risk students: Issues, approaches, emerging innova- and Psychiatry, 44, 2–32.
tions, and professional practices. Journal of School Shrout, P. E., Stadler, G., Lane, S. P., McClure,
Psychology, 45, 193–223. M. J., Jackson, G. L., Clavél, F. D., … Bolger, N.
Shaffer, D., Fisher, P., Lucas, C. P., Dulcan, M. K., & (2018). Initial evaluation bias in subjective reports.
Schwab-Stone, M. E. (2000). NIMH diagnostic Proceedings of the National Academy of Sciences of
interview schedule for children version IV (NIMH the United States of America, 115, E15–E23.
DISC-IV): Description, differences from previous Silverman, W. K. (1993). DSM and classification of
versions, and reliability of some common diagno- anxiety disorders in children and adults. In C. G. Last
ses. Journal of the American Academy of Child and (Ed.), Anxiety across the lifespan: A developmental
Adolescent Psychiatry, 39, 28–38. perspective (pp. 7–36). New York, NY: Springer.
Shaffer, D., Fisher, P., Piacentini, J., Schwab-Stone, M., & Silverman, W. K., & Albano, A. M. (1996). Anxiety dis-
Wicks, J. (1991). NIMH diagnostic interview schedule orders interview schedule for children for DSM–
for children-version 2.3. New York, NY: New York IV: (child and parent versions). SanAntonio, TX:
State Psychiatric Institute. Psychological Corporation/Graywind.
Shaffer, D., Garland, A., Gould, M., Fisher, P., & Silverman, W. K., & Nelles, W. B. (1989). The stability
Trautman, P. (1988). Preventing teenage suicide: A of mothers’ rating of child fearfulness. Journal of
critical review. Journal of the American Academy of Anxiety Disorders, 3, 1–5.
Child and Adolescent Psychiatry, 27, 675–687. Silverman, W. K., & Ollendick, T. H. (2005). Evidence-
Shaffer, D., Gould, M. S., Brasic, J., Ambrosini, P., Fisher, based assessment of anxiety and its disorders in chil-
P., Bird, H., & Aluwahlia, S. (1983). A children’s dren and adolescents. Journal of Clinical Child and
global assessment scale (CGAS). Archives of General Adolescent Psychology, 34, 380–411.
Psychiatry, 40, 1228–1231. Singleton, L. C., & Asher, S. R. (1979). Peer preferences
Shaffer, D., Schwab-Stone, M., Fisher, P., Cohen, P., and social interaction among third-grade children in
Piacentini, J., Davies, M., … Regier, D. (1993). an integrated school district. Journal of Educational
The Diagnostic Interview Schedule for Children- Psychology, 69, 330–336.
Revised Version (DISC-R): I. Preparation, field test- Skinner, B. F. (1963). Operant behavior. American
ing, interrater reliability, and acceptability. Journal Psychologist, 18(8), 503–515.
of the American Academy of Child and Adolescent Smith, B. H., Barkley, R. A., & Shapiro, C. J. (2007).
Psychiatry, 32, 643–650. Attention-deficit/hyperactivity disorder. In E. J. Mash
Shapiro, E. S. (1987). Behavioral assessment in school & R. A. Barkley (Eds.), Assessment of childhood dis-
psychology. Hillsdale, NJ: Lawrence Erlbaum. orders (4th ed., pp. 53–131). New York: Guilford.
Shapiro, E. S., & Skinner, C. H. (1990). Principles of Smith, S. R. (2007). Making sense of multiple informants
behavioral assessment. In C. R. Reynolds & R. W. in child and adolescent psychopathology: A guide for
References 421
clinicians. Journal of Psychoeducational Assessment, of childhood disorders (3rd ed., pp. 335–407).
25, 139–149. New York, NY: Guilford Press.
Smith, S. R., Baity, M. R., Knowles, E. S., & Hilsenroth, Steele, R. G., Nesbitt-Daly, J. S., Daniel, R. C., &
M. J. (2001). Assessment of disordered thinking in Forehand, R. (2005). Factor structure of the Parenting
children and adolescents: The Rorschach Perceptual- Scale in a low-income African-American sample.
Thinking Index. Journal of Personality Assessment, Journal of Child and Family Studies, 14, 535–549.
77, 447–463. Steinberg, L., Lamborn, S., Dornbusch, S., & Darling,
Snyder, J. (1991). Discipline as a mediator of the impact of N. (1992). Impact of parenting practices on adoles-
maternal stress and mood on child conduct problems. cent achievement: Authoritative parenting, school
Development and Psychopathology, 3(3), 263–276. involvement, and encouragement to succeed. Child
Snyder, M., Tanke, E. D., & Berscheid, F. (1977). Social Development, 63, 1266–1281.
perception and interpersonal behavior on the self- Stiffler, M. C., & Dever, R. W. (2015). Universal emo-
fulfilling nature of social stereotypes. Journal of tional and behavioral screening for children and
Personality and Social Psychology, 35, 656–666. adolescents: Prospects and pitfalls. New York, NY:
Solis, M. L., & Abidin, R. R. (1991). The Spanish ver- Springer.
sion of the Parenting Stress Index: A psychometric Stockings, E., Degenhardt, L., Lee, Y. Y., Mihalopoulos,
study. Journal of Clinical Child Psychology, 20, C., Liu, A., Hobbs, M., & Patton, G. (2015). Symptom
372–378. screening scales for detecting major depressive disor-
Sourander, A., Helstelae, L., & Helenius, H. (1999). der in children and adolescents: A systematic review
Parent-adolescent agreement on emotional and behav- and meta-analysis of reliability, validity and diag-
ioral problems. Social Psychiatry and Psychiatric nostic utility. Journal of Affective Disorders, 174,
Epidemiology, 34, 657–663. 447–463.
Southam-Gerow, M. A., & Chorpita, B. F. (2007). Anxiety Stone, A. L., Becker, L. G., Huber, A. M., & Catalano,
in children and adolescents. In E. J. Mash & R. A. R. F. (2012). Review of risk and protective factors of
Barkley (Eds.), A ssessment of childhood disorders substance use and problem use in emerging adulthood.
(4th ed., pp. 347–397). New York, NY: Guilford. Addictive Behaviors, 37, 747–775.
Spanier, G. B. (1976). Measuring dyadic adjustment: New Stone, L. L., van Daal, C., van der Maten, M., Engels,
scales for assessing the quality of marriage and similar R. C., Janssens, J. M. A. M., & Otten, R. (2014). The
dyads. Journal of Marriage and the Family, 38(1), 15. berkeley puppet interview: A screening instrument for
Sparrow, S. S., Cicchetti, D. V., & Saulnier, C. A. (2016). measuring psychopathology in young children. Child
Vineland adaptive behavior scales (3rd ed.). Circle & Youth Care Forum, 43, 211–225.
Pines, MN: Pearson. Stone, W. L., Coonrod, E. E., Pozdol, S. L., & Turner,
Spiel, C. F., Evans, S. W., & Langbarg, J. M. (2014). L. M. (2003). The Parent Interview for Autism–clinical
Evaluating the content of Individualized Education version (PIA–CV): A measure of behavioral change
Programs and 504 Plans of young adolescents with for young children with autism. Autism, 7, 9–30.
attention-deficit/hyperactivity disorder. School Stone, W. L., & Hogan, K. L. (1993). A structured parent
Psychology Quarterly, 29, 452–468. inter-view for identifying young children with autism.
Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. Journal of Autism and Developmental Disorders, 23,
(1970). STAI manual for the State-Trait Inventory. 639–652.
Palo Alto, CA: Consulting Psychologist Press. Storey, R., Gapen, M., & Sacco, J. S. (2014). Projective
Spitzer, R., & Cantwell, D. (1980). The DSM-III clas- techniques and psychological assessment in disadvan-
sification of psychiatric disorders of infancy, child- taged communities. International Journal of Applied
hood, and adolescence. Journal of the American Psychoanalytic Studies, 11, 114–129.
Academy of Child and Adolescent Psychiatry, 19, Storment, M., & Reinke, W. M. (2013). Implementing tier
356–370. 2 social behavioral interventions: Current issues, chal-
Splett, J. W., Smith-Millman, M., Raborn, A., Brann, lenges, and promising approaches. Journal of Applied
K. L., Flaspohler, P. D., & Maras, M. A. (2018). School Psychology, 29, 121–125.
Student, teacher, and classroom predictors of between- Strange, D., van Papendrecht, H. H., Crawford, E.,
teacher variance of students’ teacher-rated behavior. Candel, I., & Hayne, H. (2010). Size doesn’t mat-
School Psychology Quarterly, 33, 460–468. ter: Emotional content does not determine the size of
Stanger, C., Dumenci, L., Kamon, J., & Burstein, M. objects in children’s drawings. Psychology, Crime,&
(2004). Parenting and children’s externalizing prob- Law, 16, 459–476.
lems in substance-abusing families. Journal of Clinical Strauss, C. C., Lahey, B. B., Frick, P. J., Frame, C. L., &
Child and Adolescent Psychology, 33, 590–600. Hynd, G. W. (1988). Peer social status of children with
Stark, K. D., Sander, J., Hauser, M., Simpson, J., anxiety disorders. Journal of Consulting and Clinical
Schnoebelen, S., Glenn, R., … Molnar, J. (2006). Psychology, 56, 137–141.
Depressive disorders during childhood and adoles- Stredney, R. V., & Ball, J. D. (2005). The utility of the
cence. In E. J. Mash & R. A. Barkley (Eds.), Treatment Rorschach Coping Deficit Index as a measure of
422 References
depression and social skills deficits in children and Effects of paternal contact. Journal of Child and
adolescents. Assessment, 12, 295–302. Family Studies, 5, 229–240.
Stringaris, A., & Goodman, R. (2009). Longitudinal Tarescavage, A. M., & Ben-Porath, Y. S. (2014).
outcome of youth oppositionality: Irritable, head- Psychotherapeutic outcomes measures: A criti-
strong, and hurtful behaviors have distinctive predic- cal review for practitioners. Journal of Clinical
tions. Journal of the American Academy of Child and Psychology, 70, 808–830.
Adolescent Psychiatry, 48, 404–412. Taylor, C., Allen, A., Kilgus, S., von der Embse, N., &
Sullivan, A. D. W., Parent, J., Forehand, R., & Compas, Garbacz, S. A. (2016). Development and valida-
B. E. (2018). Does interparental conflict decrease fol- tion of a parent version of the Social, Academic,
lowing changes in observed parenting from a preven- and Emotional Behavior Risk Screener (SAEBRS).
tive program? Behaviour Research and Therapy, 106, Assessment for Effective Intervention, 42, 186–192.
64–70. Tellegen, A., & Ben-Porath, Y. S. (1992). The new uni-
Sund, A. M., Larsson, B., & Wichstrøm, L. (2011). form T-scores for the MMPI-2: Rationale, derivation,
Prevalence and characteristics of depressive disor- and appraisal. Psychological Assessment, 4, 145–155.
ders in early adolescents in central Norway. Child and Teufel-Shone, N., Staten, L. K., Irwin, S., Rawiel, U.,
Adolescent Psychiatry and Mental Health, 5, 28–40. Bravo, A. B., & Waykayuta, S. (2005). Family cohe-
Sundberg, N. D. (1961). The practice of psychologi- sion and conflict in an American Indian community.
cal testing in clinical settings in the United States. American Journal of Health Behavior, 29, 413–422.
American Psychologist, 16, 79–83. Reprinted in L. D. Thompson, C. E. (1949). Manual for thematic appercep-
Goodstein & R. I. Lanyon (Eds.), Readings in person- tion test: Thompson modification. Cambridge, MA:
ality assessment. New York, NY: Wiley. Harvard University Press.
Suveg, C., Payne, M., Thomassin, K., & Jacob, M. L. (2010). Tibbetts, S. G., & Piquero, A. R. (1999). The influence
Electronic diaries: A feasible method of assessing emo- of gender, low birth weight, and disadvantaged envi-
tional experiences in youth? Journal of Psychopathology ronment in predicting early onset of offending: A test
and Behavioral Assessment, 32, 57–67. of Moffitt’s interactional hypothesis. Criminology, 37,
Suveg, C., Zeman, J., Flannery-Schroeder, E., & Cassano, 843–877.
M. (2005). Emotion socialization in families of chil- Tobin, T. J., & Sugai, G. M. (1999). Using sixth-grade
dren with an anxiety disorder. Journal of Abnormal school records to predict school violence, chronic dis-
Child Psychology, 33, 145–155. cipline problems, and high school outcomes. Journal
Swensen, C. H. (1968). Empirical evaluations of human of Emotional and Behavioral Disorders, 7, 40–53.
figure drawings: 1957–1966. Psychological Bulletin, Toth, K., & King, B. H. (2010). Intellectual disability
70, 20–44. Reprinted in L. D. Goodstein & R. I. (mental retardation). In M. K. Dulcan (Ed.), Dulcan’s
Lanyon (Eds.), Readings in personality assessment. textbook of child and adolescent psychiatry (pp. 151–
New York, NY: Wiley. 171). Arlington, VA: American Psychiatric Publishing.
Tackett, J. (2011). Parent informants for child personality: Tripp, G., Schaughency, E., & Clarke, B. (2006). Parent
Agreement, discrepancies, and clinical utility. Journal and teacher rating scales in the evaluation of attention-
of Personality Assessment, 93, 539–544. deficit hyperactivity disorder: Contribution to diag-
Tackett, J. L., Herzhoff, K., Reardon, K. W., Smack, A. J., nosis and differential diagnosis in clinically referred
& Kushner, S. C. (2013). The relevance of informant children. Journal of Developmental & Behavioral
discrepancies for the assessment of adolescent per- Pediatrics, 27, 209–218.
sonality pathology. Clinical Psychology: Science and Tulbure, B. T., Szentagotai, A., Dobrean, A., & David, D.
Practice, 20, 378–392. (2012). Evidence based clinical assessment of child
Takarangi, M. K. T., Garry, M., & Loftus, E. F. (2006). and adolescent social phobia: A critical review of rat-
Dear diary, is plastic better than paper? I can’t remem- ing scales. Child Psychiatry & Human Development,
ber: Comment on Green, Rafaeli, Bolger, Shrout, and 43, 795–820.
Reis (2006). Psychological Methods, 11, 119–122. Tung, I., Li, J. J., Meza, J. I., Jezior, K. L., Klanmahd, J. S.
Talkovsky, A. M., Green, K. L., Osegueda, A., & Norton, V., Hentschel, P. G., … Lee, S. S. (2016). Patterns of
P. J. (2017). Secondary depression in transdiagnostic comorbidity among girls with ADHD: A meta-analy-
group cognitive behavioral therapy among individuals sis. Pediatrics, 138, 1–13.
diagnosed with anxiety disorders. Journal of Anxiety Tupes, E. C., & Christal, R. E. (1961). Recurrent person-
Disorders, 46, 56–64. ality factors based on trait ratings (Technical Report
Tallent, N. (1993). Psychological report writing (4th ed.). ASD-TR-61–97). Lackland Air Force Base, TX:
Englewood Cliffs, NJ: Prentice-Hall. U.S. Air Force.
Tannock, R. (2000). Attention deficit disorders with Tyrrell, M. (2005). School phobia. The Journal of School
anxiety disorders. In T. Brown (Ed.), Attention deficit Nursing, 21, 147–151.
disorders and comorbidities in children, adolescents United States Census Bureau, (2001). Current Population
and adults (pp. 125–170). Washington, DC: American Survey, 2001. Washington, DC: Author.
Psychiatric Press. United States Census Bureau. (2013). American commu-
Tapscott, M., Frick, P. J., Wootton, J., & Kruh, I. (1996). nity survey, 2013 1-year period estimates. Washington,
The intergenerational link to antisocial behavior: DC: Author.
References 423
United States Department of Education. (2006). IDEA erized peer-nomination instrument. Psychological
regulations: Early intervention services. Office of Assessment, 26, 628–641.
Special Education Programs. Retrieved from http:// Viglione, D., Blume-Marcovici, A., Miller, H., Giromini,
idea.ed.gov/ L., & Meyer, G. (2012). An inter-rater reliability study
United States Public Health Service. (2000). Report of for the Rorschach Performance Assessment System.
the surgeon general’s conference on children’s mental Journal of Personality Assessment, 94, 607–612.
health: A national action agenda. Washington, DC: Vuontela, V., Carlson, S., Troberg, A., Fontell, T., Simola,
Department of Health and Human Services. Retrieved P., Saarinen, S., & Aronen, E. T. (2013). Working
from http://www.surgeongeneral.gov/topics/cmh/chil- memory, attention, inhibition, and their relation to
dreport.htm adaptive functioning and behavioral/emotional symp-
Vaillancourt, T., & Hymel, S. (2006). Aggression and toms in school-aged children. Child Psychiatry and
social status: The moderating roles of sex and peer- Human Development, 44, 105–122.
valued characteristics. Aggressive Behavior, 32, Wagner, M., Kutash, K., Duchnowski, A., & Epstein, M.
396–408. (2005). The special education elementary longitudinal
Valla, J., Bergeron, L., & Smolla, N. (2000). The study and the national longitudinal transition study:
Dominic-R: A pictorial interview for 6- to 11-year-old Study designs and implications for children and youth
children. Journal of the American Academy of Child with emotional disturbance. Journal of Emotional and
and Adolescent Psychiatry, 39, 85–93. Behavioral Disorders, 13, 25–41.
Van de Looij-Jansen, P. M., Goedhart, A. W., de Wilde, Waldman, I. D., & Gizer, I. R. (2006). The genetics of
E. J., & Treffers, P. D. A. (2011). Confirmatory fac- attention deficit hyperactivity disorder. Clinical
tor analysis and factorial invariance analysis of the Psychology Review, 26, 396–432.
adolescent self-report Strengths and Difficulties Waldman, I. D., Poore, H. E., van Hulle, C., Rathouz,
Questionnaire: How important are method effects and P. J., & Lahey, B. B. (2016). Externalizing validity of
minor factors? British Journal of Clinical Psychology, a hierarchical dimensional model of child and ado-
50(2), 127–144. lescent psychopathology: Tests using confirmatory
van den Berg, Y. H. M., Lansu, R. A. M., & Cillessen, factor analyses and multivariate behavior genetic
A. H. N. (2015). Measuring social status and social analyses. Journal of Abnormal Psychology, 125,
behavior with peer and teacher nomination methods. 1053–1066.
Social Development, 24, 815–832. Walker, C. A., & Davies, J. (2010). A critical review of
Van Hasselt, V. B., Hersen, M., Bellack, A. S., Rosenblum, the psychometric base of the Child Abuse Potential
N. D., & Lamparski, D. (1979). Tripartite assessment Inventory. Journal of Family Violence, 25, 215–227.
of the effects of systematic desensitization in a multi- Walker, C. E., Milling, L. S., & Bonner, B. L. (1988).
phobic child: An experimental analysis. Journal of Incontinence disorders: Enuresis and encopresis. In
Behavior Therapy and Experimental Psychiatry, 10, D. K. Routh (Ed.), Handbook of pediatric psychology
51–55. (pp. 363–398). New York, NY: Guilford Press.
Van Kammen, W. B., Loeber, R., & Stouthamer-Loeber, Walker, H., & Severson, H. (1990). Systematic screening
M. (1991). Substance use and its relationship to con- for behavior disorders (SSBD). Longmont, CO: Sopris
duct problems and delinquency in young boys. Journal West.
of Youth and Adolescence, 20, 399–413. Walker, H. M., & Severson, H. (1992). Systematic screen-
Van Roy, B. V., Veenstra, M., & Clench-Aas, J. (2008). ing for behavior disorders (2nd ed.). Longmont, CO:
Construct validity of the five-factor Strengths and Sopris West.
Difficulties Questionnaire (SDQ) in pre-, early, and Walker, H. M., Severson, H. H., & Feil, E. G. (2014).
late adolescence. Journal of Child Psychology and Systematic screening for behavior disorders (SSBD):
Psychiatry, 49, 1304–1312. Administrator’s guide. Universal screening for
van Steensel, F. J. A., & Heeman, E. J. (2017). Anxiety preK-9 (2nd ed.). Eugene, OR: Pacific Northwest
levels in children with autism spectrum disorder: A Publishing.
meta-analysis. Journal of Child and Family Studies, Walker, J. L., Lahey, B. B., Russo, M. F., Frick, P. J.,
26, 1753–1767. Christ, M. A. G., McBurnett, K., … Green, S. M.
Vasay, M. W., & Lonigan, C. J. (2000). Considering (1991). Anxiety, inhibition, and conduct disorder in
the clinical utility of performance-based measures children: I. Relations to social impairment. Journal
of childhood anxiety. Journal of Clinical Child of the American Academy of Child & Adolescent
Psychology, 29, 493–508. Psychiatry, 30(2), 187–191.
Verhulst, F. C., Dekker, M. C., & van der Ende, J. (1997). Walsh, C., MacMillan, H. L., & Jamieson, E. (2003).
Parent, teacher, and self-reports as predictors of signs The relationship between parental substance abuse
of disturbance in adolescents: Whose information car- and child maltreatment: Findings from the Ontario
ries the most weight? Acta Psychiatrica Scandinavica, Health Supplement. Child Abuse and Neglect, 27,
96, 75–81. 1409–1425.
Verlinden, M., Veenstra, R., Ringoot, A. P., Jansen, P. W., Walsh, K. S., & Rau, S. (2018). Chapter 6 – Measurement
Raat, H., Hein, H., … Tiemeier, H. (2014). Detecting considerations in pediatric research on autism spec-
bullying in early elementary school with a comput- trum disorders. In A. Shekhar (Ed.), Progress in
424 References
brain research (Vol. 241, pp. 193–220). London, UK: West, M. O., & Prinz, R. J. (1987). Parental alcohol-
Elsevier. ism and childhood psychopathology. Psychological
Waschbusch, D. A. (2002). A meta-analytic examination Bulletin, 102, 204–218.
of comorbid hyperactive-impulsive-attention prob- Whalen, D. K., Henker, B., Jamner, L. D., Ishikawa, S. S.,
lems and conduct problems. Psychological Bulletin, Floro, J. N., Swindle, R., et al. (2006). Toward map-
128, 118–150. ping daily challenges of living with ADHD: Maternal
Wechsler, D. (2014). Wechsler intelligence scale for chil- and child perspectives using electronic diaries. Journal
dren, 5th edition (WISC-5). New York, NY: Pearson. of Abnormal Child Psychology, 34, 115–130.
Weems, C. F., Silverman, W. K., Rapee, R. R., & Pina, Whipple, E. E., & Webster-Stratton, C. (1991). The role
A. A. (2003). The role of control in childhood anxi- of parental stress in physically abusive families. Child
ety disorders. Cognitive Therapy and Research, 27, Abuse & Neglect, 15, 277–291.
557–568. Whitcomb, S. A., & Merrell, K. W. (2013). Behavioral,
Weiner, I. B. (1986). Assessing children and adolescents social, and emotional assessment of children and ado-
with the Rorschach. In H. M. Knoff (Ed.), The adjust- lescents (4th ed.). New York, NY: Routledge/Taylor
ment of child and adolescent personality (pp. 141– & Francis.
172). New York, NY: Guilford Press. Whiteside-Mansell, L., Ayoub, C., McKelvey, L.,
Weiner, I. B. (2001). Advancing the science of psycho- Faldowski, R. A., Hart, A., & Shears, J. (2007).
logical assessment: The Rorschach Inkblot Method as Parenting stress of low-income parents of toddlers and
exemplar. Psychological Assessment, 13(4), 423–432. preschoolers: Psychometric properties of a short form
Weiss, B., & Garber, G. (2003). Developmental dif- of the Parenting Stress Index. Parenting: Science and
ferences in the phenomenology of depression. Practice, 7, 27–56.
Development and Psychopathology, 15, 403–430. Wickerd, G., & Hulac, D. (2017). Generalizability and
Weiss, B., Harris, V., & Catron, T. (2002). Development dependability of a multi-item direct behavior rating
and initial validation of the peer-report measure of scale in a kindergarten classroom setting. Journal of
internalizing and externalizing behavior. Journal of Applied School Psychology, 33, 109–123.
Abnormal Child Psychology, 30, 285–294. Willcutt, E. G., Nigg, J. T., Pennington, B. F., Solanto,
Weissman, M. M., Warner, V., & Fendrich, M. (1990). M. V., Rohde, L. A., Tannock, R. L., … Lahey,
Applying impairment criteria to children’s psychiatric B. B. (2012). Validity of DSM-IV attention-deficit/
diagnosis. Journal of the American Academy of Child hyperactivity disorder symptom dimensions and sub-
Adolescent Psychiatry, 29, 789–795. types. Journal of Abnormal Child Psychology, 121,
Weissman, M. M., Warner, V., Wickramaratne, P., & 991–1010.
Prusoff, B. A. (1988). Early-onset major depression Williams, C. L., Butcher, J. N., Ben-Porath, Y. S., &
in parents and their children. Journal of Affective Graham, J. R. (1992). MMPI-A Content Scales:
Disorders, 15, 269–277. Assessing psychopathology in adolescents.
Weissman, M. M., Wickramaratne, P., Adams, P., Wolk, Minneapolis, MN: University of Minnesota Press.
S., Verdeli, H., & Olfson, M. (2000). Brief screen- Wirt, R. D., Lachar, D., Klinedinst, J. K., & Seat, P. S.
ing for family psychiatric history: The family history (1990). Personality inventory fir children-1990 edition.
screen. Archives of General Psychiatry, 57, 675–682. Los Angeles, CA: Western Psychological Services.
Weisz, J. R., & Hawley, K. M. (1998). Finding, evalu- Wolf, T. H. (1966). Intuition and experiment: Alfred
ating, refining, and applying empirically supported Binet’s first efforts in child psychology. Journal of the
treatments for children and adolescents. Journal of History of the Behavioral Sciences, 2(3), 233–239.
Clinical Child Psychology, 27, 206–216. Woodworth, R. S. (1918). Personal data sheet. Chicago,
Weisz, J. R., Sandler, I. N., Durlak, J. A., & Anton, B. S. IL: Stoelting.
(2005). Promoting and protecting youth mental health World Health Organization. (2018). The ICD-11 clas-
through evidence-based prevention and treatment. sification of mental and behavioural disorders:
American Psychologist, 60, 628–648. Diagnostic criteria for research. Geneva, Switzerland:
Weller, E. B., Weller, R. A., Fristad, M. A., Rooney, M. T., World Health Organization.
& Schecter, J. (2000). Children’s interview for psychi- Yap, M. B. H., & Jorm, A. F. (2015). Parental factors
atric syndromes (ChIPS). Journal of the American associated with childhood anxiety, depression, and
Academy of Child and Adolescent Psychiatry, 39, internalizing problems: A systematic review and
76–84. meta-analysis. Journal of Affective Disorders, 175,
Wells, M. G., Burlingame, G. M., & Rose, P. M. 424–440.
(2003). Administration and scoring manual for the Young, M. E., Bell, Z. E., & Fristad, M. A. (2016).
Y-OQ-SR-2.0 (Youth Outcome Questionnaire-Self Validation of a brief structured interview: The
Report). Wilmington, DE: American Professional Children’s Interview for Psychiatric Syndromes
Credentialing Services. (ChIPS). Journal of Clinical Psychology in Medical
Wender, P. H. (1995). Attention deficit disorder in adults. Settings, 23, 327–340.
London: Oxford University. Youngstrom, E., Zhao, J., Mankoski, R., Forbes, R. A.,
Marcus, R. M., Carson, W., … Findling, R. L. (2013).
References 425
Clinical significance of treatment effects with aripip- ing temporally stable peer status groups of girls. The
razole versus placebo in a study of manic or mixed Journal of Early Adolescence, 27(1), 90–114.
episodes associated with pediatric bipolar I disorder. Zettergren, P., Bergman, L. R., & Wangby, M. (2006).
Journal of Child and Adolescent Psychopharmacology, Girls’ stable peer status and their adult adjustment: A
23, 72–79. longitudinal study from age 10 to age 43. International
Youngstrom, E. A. (2013). Future directions in psy- Journal of Behavioral Development, 30, 315–325.
chological assessment: Combining evidence-based Zlomke, K., Bauman, S., & Lamport, D. (2015).
medicine innovations with psychology’s historical Adolescents’ perceptions of parenting behavior:
strengths to enhance unity. Journal of Clinical Child Validation of the Alabama Parenting Questionnaire
and Adolescent Psychology, 42(1), 139–159. adolescent self-report. Journal of Child and Family
Zander, E., Willfors, C., Berggren, S., Coco, C., Holm, Studies, 24, 3159–3169.
A., Jifält, I., et al. (2017). The interrater reliability of Zlomke, K. R., Lamport, D., Bauman, S., Garland, B., &
the Autism Diagnostic Interview-Revised (ADI-R) in Talbot, B. (2014). Parenting adolescents: Examining
clinical settings. Psychopathology, 50, 219–227. the factor structure of the Alabama Parenting
Zettergren, P. (2016). Cluster analysis in sociometric Questionnaire for adolescents. Journal of Child and
research: A pattern-oriented approach to identify- Family Studies, 23, 1484–1490.
Index