See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/12768072

Teaching Study Skills and Test-Taking Strategies to Elementary School Students

The Testbusters Program

Article  in  Behavior Modification · November 1999

DOI: 10.1177/0145445599234007 · Source: PubMed

CITATIONS READS

45 2,870

3 authors, including:

Deborah C Beidel
University of Central Florida
287 PUBLICATIONS   15,864 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Evaluation of Social Anxiety Disorder View project

All content following this page was uploaded by Deborah C Beidel on 03 February 2014.

The user has requested enhancement of the downloaded file.

BEHAVIOR
Beidel et al. / MODIFICATION
THE TESTBUSTERS
/ October
PROGRAM
1999

A pilot program to reduce test anxiety and related social-evaluative concerns is presented. Test-
busters is a program designed specifically for elementary and middle school children in grades 4
through 7 that teaches effective study habits, study skills, and test-taking strategies and includes
a behavioral contract to ensure consistent study behavior. The assessment strategy includes self-
report instruments, a behavioral assessment, and children’s grade point averages. After a
6-month waiting period, eight children with moderate to severe test anxiety participated in the
11-week program. The results indicated that Testbusters decreased general levels of test anxiety
and self-ratings of distress when taking a test. Overall grade point average improved signifi-
cantly and grades in the majority of the subjects showed positive improvement. There was no
change in overall self-esteem or judgments of cognitive competence. The results are discussed in
terms of the use of a skills strategy to decrease anxiety and improve academic achievement.

Teaching Study Skills and

Test-Taking Strategies to
Elementary School Students
The Testbusters Program

DEBORAH C. BEIDEL
SAMUEL M. TURNER
University of Maryland

JILL C. TAYLOR-FERREIRA
Pressley Ridge School

Test anxiety, an extreme fear of performing poorly on examinations,

is a common form of anxiety among elementary school children, par-
ticularly third through sixth grades (Barrios, Hartmann, & Shigatomi
1981; Sarason, Davidson, Lighthall, Waite, and Ruebush, 1960). In 1960,
Sarason, Davidson, Lighthall, Waite, & Ruebush (1960) reported that
20% of elementary school children suffered from test anxiety. More
recently, test anxiety has been reported to affect between 34% to 41% of
third through sixth grade children (Beidel, 1991; Turner, Beidel,

BEHAVIOR MODIFICATION, Vol. 23 No. 4, October 1999 630-646

© 1999 Sage Publications, Inc.

630
 Beidel et al. / THE TESTBUSTERS PROGRAM 631

Hughes, & Turner, 1993). The reasons for this apparent increase in
prevalence during the past 30 years are unclear but may be related to
parental or teacher pressure to achieve, increased expectations regard-
ing the complexity of work to be mastered at earlier developmental
ages, currently used testing strategies, or the great emphasis placed on
children to achieve above the national average when standardized
testing occurs (Turner et al., 1993). Although its origin is unclear, it is
clear that test anxiety affects a substantial proportion of school age
children.
Test anxiety affects children of all academic achievement and intel-
lectual levels (Sarason et al., 1960) and is roughly equally prevalent in
African American and Caucasian children (Beidel, Turner, & Trager,
1994). Test anxiety has negative effects on academic achievement
(Sarason et al., 1960). In addition, test anxiety has been related to aca-
demic failure, low self-esteem, dependency, and passivity (Campbell,
1986). In its most extreme form, test anxiety can be a contributing fac-
tor to school refusal (Ollendick & Meyer, 1984). Furthermore, many
children with test anxiety, in comparison to nontest-anxious controls,
actually suffer from a broader pattern of psychological distress includ-
ing higher general anxiety, more nontesting fears, and decreased esti-
mates of their cognitive abilities. Finally, two studies (Beidel &
Turner, 1988; Beidel, 1991) have shown that 50% to 63% of children
with test anxiety actually met criteria for an anxiety disorder accord-
ing to the Diagnostic and Statistical Manual of Mental Disorders
(DSM) (American Psychiatric Association, 1980, 1987). In the major-
ity of cases, these children met diagnostic criteria for social phobia or
overanxious disorder, two anxiety disorders that are characterized by
irrational and unreasonable fears of doing or saying something that
will be negatively evaluated by others. Thus, similar to test anxiety,
these disorders involve fear of negative evaluation of performance (in
this case, social interactions rather than academic knowledge) by oth-
ers. In summary, although most children may experience mild test
anxiety at some time in their lives, there is a substantial subgroup for
whom test anxiety is a chronic and severe condition and for whom its
detrimental effects interfere with academic achievement and perhaps
social development.
632 BEHAVIOR MODIFICATION / October 1999

Improving children’s academic achievement has been a primary

goal of educational research, and behavioral interventions often have
been used in this regard. For example, goal setting and contingency
contracting have been found to be effective in improving 9- to 11-
year-old children’s homework performance (Miller & Kelley, 1994).
Adams, Carmine, and Gersten (1982) reported that 4 days of training
in the Survey, Question, Read, Review, and Recite (SQ3R) study
method improved short-term recall when fifth graders took a short
answer test. SQ3R also was effective in improving oral recall in three
learning disabled students (Alexander, 1985). Although such inter-
ventions appear effective, none of these studies addressed the use of
these procedures with test-anxious elementary school students.
To date, strategies designed to treat test-anxious children have used
one of two approaches: (a) efforts to decrease anxiety or (b) efforts to
increase test-wiseness. Briefly, with respect to decreasing anxiety,
desensitization procedures have been somewhat effective in reducing
anxiety and improving test performance in elementary school stu-
dents with test anxiety (Barabasz, 1973; Kondas, 1967). Similarly,
Wilson and Rotter (1986) reported that anxiety management training
and study skills counseling may be effective. However, none of these
studies included academic achievement as an outcome variable,
despite its obvious importance. Furthermore, anxiety-reduction pro-
cedures require substantial clinical expertise and are not easily admin-
istered by most school personnel.
Test-wiseness programs encompass strategies designed to increase
children’s sophistication with testing procedures and have been pro-
moted as an effective tool for increasing test performance and school
achievement. Based on a meta-analysis of 24 studies, test-wiseness
programs generally are effective in enhancing achievement scores
(Samson, 1985). However, there are several caveats that must be kept
in mind. First, programs of 5 weeks or longer produce significantly
greater academic achievement gains than shorter programs. Second,
average or above average achieving students understood and were
more likely to benefit from the program than below average students
(Ritter & Idol-Maestas, 1986). Third, and perhaps most important, the
results of test-wiseness strategies (and anxiety-reduction strategies)
are enhanced for students who already possess good study habits (e.g.,
 Beidel et al. / THE TESTBUSTERS PROGRAM 633

Cavallaro & Meyers, 1986). Thus, teaching anxiety-reduction strate-

gies will be less effective if children do not possess the study habits
and study skills that are a necessary prerequisite for success.
Skill and anxiety often have an orthogonal relationship (Hersen &
Bellack, 1976). That is, skill and anxiety can independently vary
along a high-low dimension resulting in the formation of four groups
(high skill-low anxiety, low skill-low anxiety, high skill-high anxiety,
or low skill-high anxiety). In the case of the high skill-high anxiety
group, poor performance is theorized to result from high anxiety that
inhibits skill (i.e., the athlete chokes during the big game). In such cases,
anxiety reduction strategies may reduce arousal such that skilled behav-
ior can be produced. In the case of low skill-high anxiety, however, poor
performance results from a lack of skill, which in turn increases arousal.
Thus, attempting to decrease arousal will be ineffective if skill build-
ing is not addressed (because arousal is a direct result of poor skill).
There is little reason to believe that children will develop good study
habits spontaneously and, to our knowledge, programs designed to
teach study skills at the elementary school level currently do not exist.
Testbusters (Beidel & Taylor-Ferreira, 1996) is an 11-week study
skills and test-taking program designed specifically for elementary
and middle school students in 4th through 7th grades. The program
includes behavioral contracting, study skills training, and test-
wiseness training. The aim of the program is to teach specific skills,
thereby decreasing anxiety, improving academic achievement, and
improving academic self-concept. The purpose of this study was to
determine if the Testbusters program decreased anxiety and improved
academic performance in test-anxious children.

METHOD

PARTICIPANTS

The participants were eight children (4 boys and 4 girls) between

the ages of 9 and 12 (M age = 10.9, SD = 1.3) in grades 4 through 6 who
were significantly test anxious as measured by scores on the Test
Anxiety Scale for Children (TASC) (Sarason et al., 1960). The Test-
634 BEHAVIOR MODIFICATION / October 1999

busters program was advertised to all parents of children in grades 4

through 6 who had participated in an assessment study of reliability
of anxiety measures (see below). The participants initially were iden-
tified by their parents as (a) commonly expressing anxiety about a
range of testing situations and (b) their performance in school was
impaired by their test anxiety. Determination of test-anxious status
according to the TASC was made using cutoffs validated by Beidel
and Turner (1988). To be considered test anxious, girls had to score 16
or higher on the TASC, whereas boys had to score 10 or higher. The
difference reflects the tendency of boys to minimize their fears. The
overall mean TASC score for the 8 subjects was 18.00 (SD = 4.9; see
Table 1 for a breakdown by gender), indicating the sample was moder-
ately to severely test anxious.
In addition to their selection using the TASC, children were inter-
viewed with the Anxiety Disorders Interview Schedule for Children
(Silverman & Nelles, 1988) to confirm their test-anxious status. This
semistructured interview contains a set of questions assessing anxiety
in various school situations, including testing situations. During the
diagnostic interview, all children endorsed at least moderate distress
in testing situations. In addition, children reported anxiety answering
questions orally and writing on the board. Twenty-five percent of all
interviews were rated by a second clinician unaware of the first’s diag-
nosis. Although a specific interrater reliability coefficient is not avail-
able for these 8 subjects due to sample size, the Kappa coefficient for
the larger sample was k = .87.
All children were enrolled in regular classroom settings in public
elementary schools in southwestern Pennsylvania. All were Cauca-
sian and were from upper-middle-class socioeconomic status back-
grounds as assessed by the Hollingshead (1975) scale. None was tak-
ing psychotropic medication for anxiety. Among the children who
qualified for the study (n = 25), these 8 children represent those whose
parents were interested in having them participate and who could
attend during the scheduled group sessions. Analyses of demographic
and academic achievement data indicated no differences between
these children and those who did not participate in the program. The
children previously had been participants in a larger study assessing
the short- and long-term reliability of various measures of test anxiety
 Beidel et al. / THE TESTBUSTERS PROGRAM 635

TABLE 1
Means and Standard Deviations for Initial,
6-Month Reliability, and Posttreatment Assessments
Variable Initial 6-Month Posttreatment

Test Anxiety Scale for Children 18.0 (4.9) 15.1 (4.1) 6.4 (4.9)*
Boys 12.7 (0.6) 13.3 (3.1) 3.7 (3.2)
Girls 20.8 (3.1) 14.5 (3.1) 8.0 (6.1)
Perceived Self-Competence Scale
for Children
Cognitive 2.5 (0.2) 2.6 (0.2)
Social 2.3 (0.3) 2.3 (0.2)
Physical 2.6 (0.4) 2.7 (0.2)
General self-esteem 2.4 (0.4) 2.4 (0.1)
Pulse rates during:
Vocabulary test 82.9 (10.1) 82.6 (6.9) 86.2 (7.4)
Read aloud 95.7 (6.7) 85.8 (8.2) 88.6 (9.3)
Self-Assessment Manikin ratings during:
Vocabulary test 1.9 (0.4) 1.8 (0.4) 1.1 (0.6)*
Reading task 2.4 (0.9) 2.1 (0.4) 1.4 (0.5)
Grades
English 2.5 (0.8) 2.7 (0.5) 3.0 (0.5)
Mathematics 3.4 (0.7) 3.4 (0.5) 3.1 (0.4)
Reading 2.4 (0.9) 2.3 (0.7) 2.9 (1.0)
Science 2.5 (0.9) 2.0 (0.9) 3.1 (1.1)
Social studies 2.6 (0.9) 2.5 (1.1) 3.4 (0.7)
Overall grade point average 2.7 (0.8) 2.6 (0.6) 3.1 (0.6)*

*Initial to posttreatment comparisons significantly different at p < .01 (Bonferroni correction).

(self-report and psychophysiological assessment). Thus, this 6-month

period provides an opportunity to address the stability of these
measures.

MEASURES

Self-report inventories included the TASC, a 30-item true-false

inventory assessing fears surrounding tests and other evaluative situa-
tions (such as reading in front of the class or writing on the board), and
the Perceived Self-Competence Scale for Children (PSCSC) (Harter,
1982), a 28-item inventory that assesses children’s self-concept using
a 4-point rating scale in four areas: social competence, cognitive com-
petence, physical competence, and general self-esteem.
636 BEHAVIOR MODIFICATION / October 1999

The assessment strategy included a behavioral assessment

designed to determine children’s distress when actually engaged in
social evaluative tasks. The assessment required the children to com-
plete an age-appropriate vocabulary test and to read aloud in front of
an audience. The latter task was included because approximately 50%
to 60% of children with test anxiety also have fears in public perfor-
mance situations such as reading aloud (Beidel & Turner, 1988).
Thus, the task allowed for assessment of generalization effects
(whether the treatment program would affect a second related area of
functioning). The age-appropriate vocabulary test was the California
Achievement Test. For the reading task, children were required to read
aloud the story of Jack and the Beanstalk to an audience of three young
adults. Each task was 10 minutes in length and both tasks were preceded
by a 10-minute baseline period. During baseline, children rested quietly
in a sitting position. Order of the tasks was randomly determined.
During baseline and the two behavioral tasks, pulse rates were
assessed every 2 minutes using an IBS Corporation SD700A auto-
mated blood pressure/pulse rate monitor. Pulse rate was assessed at
the start of the baseline or task and after 2, 4, 6, 8, and 10 minutes,
respectively. After each task, the child was asked to rate perceived
level of distress using the Self-Assessment Manikin (SAM) (Lang &
Cuthbert, 1984) as modified and validated by Beidel and S. M. Turner
(1988). The modified version consists of five unisex pictures depict-
ing various levels of anxious distress, where 1 = no anxiety and 5 =
maximum anxiety.
Children’s academic achievement was assessed using five subjects
common to the various grade levels: English, reading, social studies,
mathematics, and science. Letter grades were converted to a numeri-
cal scale where A = 4, B = 3, C = 2, D = 1, and F = 0.

SIX-MONTH RELIABILITY ASSESSMENT

Prior to treatment, all children participated in a 6-month repeat

assessment designed to determine the reliability of pulse rate and
self-report ratings during the behavioral tasks. In addition, the chil-
dren completed a second TASC. Grades also were available during
this 6-month time period. Thus, this 6-month reliability phase served
 Beidel et al. / THE TESTBUSTERS PROGRAM 637

as a type of control condition (i.e., subjects served as their own con-

trol). If change did not occur during this period of time but did occur
after the intervention, there is some increased confidence that the
treatment outcome was the result of the intervention and not merely
the result of time or placebo effects. It must be stated clearly, however,
that a design that included a control group would offer a superior
degree of control. Children were not aware that they would be offered
the opportunity to participate in a treatment program until after the
6-month stability period.

TREATMENT

The Testbusters treatment program is an 11-week study skills and

test-taking strategy program designed specifically for children in
grades 4 through 7. It emphasizes the acquisition of skill and mini-
mizes attention to academic achievement. The philosophy underlying
the program is that improving skill will decrease anxiety and improve
self-esteem thereby allowing each child to fulfill his or her academic
potential.
Testbusters can be implemented individually or in small groups. In
this study, a group format was used. The program includes the acquisi-
tion of several specific skills. First, to establish good and consistent
study habits, children’s current study behaviors were assessed. Poor
study habits (such as studying in front of the TV or in a high traffic
area) were replaced with proper study hygiene (studying in a quiet
room at a desk with minimal distractions). Second, children were
required to study an additional 15 to 20 minutes per night after com-
pleting their nightly homework, using an academic subject in which
they were having difficulty or that created distress. Thus, this repre-
sented an additional study period over and above regular nightly
homework. The children were expected to complete this additional
studying for the entire 11-week program. To ensure compliance, a
contract between child and parent was established, where children
received a small reward each night that they completed their study
assignment, and a somewhat larger reward if they studied every week-
night during the entire week. Small rewards included such things as a
delay in bedtime equal to the extra time spent studying (e.g., 15 min-
638 BEHAVIOR MODIFICATION / October 1999

utes of studying meant staying up 15 minutes later), a special food as a

bedtime snack, or extra telephone time. Larger (weekly) rewards
included renting an extra video on the weekend, special weekend
breakfasts, time alone with Mom or Dad, and a host of others.
Rewards were negotiated individually with parent and child and repre-
sented incentives that uniquely motivated each child.
The initial assessment of children’s study behaviors indicated that
all children were committing common study errors such as studying
with the television or music on, studying in a high traffic area such as
the kitchen, or spending as much time on study breaks as on studying.
Thus, study contracts included improvement in study habits as well as
consistent study behavior. Compliance was monitored by the child’s
parent. Children were questioned at each group meeting regarding
compliance with the study contract and all children reported receiv-
ing their rewards consistently (i.e., they were studying correctly and
consistently).
Once good and consistent study habits were established, children
were instructed in the SQ3R (Carmen & Adams, 1972) method of
studying. The first two procedures, survey and question, provide a
way of organizing material to make study time more effective. Survey
means to look over the entire assignment to get a general idea of its
content. Question means to use text information such as bold-faced,
italicized, and underlined words; headings; subheadings; maps; and
figures as clues for identifying the critical information contained in
the text. Children are taught to turn these clues into questions (e.g., the
heading “The Pilgrims Land at Plymouth Rock” becomes “When did
the Pilgrims land at Plymouth Rock?”). Read means that then reading
the material becomes reading with a purpose (i.e., to find the answer to
the questions). Review means that at the end of the reading assign-
ment, children go back to their list of questions to see if they can
answer them. Recite means that they have someone else quiz them on
the material they learned that evening. Instruction in SQ3R took place
over 3 weeks (weeks 3 through 5 of the program) and included home-
work assignments incorporated into the child’s extra study time.
Although SQ3R could be considered merely a specific study tactic,
this procedure also teaches organizational and decision-making skills
such as understanding the nature of the study assignment. In addition,
 Beidel et al. / THE TESTBUSTERS PROGRAM 639

because this method is useful only for certain school subjects that
require reading text passages (in comparison to mathematics), chil-
dren learn to identify the need for different approaches to studying dif-
ferent types of information. Finally, the review and recitation aspects
of SQ3R helps children determine whether the material has been
learned or whether further attention is needed. Thus, although SQ3R
may appear to be an old-fashioned procedure, it nonetheless includes
attention to the more contemporary theories of studying.
Following the acquisition of SQ3R, children were instructed in
behaviors designed to ensure physical and emotional readiness prior
to taking a test. Thus, they were instructed about getting to bed on time
the night before a test, wearing comfortable clothes to school the next
day, eating a good breakfast and lunch the day of the test, relaxing
(rather than cramming) a few minutes before a test, listening to all
directions provided by the teacher, and so on. The final 4 weeks of the
program focused on improving specific test-taking skills. Children
were instructed in how to use, to their advantage, all of the test time
allotted. In addition, they were taught to carefully read all instructions,
eliminate incorrect options from multiple choice tests to increase the
chance of identifying the correct answer, and to identify the use of
absolute words such as never or always that provide clues about the
veracity of true-false statements. The final lesson addressed how to
review tests handed back by the teacher to identify testing mistakes to
correct performance on the next exam. A detailed description of the
Testbusters program is contained in Beidel and Taylor-Ferreira (1996).

RESULTS

SIX-MONTH RELIABILITY ASSESSMENT

Comparisons between the initial assessment and the 6-month reli-

ability assessment were conducted using pairwise-comparison t tests.
As noted, TASC scores, pulse rates and SAM ratings during the
vocabulary test and the read aloud task, and grades were available for
the reliability assessment. Because of the ratio of t tests to participants,
the Bonferroni correction factor was used to control the experiment-
640 BEHAVIOR MODIFICATION / October 1999

wise error rate. The target significance level was p = .01. All data are
presented in Table 1. Pulse rate data were reduced to a mean score by
averaging readings across the 10-minute interval. A repeated mea-
sures ANOVA across the six readings within each task indicated that
the first reading was significantly higher than all others (perhaps rep-
resenting a startle response). Thus, the mean pulse rate represented the
average of the 2-, 4-, 6-, 8-, and 10-minute readings. There was a sig-
nificant decrease in pulse rate during the reading task at the 6-month
follow-up period, t(df = 7) = 4.15, p < .005. There were no differences
on pulse rate during the vocabulary task, SAM ratings during either
task, TASC scores, or grades during the 6-month reliability phase. The
outcome for these 8 children is consistent with the larger cohort who
completed the 6-month reliability assessment. In that study, 6-month
reliability coefficients (for a sample of 150 school children) were
moderate to high and significant for self-report measures but low and
nonsignificant for pulse rate (Beidel, Turner, & Fink, 1996).

TREATMENT OUTCOME

Again, an adjusted target significance level of .01 (Bonferroni cor-

rection) was used. All means and standard deviations are presented in
Table 1. Pulse rates were calculated as described earlier. Pre- and post-
treatment TASC scores were compared using a paired comparison t
test. The results indicated that TASC scores decreased significantly as
a result of the Testbusters program, t(df = 7) = 5.24, p < .001. There
were, however, no changes on either the cognitive, t(df = 7) = 1.1, p >
.05; social, t(df = 7) = 0.2, p > .05; physical, t(df = 7) = 1.21, p >.05; or
general self-esteem, t(df = 7) = 0.01, p > .05, subscales of the PSCSC.
Paired comparison t tests also indicated there was no decrease in
pulse rate during either the vocabulary, t(df = 1.88, p >.05, or read
aloud task, t(df = 7) = 0.09, p > .05, when decreases as a result of the
6-month reliability assessment were controlled. Self-ratings of distress
decreased significantly for the vocabulary task, t(df = 7) = 3.0, p < .01,
and the read aloud task showed a trend in the appropriate direction,
t(df = 7) = 2.37, p < .05.
Using paired comparison t tests, there was significant improvement
in overall grade point average, t(df = 7) = 3.9, p < .01. In addition, there
 Beidel et al. / THE TESTBUSTERS PROGRAM 641

were strong trends for similar improvement on individual academic

subjects including English, t(df = 7) = 2.65, p < .05; reading, t(df = 7) =
2.65, p < .05; science, t(df = 7) = 2.38, p < .05; and social studies, t(df =
7) = 3.0, p < .025. There was, however, no change in mathematics
grades, t(df = 7) = 1.0, p > .05).

DISCUSSION

The results of this study indicate that the Testbusters program may
be an effective program for decreasing test anxiety. Although the pro-
gram did not include any standardized anxiety-reduction strategies or
coping procedures, scores on the Test Anxiety Scale for Children
decreased significantly after the intervention. The mean score at post-
treatment for this sample is lower than the scores of unselected sam-
ples of elementary school children in the same area of the country (Ms
ranging from 11.4 to 12.6) (Turner et al., 1993) and fall within the
range of those usually considered as not test anxious. Thus, the
decrease in test anxiety in this study implies not just a statistically sig-
nificant decrease but a change that is of meaningful clinical
significance.
Academic achievement also showed significant improvement after
the intervention. Overall grade point averages across the five subjects
improved significantly from a C to a B. In addition, although achieve-
ment was de-emphasized in this program in favor of skill acquisition,
pre- to postcomparison of grade changes showed strong trends for
improvement in every individual subject with the exception of mathe-
matics. With respect to the latter, an important component of the Test-
busters program is the study contract and the SQ3R study skills pro-
gram. SQ3R requires the use of written study materials that allow the
child to form questions, read for the answers, and then review the
material. Mathematics books do not easily lend themselves to the
application of the SQ3R method and children were discouraged from
selecting mathematics when applying the Testbusters program. Thus,
math grades would not be expected to improve. However, given that
many children identify mathematics and math tests as substantial
sources of concern, study skills and test-taking strategies appropriate
642 BEHAVIOR MODIFICATION / October 1999

for this subject should be developed because Testbusters is not

designed to address those specific concerns.
The behavioral assessment data revealed that after treatment, chil-
dren with test anxiety reported decreased anxiety when taking an age-
appropriate vocabulary test. This decrease was significant although
the children did not report substantial anxiety when performing these
tasks at pretreatment. One reason for the somewhat low distress rat-
ings (for both read aloud and test taking tasks) at pretreatment may be
the analogue nature of the task. Thus, although the vocabulary test was
age appropriate and the children were told that their score would be
compared with those of all others participating in the study, they knew
that the test was not related to their school grades. Thus, the test situa-
tion was less relevant than that found in a traditional classroom setting.
Support for this hypothesis is bolstered by the pulse rates during the
vocabulary task, which are within the range of values normally con-
sidered to be an average resting heart rate. This floor effect at pretreat-
ment probably prevented any decrease as a result of treatment. Recent
advances in technology now make possible the ability to unobtru-
sively collect heart rate data in a naturalistic setting. Future investiga-
tions should use this technology to determine more accurately physio-
logical reactivity and generalization to naturalistic settings. In
summary, although the children did not endorse the same degree of
distress when taking this age-appropriate vocabulary test that they did
when taking an actual test in their classroom, distress ratings still
decreased significantly after treatment.
Finally, the data did not reveal any improvement in the children’s
perceptions of their cognitive, social, and physical competence, or
their general self-esteem. It is likely that 11 weeks is too short a period
of time for broad changes in self-perception to occur. Furthermore, it
is likely that similar to other behavioral change processes, changes in
external behavior precede changes in cognitive processes such as per-
ceptions of competence or self-esteem. In this case, feedback from the
external environment, in the form of improved grades, may be neces-
sary before the children can internalize these changes and attribute
them to improvement in their own abilities. Follow-up studies of con-
siderable duration are needed to determine if change in self-concept
occurs at some later date.
 Beidel et al. / THE TESTBUSTERS PROGRAM 643

This study has several limitations that must be kept in mind when
evaluating the results. First and most importantly, there was no control
group. Thus, it cannot be stated with absolute certainty that the
changes occurred as a result of the treatment program. Although the
finding that only minimal changes occurred during the 6-month reli-
ability assessment phase provides some limited control for validity
threats such as time, such a design cannot take the place of a controlled
trial. In addition to the positive effects of treatment, the reported
changes could have been due to reactive effects of attention, repeated
measurement, or cognitive and physical maturation. For example,
behaviors such as smoking cigarettes are known to decrease in fre-
quency with clinical attention such as self-monitoring. Similarly,
repeated assessment of the same variable may result in a change in the
dependent variable (i.e., a practice effect). The latter explanation may
be more plausible for self-report and behavioral assessment data but
less plausible in explaining improved academic achievement. Finally,
the children were 9 months older at the end of the study than when the
initial assessment occurred. Thus, this increased maturity could have
affected their scores. Although there are no data to suggest test anxiety
decreases with age, it is possible that the children may have felt more
comfortable in the classroom setting and with their teacher and their
peers. Again, this may have decreased general arousal but it is less
likely to affect academic achievement.
Because there are a number of possible explanations for the posi-
tive changes described in this study other than the Testbusters pro-
gram, this intervention appears promising, but a controlled trial is
needed prior to making more conclusive statements. Furthermore, the
sample size was small and consisted of children from an upper-
middle-class background, all of whom were Caucasian. Future inves-
tigations should address the applicability of the program with children
from a broader range of socioeconomic status, ethnicity, and social
settings. Also, these children had average grades at the start of the pro-
gram and improved to above average as a result of the program. This
suggests that the acquisition of these skills may be useful for children
struggling to achieve academically. However, its applicability to chil-
dren who are low academic achievers or who have learning disabilities
remains to be addressed. Another consideration is that approximately
644 BEHAVIOR MODIFICATION / October 1999

50% to 60% of children with test anxiety also have other anxiety disor-
ders such as social phobia or overanxious disorder. It is unlikely that a
problem-focused program such as Testbusters would be appropriate
for the remediation of these comorbid fears. Furthermore, it is unclear
if children with comorbid disorders improve their study skills given
the nature of their comorbid conditions. Further studies are necessary
to address these issues.
In summary, Testbusters is a promising intervention for the treat-
ment of test anxiety in elementary school students. Because of its mul-
ticomponent nature (i.e., reinforcement for studying, study skills, and
test-taking skills training), the current study does not allow one to
determine the relative efficacy of these various components. Such dis-
mantling studies usually follow the determination that an overall
package is effective. The program’s effects appear to include improv-
ing academic achievement as well as decreasing anxiety. Intervention
at the elementary school level may have preventive as well as thera-
peutic effects, preventing the correlates of test anxiety (poor self-
concept and attenuated academic achievement) from further impact-
ing the social, emotional, and academic development of young
children.

REFERENCES

Adams, A., Carmine, D., & Gersten, R. (1982). Instructional strategies for studying content area
texts in the intermediate grades. Reading Research Quarterly, 18, 27-55.
Alexander, D. F. (1985). The effect of study skill training on learning disabled students’retelling
of expository material. Journal of Applied Behavior Analysis, 18, 263-267.
American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disor-
ders (3rd ed.). Washington, DC: Author.
American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disor-
ders (3rd ed., rev.). Washington, DC: Author.
Barabasz, A. F. (1973). Group desensitization of test anxiety in elementary school. Journal of
Psychology, 83, 295-301.
Barrios, B. A., Hartman, D. B., & Shigatomi, C. (1981). Fears and anxieties in children. In E. J.
Mash & L. G. Terdal (Eds.), Behavioral assessment of childhood disorders (pp. 259-304).
New York: Guilford.
Beidel, D. C. (1991). Social phobia and overanxious disorder in school-age children. Journal of
the American Academy of Child and Adolescent Psychiatry, 30, 545-552.
Beidel, D. C., & Taylor-Ferreira, J. C. (1996). Testbusters: Study smarter not harder. Mt. Pleas-
ant, SC: Turndel.
 Beidel et al. / THE TESTBUSTERS PROGRAM 645

Beidel, D. C., & Turner, S. M. (1988). Comorbidity of test anxiety and other anxiety disorders in
children. Journal of Abnormal Child Psychology, 16, 275-287.
Beidel, D. C., Turner, S. M., & Fink, C. M. (1996). The assessment of childhood social phobia:
Construct, convergent and discriminative validity of the Social Phobia and Anxiety Inven-
tory for Children (SPAI-C). Psychological Assessment, 3, 235-240.
Beidel, D. C., Turner, S. M., & Trager, K. N. (1994). Test anxiety and childhood anxiety disor-
ders in African American and White school children. Journal of Anxiety Disorders, 8,
169-179.
Campbell, S. B. (1986). Developmental issues. In R. Gittelman (Ed.), Anxiety disorders in chil-
dren (pp. 24-57). New York: Guilford.
Carmen, R. A., & Adams, R.W.J. (1972). Study skills: A student’s guide for survival. New York:
John Wiley.
Cavallaro, D. M., & Meyers, J. (1986). Effects of study habits on cognitive restructuring and
study skills training in the treatment of test anxiety with adolescent females. Techniques: A
Journal for Remedial Education and Counseling, 2, 145-155.
Harter, S. (1982). The perceived competence scale for children. Child Development, 53, 87-97.
Hersen, M., & Bellack, A. S. (1976). Social skills training for chronic psychiatric patients:
Rationale, research findings, and future directions. Comprehensive Psychiatry, 17, 559-580.
Hollingshead, A. (1975). Four factor index of social status. Unpublished manuscript, Yale Uni-
versity, New Haven, CT.
Kondas, O. (1967). Reduction of examination anxiety and stage fright by group desensitization
and relaxation. Behaviour Research and Therapy, 5, 275-281.
Lang, P. J., & Cuthbert, B. N. (1984). Affective information processing and the assessment of
anxiety. Journal of Behavioral Assessment, 6, 386-395.
Miller, D. L., & Kelley, M. L. (1994). The use of goal setting and contingency contracting for
improving children’s homework performance. Journal of Applied Behavior Analysis, 27,
73-84.
Ollendick, T. H., & Meyer, J. (1984). School phobia. In S. M. Turner (Ed.), Behavioral theories
and treatment of anxiety (pp. 367-411). New York: Plenum.
Ritter, S., & Idol-Maestas, L. (1986). Teaching middle school students to use a test-taking strat-
egy. Journal of Educational Research, 79, 350-357.
Samson, G. E. (1985). Effects of training in test-taking skills on achievement test performance:
A quantitative synthesis. Journal of Educational Research, 78, 261-266.
Sarason, S. B., Davidson, K. S., Lighthall, F. F., Waite, R. R., & Ruebush, B. K. (1960). Anxiety
in elementary school children. New York: John Wiley.
Silverman, W. K., & Nelles, W. B. (1988). The Anxiety Disorders Interview Schedule for Chil-
dren. Journal of the American Academy of Child and Adolescent Psychiatry, 27, 772-778.
Turner, B. G., Beidel, D. C., Hughes, S., & Turner, M. W. (1993). Test anxiety in African Ameri-
can school children. School Psychology Quarterly, 8, 140-152.
Wilson, N. H., & Rotter, J. C. (1986). Anxiety management training and study skills counseling
for students on self-esteem and test anxiety and performance. School Counselor, 34, 18-31.

Deborah C. Beidel, Ph.D., is professor of psychology and codirector, Maryland Center

for Anxiety Disorders at the University of Maryland. Dr. Beidel is the author or coauthor
of more than 100 journal articles, book chapters, or books, including Shy Children, Pho-
bic Adults: Nature and Treatment of Social Phobia, coauthored with Samuel M. Turner,
Ph.D. She has served on the editorial boards of numerous psychological journals includ-
 646 BEHAVIOR MODIFICATION / October 1999

ing Journal of Consulting and Clinical Psychology, Professional Psychology: Research

and Practice, and Journal of Anxiety Disorders. She currently serves as chair of the Anxi-
ety Disorders Association of America’s Children Task Force and as chair of the National
Institute of Mental Health’s Child and Adolescent Psychosocial Interventions Research
Consortium. Her areas of research include the psychopathology and treatment of anxiety
disorders in adults and children. She is a diplomate of the American Board of Profes-
sional Psychology.

Samuel M. Turner, Ph.D., is professor of psychology and codirector, Maryland Center

for Anxiety Disorders at the University of Maryland. Dr. Turner is the author and editor
of more than 150 journal articles, book chapters, and books, including Shy Children,
Phobic Adults: Nature and Treatment of Social Phobia, coauthored with Deborah C. Bei-
del, Ph.D. He is associate editor of American Psychologist and has served on the edito-
rial boards of numerous psychological journals including Journal of Consulting and
Clinical Psychology, Behavior Therapy, and Behavior Research and Therapy. He has
served on the National Institute of Mental Health’s Scientific Advisory Committee as
well as numerous review panels. His areas of research include the psychopathology and
treatment of anxiety disorders in adults and children. He is a diplomate of the American
Board of Professional Psychology.

Jill C. Taylor-Ferreira, Ph.D., is Director of Research and Evaluation at the Pressley

Ridge Schools in Pittsburgh. She is the author of a number of studies on the treatment of
children with developmental disabilities and dual diagnoses.

View publication stats