This article was downloaded by: [University of West Florida]

On: 07 October 2014, At: 08:42

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Child & Family Behavior Therapy

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/wcfb20

Kazdin, A. E. (2011). Single-Case

Research Designs, Second Edition.
a a
Valerie A. Evans PhD & Saul Axelrod PhD, BCBA-D
a
Temple University, College of Education , Philadelphia , PA
Published online: 14 Mar 2012.

To cite this article: Valerie A. Evans PhD & Saul Axelrod PhD, BCBA-D (2012) Kazdin, A. E. (2011).
Single-Case Research Designs, Second Edition. , Child & Family Behavior Therapy, 34:1, 76-79, DOI:
10.1080/07317107.2012.654458

To link to this article: http://dx.doi.org/10.1080/07317107.2012.654458

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 Child & Family Behavior Therapy, 34:76–91, 2012
Copyright # Taylor & Francis Group, LLC
ISSN: 0731-7107 print=1545-228X online
DOI: 10.1080/07317107.2012.654458

Book Reviews

Kazdin, A. E. (2011). Single-Case Research Designs, Second Edition. New

York, NY: Oxford University Press, xi þ 452 pp., $72.95.

Alan Kazdin, Professor of Psychology and Child Psychiatry at Yale University,

is Director of the Yale Parenting Center and Child Conduct Clinic. The author
Downloaded by [University of West Florida] at 08:42 07 October 2014

of Research Design in Clinical Psychology (2002), Kazdin is an established

authority on research. Single-Case Research Designs (2011) is the second
edition of the book, the first published in 1982. The revision includes a broad
range of disciplines for application and new design options.
There are two basic types of research designs: group-comparison and
single-case, also called the single-subject design. Group-comparison designs
compare a treatment group to a control group, drawing conclusions about
the effect of the intervention (the independent variable) on the target
behavior (the dependent variable). In an attempt to create equal groups,
group-comparison researchers randomly select a large number of parti-
cipants and then randomly assign them to either a control or treatment
group. Measurements of the dependent variable are typically collected at
the beginning and at end of treatment, or just at the end of treatment. The
comparison of the degree to which the treatment and control groups
produce different outcomes allows group-comparison researchers to draw
conclusions about the effects of the intervention.
By contrast, single-subject designs use participants as their own con-
trols. Rather than comparing groups of participants, single-subject research-
ers look at how the intervention changes the behavior of each participant
over time and across conditions. Although there is often more than one par-
ticipant in single-subject experiments, and there are sometimes large num-
bers of participants, each subject’s data are typically analyzed individually.
The most common single-subject design are the reversal design, or ABAB
design (Chapter 6). The reversal design records several data points on each
subject across time in two alternating conditions: baseline (A) and inter-
vention (B). During baseline, several data points (three at minimum) are
recorded and plotted to show visually consistent rates in the dependent vari-
able across time. Following baseline, the independent variable is introduced,
and the dependent variable is measured either for a fixed period of time or
until it shows considerable behavioral improvement. The independent
variable is then withdrawn, and baseline rates are measured until stable.

76
 Book Reviews 77

The independent variable is then re-introduced, and the effects are again mea-
sured in the final phase. The reversal design allows the researcher to
demonstrate change in the dependent variable through withdrawal and
reinstatement of the independent variable. Other rigorous single-subject
designs include the multiple-baseline (Chapter 7) and alternating-treatments
(Chapter 9). Unlike group-comparison designs, the single-subject research
frequently relies on visual analysis, not inferential statistics, to draw
conclusions about changes in rates of the dependent variable.
Research designs are evaluated based on the degree to which change in
the dependent variable may be attributed to the introduction of the inde-
pendent variable. The degree to which the change in behavior may be attrib-
uted to the intervention, and not to confounding factors, is known as internal
Downloaded by [University of West Florida] at 08:42 07 October 2014

validity. Both group-comparison and single-subject researchers are greatly

concerned about internal validity, but achieve it in distinctly different
manners, as described above.
Single-Case Research Designs (2011) was written for professionals and
students with a basic foundation in research design. The preface more
specifically targets group-comparison researchers, with its discussion of com-
mon assumptions and general advantages to single-subject research. Follow-
ing the preface, the author aims to reach a broader audience. The book
maintains a balanced presentation of single-subject designs throughout its
five parts. Beginning with a background to the designs, Kazdin provides
an historical context to research-design development. The development of
contemporary single-subject designs is attributed to the work of B. F. Skinner
and subsequent researchers in Applied Behavior Analysis (ABA; e.g., Baer,
Wolf, & Risley, 1968). The book’s natural progression to assessment is logical
and necessary to understand how single-subject research is distinguished
from group research. Kazdin provides detailed analyses for methods of defin-
ing and measuring the dependent variable with regard to design options,
reliability, and validity. The large middle part of the book is dedicated to
the description of single-subject designs. Design methods are described in
detail with many applied examples. Although Kazdin promotes the notion
of distinguishing single-subject research designs from group designs based
on the research question at hand, he also cites benefits of single-subject
designs—including lower cost, multiple measurements, and flexibility to
respond to design issues as they arise during the study. The fourth part
covers the evaluation of single-subject data, and the book concludes with
perspectives and a critical look at design problems.
Although the number of published research articles using single-subject
research designs seems to be increasing, the research methods are still, to
some degree, restricted to ABA and special-education research topics. A bar-
ometer for change is when single-subject research is applied to studies out-
side these areas. Kazdin’s examples for application are mostly clinical in
nature but apply to all areas of practical concern. Additionally, Kazdin
 78 Book Reviews

provides many examples of familiar behaviors, such as recycling and

completing homework assignments.
Kazdin makes efforts to bridge the gap between group-comparison and
single-subject designs by highlighting, for the group-comparison design
researcher, the contributions of single-subject designs that may be borrowed
by group-comparison experimenters. These contributions include social-
validation techniques and demonstrations of maintenance. Social validation
means, in part, that the outcomes of the study will have practical importance
in the participant’s life. It is not uncommon for single-subject studies to
involve input from family and teachers when defining the dependent
variable and developing the intervention. Peers also play a large role in
determining the degree to which behavior is within an acceptable range of
Downloaded by [University of West Florida] at 08:42 07 October 2014

performance. Maintenance refers to the ongoing effects of the intervention

following the study’s completion. Although demonstrations of maintenance
are considered best practice in single-subject designs, it is still optional for
publication purposes to demonstrate treatment effects following intervention
in group-comparison research.
Throughout the book, Kazdin describes the benefits of repeated
measurements, such as providing the researcher the opportunity to make
adjustments to improve outcomes. Additional or different procedures, such
as a change in the delivery of a reinforcer, may be a slight adjustment that
produces the desired effect. Between-group designs often only take a pre-
and a post-measure so treatment effects are only apparent after the inter-
vention being tested has concluded. The benefit of repeated measurements
also underscores the special attention single-subject designs give to individ-
ual differences. Kazdin cites the example of a study training six children not
to play with guns (Miltenberger et al., 2007, as cited in Kazdin, 2011). Half the
children responded to the intervention, while two others required additional
training trials, and the third required an incentive. If conducted as a between-
groups design, the findings would possibly show that statistical significance
had been achieved, even though half of the subjects originally did not attain
the desired goal. If other procedures were to be tested, such as the prompt-
ing and reinforcement provisions, the study would need to be conducted
again. Conducted as a single-subject design, however, the researchers
showed that half of the subjects needed additional supports, which were
then provided, making the intervention effective for all subjects. This type
of information has significance for practitioners applying programs to diverse
populations, especially where individual failure is unacceptable.
When social scientists discuss the attributes of single-subject versus
group-comparison research designs on the issue of external validity (i.e.,
generalizing the results of a study to the population at large), the conver-
sation often stops with the proponents of group-comparison designs
declaring victory. The argument is that since group-comparison studies
usually involve many subjects, and single-subject studies may involve as
 Book Reviews 79

few as one subject, group-comparison research more likely leads to general

truths. The argument is flawed in at least three respects. First, for group-
comparison researchers to generalize to the population at large, they must
draw samples from the population at large, a condition that is rarely met.
(Mostly, samples are chosen due to convenience and availability.) Second,
group-comparison researchers often generalize to a group average. Yet,
the group average may not resemble any individual’s performance in the
group. Third, if psychology is to be the study of the individual, as it is often
touted, no subject’s data may be dismissed through the averaging process.
People who engage in single-subject research try to generalize across
subjects by attempting to replicate outcomes, first with similar and, then with
dissimilar subjects. Yet, such replications seldom appear in print. The pos-
Downloaded by [University of West Florida] at 08:42 07 October 2014

ition of the present reviewers is that neither approach does particularly well
when external validity is carefully examined. Our advice to researchers:
Study questions of social importance. Use either a single-subject or group-
comparison design, depending on which one better answers the question
being considered. Next, replicate important studies and do not worry much
about the external validity of each study. Replicated, well-conducted studies
will eventually make truths self-evident.
Kazdin’s message is consistent: group-comparison and single-subject
designs are complementary. He explains how the research question and
variables surrounding the application of the research indicate the best meth-
odological fit. The book is rich with technique and approaches each step of
the research process analytically. No matter how familiar one might be with
research design, Kazdin gives every reader something new to consider about
single-subject designs.

REFERENCE

Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied
behavior analysis. Journal of Applied Behavior Analysis, 1(1), 91–97.

Valerie A. Evans, PhD

valeriea@temple.edu
Temple University
College of Education
Philadelphia, PA

Saul Axelrod, PhD, BCBA-D

axelrod@temple.edu
Temple University
College of Education
Philadelphia, PA