Single-Case Research Design and Analysis

New Directions for Psychology and Education

Thomas R. Kratochwill, Joel R. Levin, Thomas R. Kratochwill, Joel R. Levin

• Publisher: Routledge
• Year: 2015
• eBook ISBN: 9781317535645
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

Originally published in 1992, the editors of this volume fulfill three main goals: to take stock of progress in the development of data-analysis procedures for single-subject research; to clearly explain errors of application and consider them within the context of new theoretical and empirical information of the time; and to closely examine new developments in the analysis of data from single-subject or small n experiments.

To meet these goals, this book provides examples of applicable single-subject research data analysis. It presents a wide variety of topics and perspectives and hopes that readers will select the data-analysis strategies that best reflect their methodological approaches, statistical sophistication, and philosophical beliefs. These strategies include visual analysis, nonparametric tests, time-series experiments, applications of statistical procedures for multiple behaviors, applications of meta-analysis in single-subject research, and discussions of issues related to the application and misapplication of selected techniques.

1 Single-Case Research Design and Analysis: An Overview

Thomas R. Kratochwill
University of Wisconsin-Madison

There is little doubt that single-case designs play a major role in applied and clinical research in psychology, education, and related fields. The rapid proliferation of writings on single-case design is remarkable over the past 15 years, and especially since the author of this chapter edited a volume devoted to single-subject research and strategies for evaluating change (Kratochwill, 1978). At that time there was only a handful of professional books devoted to this methodology and related issues. At this writing, the picture has changed dramatically, with a considerable number of professional works on both design and data analysis in the literature. Table 1.1 provides a listing of major textbooks in the area of single-case research design and analysis. It can be observed that although work in this area has a solid foundation prior to 1980, many major works have appeared since that time.

An interesting phenomenon in reviewing these works is that authors have discussed single-case methodology as a specialized research application in a variety of fields. For example, beyond the more traditional discussion of the methodology within applied behavior analysis (e.g., Bailey, 1977; Johnson & Pennypacker, 1980), there are presentations in clinical psychology (e.g., Barlow & Hersen, 1985), social work (e.g., Fischer, 1978), special education (e.g., Tawney & Gast, 1984), and communicative disorders (e.g., McReynolds & Kearns, 1983). That is, whereas single-case research design once appeared to have origins within quasi-experimental investigations in the tradition of Campbell and Stanley (1963) and applied behavior analysis with origins in the experimental analysis of behavior (Sidman, 1960), this picture has rapidly changed and the contributions are apparent across a variety of professional fields.

This widening influence of single-case methodology as a primary contribution

TABLE 1.1
Major Textbooks in the Area of Single-Case Research Design and/or Analysis

Date	Author	Textbook Title
1960	Sidman, M.	Tactics of scientific research New York: Basic Books.
1969	Davidson, P. 0., & Costello, C. G. (Eds.)	N= 1: Experimental studies of single cases New York: Van Nostrand/Reinhold.
1973	Jayaratne, S., & Levy, R. L.	Empirical clinical practice. Irvington, NY:Columbia University Press.
1975	Glass, G. V., Wilson, V. L, & Gottman, J. M.	Design and analysis of time-series experiments. Boulder, CO: Colorado Associated University Press.
1975	Bailey, J. S.	A handbook of research methods in applied behavior analysis. Gainesville, FL University of Florida.
1978	Fischer, J.	Efficient casework practice: An eclectic approach. New York McGraw-Hill.
1978	Kratochwill, T. R. (Ed.)	Single subject research: Strategies for evaluating change. New York: Academic Press.
1979	Cook, T. D., & Campbell, D. T. (Eds.)	Quasi-experimentation: Design and analysis issues for field settings. Chicago: Rand McNally.
1979	Robinson, P. W., & Foster, D. F.	Experimental psychology: A smail-n approach. New York: Harper & Row.
1980	McCleary, R., & Hay, R. A., Jr.	Applied time-series analysis for the social sciences. Beverly Hills, CA: Sage.
1980	Johnson, J. M., & Pennypacker, H. S.	Strategies and tactics ofhuman behavioral research. Hillsdale, NJ: Lawrence Erlbaum Associates.
1980	McDowall, D., McCleary, R., Meidfinger, E. E., & Hay, R. A., Jr.	Interrupted time-series analysis. Beverly Hills, CA: Sage.
1981	Gottman, J. M.	Time-series analysis: A comprehensive introduction forsocial scientists. Cambridge, England:Cambridge University Press.
1981	Wodarski, J. S.	The role ofresearch in clinical practice: A practical approach forhuman services. Baltimore: University Park Press.
1982	Kazdin, A. E.	Single-case research designs: Methods for ciinicai and applied settings. NewYork: Oxford University Press.
1982	Kazdin, A. E., & Tuma, A. H. (Eds.)	Single-case research designs. San Francisco, Jossey-Bass.
1983	McReynolds, L. V., & Kearns, K. P.,	Single-subject experimental designs in communicative disorders. Baltimore: University Park Press.
1984	Barlow, D. H., Hayes, S. C., & Nelson, R. O.	The scientist practitioner: Research and accoutability in clinical and educational settings. New York: Pergamon.
1984	Tawney, J. W., & Gast, D. L.	Single subject research in special education. Columbus, OH: Merrill.
1984	Yin, R. K.	Case study research: Design and methods. Beverly Hills, CA: Sage.
1985	Barlow, D. H., & Hersen, M.	Single case experimental designs: Strategies for stydying behavior change (2nd ed.). New York: Pergamon.
1985	Behling, J. H., & Merves, E. S.	The practice of ciinicai research: The singie-case methods. Lanham, MD: University Press of America.
1986	Bromley, D. B.	The case-study method in psychology and related disciplines. New York: Wiley.
1986	Cryer, J D.	Time series analysis. Boston, MA: Duxbury Press.
1986	Poling, A., & Fuqua, R. W. (Eds.)	Research methods in applied behavior analysis: Issues and advances. New York: Plenum Press.
1986	Valsinger, J. (Ed.)	The individual subject and scientific psychology. New York: Plenum Press.

Source: Adapted from Kratochwill and Williams (1988). Personal perspectives on pitfalls and hassles in the conduct of single subject research. Joumal oftheAssociation ofPersons with Severe Handicaps, 13, 147-154. Reproduced by permission.

to scientific information in these fields is interesting. But more importantly, the application of this methodology has allowed a number of refinements of application to specific and unique problems that may not have occurred without such diverse application across a variety of scholarly fields. Indeed, several specific advances may be pinpointed. First of all, specific advances have occurred in reconceptualizing the contributions of traditional "case-study" methodology. Traditional case-study methodology has typically been relegated to a rather low level of scientific knowledge. In fact, most case-studies involve demonstrations that do not allow an investigator to rule out conventional threats to internal validity and, therefore such studies do not quality as "true experiments" (Campbell & Stanley, 1963).

Several writers have presented the contributions that case-studies make and have presented the features that allow one to draw valid inferences from case-study and single-case design (Kazdin, 1981, 1982; Kazdin, Kratochwill, & VandenBos, 1986; Kratochwill, 1985; Kratochwill, Mott, & Dodson, 1984). In this regard, two important points are necessary to reconsider the role of case-studies in contributing valid information to scientific information. First, this methodology can be distinguished along a series of types that provide contributions to research knowledge. Table 1.2 presents three major types of case-study investigations that can be distinguished in applied and clinical research.

TABLE 1.2
Types of Case-Study Investigations

Type	Characteristics
Nontherapeutic case study	Researcher is interested in nonclinical investigation. Such areas as developmental or educational psychology would be representative. Includes traditional baby biographies.
(1) descriptive/uncontrolled
(2) biography/autobiography
Assessment/diagnosis case study	Researcher uses various psychometric instruments for diagnosis or description of cognitive or social behavior.
(1) descriptivecase
Therapeutic/intervention case study	Researcher is primarily interested in a clinical disorder and may either describe natural course of disorder or develop intervention to treat client’s problem.
(1) uncontrolled
(2) pre-experimental
(3) clinical replication case

Source: Kratochwill, Mott, and Dodson (1984). Case study and single-case research in clinical and applied psychology. In A. S. Bellack and M. Hersen (Eds.), Research methods in clinical psychology(pp. 35-99). New York: Pergamon.

The most common case-study methods are those used within clinical and other areas of applied psychology and which focus on outcome evaluation of specific intervention techniques. These case-studies have the "appearance" of an experiment in that some independent variable is manipulated. Nevertheless, many of these traditional case-studies are characterized by the absence of experimental control, inasmuch as the researcher cannot rule out most rival interpretations that might account for the observed therapeutic change. Since considerable ambiguity surrounds the interpretation of case-study methodology, these studies have been correctly relegated to a rather poor standing in the scientific community. Nevertheless, when various internal validity criteria are applied to case-study or even more well-controlled single-case designs, it is clear that ruling out various threats to validity is not an all or none matter (Kazdin, 1982). Attention to a variety of methodological, conceptual, and assessment factors is required to rule out various plausible factors that could account for observed effects. That is, whether a specific threat to internal validity of an experiment has been eliminated is often a matter of degree and, depending on how the study is arranged, various threats to validity might actually be controlled or rendered implausible.

To understand how the researcher can draw valid inferences, it is useful to note that the traditional uncontrolled case-study is characterized by a variety of unsystematic assessment and anecdotal reports, as well as the lack of design structure. However, numerous tactics can be used to advance understanding of contributions of case-studies and draw more valid inferences from the investigation. Table 1.3 provides a list of some dimensions that can be taken into account in either designing the study or evaluating an intervention case-study in the

TABLE 1.3
Improving Valid Inferences from Case-Study Research

Research Characteristics	Low-Inference Strategies	High-inference Strategies
Type of data	Subjective data are used. For example, subjective anecdote description of outcomes is the fof...

Table of contents

• Cover
• Title
• Copyright
• Original Title
• Original Copyright
• Contents
• Preface
• Dedication
• 1. Single-Case Research Design and Analysis: An Overview
• 2 The Visual Analysis of Data, and Current Research into the Stimuli Controlling It
• 3. Philosophical and Statistical Foundations of Time-Series Experiments
• 4. The Intensive Examination of Social Interaction
• 5. Nonparametric Tests for Single-Case Experiments
• 6. Statistical Analysis in Single-Case Research: Issues, Procedures, and Recommendations, with Applications to Multiple Behaviors
• 7. Meta-analysis for Single-Case Research
• 8. Single-Case Research Design and Analysis: Comments and Concerns
• Author Index
• Subject Index