SINGLE-SUBJECT (SMALL-n) EXPERIMENTAL DESIGNS

• In applied behavioral analysis, the methods developed within the experimental analysis of behavior are applied to socially relevant problems.
In the remainder of this chapter we will describe single-subject experimen- tal (small-n) designs. These experimental designs have their roots in an ap- proach to the study of behavior that was developed by B. F. Skinner in the 1930s. The approach is called an experimental analysis of behavior. It presents a unique behavioral view of human nature that not only contains prescrip- tions for the way psychologists should do research but also has implications for the way society should be organized. Several of Skinner’s books, includ- ing Walden Two and Beyond Freedom and Dignity, describe how the principles derived from an experimental analysis of behavior can be put to work to im- prove society.
In the experimental analysis of behavior (unlike the group methodologies discussed in previous chapters), it is often the case that the sample is a single subject or a small number of subjects (small-n). Experimental control is demon- strated by arranging experimental conditions such that the individual’s behav- ior changes systematically with the manipulation of an independent variable. AS Skinner (1966) commented,
Instead of studying a thousand rats for one hour each, or a hundred rats for ten hours each, the investigator is likely to study one rat for a thousand hours. The procedure is not only appropriate to an enterprise which recognizes individual- ity; it is at least equally efficient in its use of equipment and of the investigator’s time and energy. The ultimate test of uniformity or reproducibility is not to be found in the methods used but in the degree of control achieved, a test which the experimental analysis of behavior usually passes easily.
In applied behavior analysis, the methods that are developed within an experimental analysis of behavior are applied to socially relevant problems. These applications are frequently referred to as behavior modification, but when ap- plied to clinical populations the term behavior therapy is preferred (Wilson, 1978). Behavior therapy is seen by many psychologists as a more effective approach to clinical treatment than that based on a psychodynamic model of therapy. Instead of seeking insight into the unconscious roots of problems, behavior therapy focuses on observable behavior. Forexample,self-stimulatory behaviors (e.g.,pro- longed body rocking, gazing at lights, or spinning) that often characterize autistic children may be conceptualized as behaviors under the control of reinforcement contingencies. In this way, clinicians and teachers may be able to control their frequency of occurrence by using behavior modification techniques (see Lovaas, Newsom, & Hickman, 1987). Numerous studies have been published showing how behavior modification and behavior therapy can be employed successfully to change the behavior of stutterers, normal and mentally impaired children and adults, psychiatric patients, and many others. Approaches based on applied behavior analysis have also been successfully used by school psy- chologists in educational settings (see Kratochwill & Martens, 1994). A primary source for these published studies is the Journal of Applied Behavior Analysis.
Characteristics of Single-Subject Experiments
• Researchers manipulate an independent variable in single-subject experiments; therefore, these designs allow more rigorous control than case studies.
• In single-subject experiments, baseline observations are first recorded to
describe what an individual’s behavior is like (and predicted to be like in the future) without treatment.
• Baseline behavior and behavior following the intervention (treatment) are
compared using visual inspection of recorded observations.
ADVANTAGES OF SINGLE-SUBJECT DESIGNS OVER GROUP DESIGNS: LESS CAN BE MORE
Single-subject experimental designs may be more appropriate than multiple-group designs for certain kinds of applied research (see Hersen & Barlow,
1976). One such situation is when research is di- rected toward changing the behavior of a specific individual. For example, the outcome of a group experiment may lead to recommendations about what treatments are effective “in general” in modi- fying behavior. It is not possible to say, however, what the effect of that treatment would be on any particular individual based on a group average. Kazdin (1982) summarizes this characteristic of single-subject experiments well: “Perhaps the most obvious advantage [of single-case experimental designs] is that the methodology allows investiga- tion of the individual client and experimental evalu- ation of treatment for the client”
Another advantage of single-subject experi- ments over multiple-group experiments involves
the ethical problem of withholding treatment that can arise in clinical research. In a multiple-group design, a potentially beneficial treatment must be withheld from individuals in order to provide a control group that satisfies the requirements of internal validity. Because single-subject ex- perimental designs contrast conditions of “no- treatment” and “treatment” within the same individual, the problem of withholding treatment can be avoided. Moreover, investigators doing clinical research often have difficulty gaining ac- cess to enough clients to do a multiple-group ex- periment. For instance, a clinician may be able to identify only a few clients experiencing claustro- phobia (excessive fear of enclosed spaces). The single-subject experiment provides a practical solution to the problem of investigating cause- effect conclusions when only a few participants are available.