small N designs Flashcards
research in psychology began with a small N
Research in psychology’s earliest years normally involved a small number of participants, no statistical summaries of data, and, because they had yet to be invented, no inferential statistical analyses. The additional participants served the purpose of replication. Dresslar’s facial vision study and Thorndike’s research on cats escaping from puzzle boxes are good examples.
reasons for small N designs
Individual‐subject validity is the extent to which a general conclusion drawn from research with large N applies to the individual participants in the study. One reason for favoring small N designs is that when summarizing data for a large number of individuals, individual‐subject validity can be weak. This occurs in some concept learning research, for instance, in which the summarized data imply gradual learning but an examination of individual behavior implies a more rapid change in performance. Small N designs are also favored when studying participants from rare populations or when practical considerations (e.g., expense) make a large N design impossible.
the experimental analysis of behaviour
the philosophical basis for some small N research is the position taken by B. F. Skinner, who argued that if a researcher is able to establish sufficient control over environmental influences, then orderly behavior will occur and can be easily observed (i.e., without statistical analysis). In operant conditioning, behavior is influenced by its consequences and comes under the control of the environment in which it occurs. As a dependent variable, Skinner relied on rate of response. Most of Skinner’s work was in basic research (e.g., schedules of reinforcement), but he was a strong advocate for the technological ideal—the practical application of operant principles and techniques to bring about societal change.
small N designs in applied behaviour analysis
Applied behavior analysis is the application of operant principles to improve behavior. Small N or single‐subject designs are used to evaluate the effectiveness of these applied programs. All single‐ subject designs carefully define terms, establish baselines, and then apply an intervention technique to be evaluated. Withdrawal designs establish a baseline, apply a treatment, and then withdraw the treatment. If the behavior changes along with the changing conditions, it is assumed the treatment was effective in altering the behavior. A‐B‐A‐B designs are preferred over A‐B‐A designs because the study ends with the treatment in effect and the treatment is tested twice. Multiple baseline designs are often used when withdrawal designs are not feasible or ethical. Multiple baselines can be established for several individuals, several behaviors, or several environmental settings. Changing criterion designs are used when the target behavior must be shaped gradually (e.g., weight loss). Alternating treatment designs allow for the comparison of two or more treatments in a single study. Small N designs have been criticized for being unable to evaluate interactive effects, relying on a single dependent variable, and having limited external validity.
case study designs
Most case studies are detailed analyses of the behavior of single individuals, although some case studies investigate unique events or identifiable groups. They are often used in a clinical setting to study individuals with relatively common disorders (e.g., head injury from boxing) or to analyze of people with rare attributes (e.g., the memory abilities of S.). They can be useful sources of information and can serve to falsify claims and generate hypotheses for further research, but they are subject to the biases of the investigator, memory failures, and lack of generalizability (external validity).
