Single-Subject Designs and Developmental Psychology
Michael Perone West Virginia University
The objective of research design is straightforward: The researcher must collect, analyze, and present data in such a manner as to persuade an audience of critics that the conclusions drawn from the data are reasonable. The study of research methods is just an analysis of the characteristics of cases that are, or logically should be, persuasive. Methodologists have eased the researcher's task by providing rules for experimental design and analysis. Adherence to these rules is likely to be reinforced, first because they do facilitate productive research, and second because the scientific community tends to reinforce behavior that corresponds with its conventions. (The community tends to punish unconventional behavior by denial of publications and research grants.)
Although conducting research according to the rules leads to scientific and social reinforcers, one must occasionally break the rules to discover whether a richer schedule of reinforcement is waiting to be tapped. Operant psychologists are beginning to study rule-governed behavior in the laboratory ( Hayes, 1989). The results to date suggest that although rules are of great value, they do have a serious drawback: They tend to insulate the rule-follower from the environment, an environment that might reinforce other forms of behavior more generously. These alternative sources of reinforcement may not be contacted when adherence to rules is strong; and without such contact, no change in behavior is likely to occur (e.g., Galizio, 1979).
Graduate training in psychology tends to produce research-related behavior that is strongly rule-governed. And whether the rules are learned by way of McNemar or Hays or Kirk or Keppel or Winer, they may be traced