ABSTRACT
In this chapter, we describe single-subject designs, a subcategory of quasi-experimental time-series designs that can be used with one or a few participants. These single-subject designs have many of the characteristics that govern traditional time-series designs with groups of participants. They have numerous repeated measures on each participant and the initiation and withdrawal of treatment. Traditionally, the data from singlesubject designs had infrequently been analyzed using statistical methods, and when those methods were applied, they were often unique to these designs. Recently, there have been increased efforts to apply statistical analyses to single-subject designs, and some of these analyses have used more traditional statistics. In addition, efforts have been made to combine results from different single-subject studies on a similar topic into a meta-analysis to increase generalizability. The topic of single-subject designs is quite complex and contains too much material to be covered completely in a single chapter. For those interested in a complete treatment of the topic, we suggest the text by Kazdin (1982). For a clinical perspective on single-subject designs, Ottenbacher (1986) and Skinner (2005) are recommended. For excellent reviews of the evaluation of singlesubject designs, we recommend Franklin, Gorman, Beasley, and Allison’s (1997) treatment of the visual analysis and Gorman and Allison’s (1997) treatment of the statistical analysis.
