Assessment of change-agent expectations in reversal-to-baseline designs

Questionnaires similar to those used in previous research to assess behavioral expectations produced by differing reversal-to-baseline rationales were presented to 160 male and female parents in rural South Dakota. Given two case histories, respondents were asked to predict, as simulated change agents, the frequency of problem behavior during reversal in an ABAB design intervention. The five rationales for reversal were: (a) withdrawal, (b) withdrawal and counter-expectancy, (c) problem switch and counter-expectancy, (d) fade and counter-expectancy and (e) inert treatment and counter-expectancy. While obtained means were close to those of a previous study, there were no significant sex or treatment effects. This fails to confirm the earlier results obtained from university undergraduates. Implications for internal validity of ABAB designs are discussed.