A multilevel meta-analysis of single-subject experimental design studies

A multilevel approach is proposed to synthesize the results of single-subject experimental research by combining either raw data or effect-size measures. A multilevel meta-analysis of single-subject data allows us to make inferences not only about the effects for the participants that are investigated, but also about the generality of the effects. It also permits a systematic exploration of the possible moderating effect of study and participant characteristics. We also discuss the possibility of combining single-subject experimental research and group-comparison research in one single meta-analysis, and the difficulties involved, especially in making effect-size measures comparable for both kinds of studies.