Analysis of cognitive change: a commentary on Keith et al. (2002)

There have been significant advances in the study of cognitive change in recent years, evolving from traditional group-level parametric methods to new techniques such as the reliable change index (RCI) and the standardized regression-based (SRB) change score. In the very interesting study of cognitive change following cardiopulmonary bypass, Keith et al. (2002) opted for a more traditional group-level analysis of change (the mixed factor analysis of variance). They cited several limitations of the individual-level analytic techniques such as the RCI. In this commentary, the author attempts to address these specific concerns as well as some of the more general difficulties presented by the analysis of cognitive change. Advantages and disadvantages of the RCI and SRB methods are discussed.