Clinical Psychology

Abstract

Rapid advancement in the understanding and amelioration of complex clinical phenomena places heavy demands on clinical scientists for thoughtful articulation of theoretical models, methodological expertise, and statistical rigor. This article provides a descriptive and prescriptive overview of the use of statistics in clinical psychology. Five classes of research questions that clinical psychologists commonly address are described: the definition and measurement of constructs, the evaluation of group differences, the prediction of group membership, the examination of a network of constructs, and the evaluation of research questions across studies or data sets. A set of four principles that underlie the use of statistics in clinical research are described: the importance of theory to the rapid advancement of clinical science; the centrality of model specification, evaluation, and comparison to clinical research; the need for researchers to master research design and the mechanics of statistical techniques; and the limitations of null-hypothesis statistical testing (NHST).