ABSTRACT
This chapter examines the conceptual and empirical foundations, clinical utility, and cost-benefits of multisource clinical assessment. One strategy to reduce the likelihood and impact of measurement and judgment errors in clinical assessment is through the use of multiple measures, methods, informants, instruments, settings, contexts, and/or times in clinical assessment. The chapter first examines sources of error variance associated with specific assessment instruments, informants, methods, and time samples, with particular attention given to mono-method assessment strategies (such as interviews, informant questionnaires, and self-report questionnaires. Time-sampling is discussed as an assessment strategy to capture dynamic aspects of a client’s behavior. The chapter reviews sources of error in measurement and clinical judgments, such as criterion contamination, semantic overlap, response bias, reification of test scores, and shared method variance. Finally, the chapter discusses strategies for understanding and drawing inferences from convergent and divergent data obtained from multiple settings and sources, their conditional nature, their implication for construct validation and clinical practice, and z-score transformation as a strategy for forming composite measures. The chapter ends with recommendations for conducting multisource assessment and for interpreting its results.
