The More Things Change, the More They Stay the Same

Abstract

The results of the clinical versus statistical prediction meta-analysis support the longstanding finding that statistical methods are superior to clinical methods in making predictions. In this reaction, the authors discuss the findings from the perspective of the context of discovery and the context of justification, the client model-building process, bias in clinician judgment, and evidenced-based practice, and they speculate about why more clinicians have not made statistical methods part of their practice.