Rational emotive behavior therapy (REBT) has achieved positive results in quantitative reviews of treatment outcome studies. In part because of methodological limitations of these studies, however, the generalizability of their favorable results to routine clinical practice is unknown. Also unknown are the clinical significance of outcomes achieved by REBT, the contribution made by its distinctive and specialized procedures, and the types of patients for whom REBT is particularly useful. Recommendations are provided for refining our knowledge of the effectiveness of REBT.Forty years ago, rational therapy (RT) appeared particularly wellsuited to the challenges of treatment outcome research. Albert Ellis had described a clear and parsimonious model of emotional disturbance and its treatment, and he promoted the scientific evaluation of his approach with his own pioneering example (Ellis, 1957). Rational therapy was, among other things, an empiricist's rejoinder to the less testable formulations of psychoanalytic theory (Ellis, 1992). It might well have seemed likely that a data-based verdict on the effectiveness of RT and its explanations of psychological distress would soon be forthcoming.However, twenty years after Ellis' inaugural effort at treatment validation, several distinguished psychotherapy scholars reviewed the