This article addresses the long-standing divide between researchers and practitioners in the field of psychotherapy, regarding what really works in treatment and the extent to which interventions should be governed by outcomes generated in a “laboratory atmosphere.” This alienation has its roots in a positivist paradigm, which is epistemologically incomplete because it fails to provide for context-based practical knowledge. In other fields of evaluation research, it has been superseded by a mixed methods paradigm, which embraces pragmatism and multiplicity. On the basis of this paradigm, we propose and illustrate new scientific standards for research on the evaluation of psychotherapeutic treatments. These include the requirement that projects should comprise several parallel studies that involve randomized controlled trials, qualitative examinations of the implementation of treatment programs, and systematic case studies. The uniqueness of this article is that it contributes a guideline for involving a set of complementary publications, including a review that offers an overall synthesis of the findings from different methodological approaches. (PsycINFO Database Record (c) 2010 APA, all rights reserved)