Realistic Evaluation and Case Studies

Abstract

Embedded in the critical realist tradition, this article aims to explore the potentialities of the case study for evaluation purposes when complexity and specificity are moderate.Three objectives are pursued. First, it is stated that the focus of the evaluation effort is not necessarily the programme itself, but can be one premise upon which the programme is built. Second, taking into account Donald Campbell's contributions, it is asserted that, under favourable conditions, not only deterministic propositions, but also probabilistic ones can be rejected as highly improbable via case studies. Third, making use of the Context—Mechanism—Outcome model proposed by Pawson and Tilley, this article demonstrates the capacity of a critical case study to sustain theory building.