Answers are grounded in this paper and cite their sources.
Background: A complexity-aware approach, realist evaluation is ideal for norms-shifting interventions (NSIs), which are not well-understood but gaining prominence in behaviour change programming in Africa and globally to foster enabling socionormative environments that sustain behaviour change. A new application of realist evaluation to NSIs uses an adapted approach employing realism values that is suitable for social and behaviour change (SBC) programme evaluation more generally. Objectives: This article shares the authors’ reflections on tailoring realist evaluation approaches for use with community-based norms-shifting programmes. It describes how realist evaluation enables co-building of programme theory that conceptually underpins NSIs, guides evaluation efforts and yields benefits beyond theory-proving. Method: Two NSIs in Niger and Senegal illustrate how locally refined theories of change (TOC) and identification of evidence gaps in causal pathways guided a series of rapid programme and quasi-experimental outcome studies. Over two years externally and internally led studies assessed intermediate or mediating norms-shifting effects and outcomes comprising the realist evaluation. Studies drew from experiential, existing and new data. Results: The tailored approach created a co-owned evaluation, from joint exploration of SBC theory to review of evidence generation. Five values applied to the research–practice partnerships reinforced a realist perspective: participatory, complexity, shared ownership, practice-oriented and valuing all forms of data. Conclusion: Bounded by TOC exploration for programme inquiry, realist evaluation embeds learning and assessment concretely into local programming and knowledge building. Integrating evaluation practice with realism values creates a nexus and a unique and significant dynamic between programme implementers and evaluators that transcends NSI research and programme practice.
No takes yet. Share an insight, caveat, or question.
Igras et al. (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: