Researchers and research managers need to know whether and how their interventions are working to be able to adapt and improve their programmes as well as to be able to satisfy their funders. This is especially true in a research-for-development context where research competes with other worthy alternatives for overseas development assistance funding and where highly complex social, economic and ecological environments add to evaluation challenges.

This paper presents a theory-based research evaluation approach that was developed and tested on four policy-relevant research activities. Each research evaluation began with documentation of a theory of change (ToC) that identified key actors, processes and results. Data collected through document reviews, key informant interviews and focus group discussions were analysed to test the ToCs against evidence of outcomes in the form of discourse, policy formulation and practice change. The approach proved valuable as a learning tool for researchers and research managers and it has facilitated communication with funders about actual and reasonable research contributions to change

Evaluations that employed a participatory approach with project scientists and partners noticeably supported team learning about past work and about possible adaptations for the future. In all four cases, the retrospective ToC development proved challenging and resulted in overly-simplistic ToCs.

This theory-based approach to research evaluation provides a valuable means of assessing research effectiveness (summative value) and supports learning and adaptation (formative value) at the project or programme scale. The approach is well suited to the research-for-development projects represented by the case studies, but it should be applicable to any research that aspires to have a societal impact.