Evaluation in multi-actor policy processes: accountability, learning and cooperation

Two main functions of evaluation are to enable accountability and collective learning. Both of these – and their combination – run into divers complications when applied in complex multi-actor policy processes. The article explores these complications and illustrates these with examples from the field of spatial policy. In doing so a third function of evaluation in such contexts is identified, viz. evaluation as an instrument of cooperation. Next, a number of theoretical ideas, supported by empirical research, are proposed in order to understand better when, why and how evaluation contribute to complex multi-actor policy processes. Based on these insights some principles are elaborated for the development of constructive evaluation arrangements. It is suggested that cooperation is a precondition for preservation of accountability and learning functions of evaluation in multi-actor settings.