Toward a plurality of methods in project evaluation: a contextualised approach to understanding impact trajectories and efficacy

Understanding the efficacy of development projects requires not only a plausible counterfactual but also an appropriate match between the shape of impact trajectory over time and the deployment of a corresponding array of research tools capable of empirically discerning such a trajectory. At present, however, the development community knows very little, other than by implicit assumption, about the expected shape of the impact trajectory from any given sector or project type, and as such is prone to routinely making attribution errors. Randomisation per se does not solve this problem. The sources and manifestations of these problems are considered, along with some constructive suggestions for responding to them.