10 thoughts on “Links I liked

  1. They went from a possible 60,000 to 2400 actually put in the database covering 120 countries over the last few decades. What was wrong with the 57,600. Forget the past especially if it is bad, question mark.

  2. What I was wondering was, from the definition on the site that it measures the net change in outcomes of a development program, does the repository include any measures on USAID and World Bank programs or is it only impact evaluations on smaller development interventions or even rct research. The former two biggies have been spending billions over the years with 1000’s of projects. Or have there been, as yet, no impact evaluations of these biggies. Who impact evaluates USAID project Y of $5million to do X in developing country T. E.g. USAID gives out 10,000 MT of seed to produce 100,000 MT of grain.
    It does also look as if some of the evaluations are of research not actual projects, e.g. the rct type of research. That is really research on research. But I may be getting it wrong.

  3. Great questions! At 3ie we love impact evaluations of actual programs, especially big ones. As you suggest, they are harder to do. We have certainly tried to capture any that have been done, but we do screen on counterfactual-based methodologies. Both institutions have been doing a great job recently funding and promoting more impact evaluations of their projects, so we expect to have many more to index in the future. If you know of existing impact evaluations of actual projects that we have missed, please do let us know!

  4. The discontent on the impact of development, no doubt, led to the rise of impact evaluation.
    It is right that a repository is made available to assist in the improvement of the impact of development. 3ie are making a good start and are to be encouraged.
    If I may suggest 3ie could write all the biggies asking them to produce impact evaluations for all their projects and lodge them in the repository, from today forwards. Then 3ie could fund assessment impact evaluations of ten per cent of these submitted evaluations. There is not enough external assessment of Ngo impacts.