CGD “Evaluation Gap” Report

Lots of people will agree withthe general message of the Center for Global Development’s Evaluation Gap Working Group report: there is a dearth of rigorous evidence that independently evaluates the outcomes of development projects.

More controversial will be their recommendation: that a new international entity be formed to address the shortage of meaningful knowledge being produced about the success and failure of development enterprises. Would such a council truly be independent? How would it be funded?

The problem that the report addresses is familiar to many in the development field, but we have yet to identify a credible mechanism to overcome the knowledge gap. This document’s proposal can serve as a starting point for debate about the nature of such mechanisms.

3 thoughts on “CGD “Evaluation Gap” Report”

  1. Unknown's avatarP.H.

    And might it not just add an additional layer of costs and delay? Would have been nice if they had at least discussed a little more ideas such as output-based contracts, market-based incentives and randomized trails.

  2. Jonathan Dingel's avatarJonathan Dingel

    They do discuss randomized trials. Check out pages 15 & 21 and appendices G & H. One excerpt:

    Remembering that the point of departure is always the policy question and context, with the methodological choice following, it is usually worth asking whether a random-assignment approach—that is, randomly choosing which individuals, families, or communities will be offered a program and which will not—is appropriate and feasible. Where this method can be applied, it ensures that impact measurements are not confounded by systematic differences between beneficiary and control groups.

    Where random assignment cannot be applied, either because it is not appropriate to the policy question or because it is not feasible, other approaches can be applied, such as controlled before-and-after studies, interrupted time-series studies, and various types of matched comparison studies (such as difference-of-differences and propensity score matching).

  3. William Savedoff's avatarWilliam Savedoff

    Disclosure: I’m one of the report’s co-authors.

    I wanted to respond to P.H.’s concerns about the new initiative creating delay’s and new layers. The report actually presents several institutional options for moving forward – an interagency committee, a subprogram of an existing institution, or an independent council. That is as far as the report goes and it presents these options with pros and cons to the people and institutions who will, hopefully, take up the challenge and do something about it.

    My own personal view is that setting up an independent council to promote good impact evaluations could actually reduce delays because I think it could be more agile and active than current institutions.

    I’d be interested in hearing what others think.

Comments are closed.