Impact evaluation repository

Curt Carnemark_World Bank_IER

This repository is an index of all published impact evaluations of development interventions. 3ie curates it to ensure the studies we include meet well-defined inclusion criteria. 3ie first launched the repository in 2014, using systematic search and screening methodologies to populate the database. We undertake regular updates, as well as add records when we identify new studies through other 3ie projects. Each record contains basic information about the evaluation and a link to the published study. 

Latest impact evaluations

Repository at a glance

  • The repository holds more than 4,600 records as of June 2018.
  • The top 10 countries where studies have been conducted are India (390), China (281), Mexico (247), Kenya (233), Bangladesh (197), South Africa (194), Brazil (193), Uganda (173), Pakistan (105) and Peru (105).
  • The number of impact evaluations conducted in Brazil, Cambodia, Cameroon, Costa Rica, Malawi and Mozambique has more than doubled since 2014.
  • Central Asia, Central Africa, the Middle East and North Africa continue to be underrepresented.

New article in JDEff: Is impact evaluation still on the rise?

3ie's Shayda Sabet and FHI360's Annette Brown examine data from our repository to highlight new trends in impact evaluation research. The authors used an updated search and screening protocol to identify new impact evaluation studies. They found that the number of studies published between 2010 and 2015 account for almost two thirds of the evidence base. Over half of all studies are from health and education sectors; however, in the last decade, more studies from underrepresented sectors are emerging. Nearly 60 per cent of the studies are from Sub-Saharan Africa and Latin America; however, recent studies are increasingly conducted in underrepresented regions such as the Middle East and North Africa. Read more

3ie populates the repository through a systematic search and screening process using 35 databases, search engines and websites to locate published development impact evaluations. We screen studies found through the search are to ensure they meet the inclusion criteria.

For an impact evaluation to be included, it must be published as a journal article, book chapter, report, or working paper; be conducted in a low- or middle-income country; examine the effectiveness of a specific development intervention; and use an experimental or quasi-experimental estimation strategy 3ie recognizes.

To find out more about 3ie’s search and screening protocol, please visit the links below

First full search (2012–2014)

Second full search (2015–2016)

Third full search (2017–2018)

In addition to our regular updating exercise, we welcome receiving information about new studies that may qualify. If you know of an impact evaluation that you think meets our inclusion criteria, please write to us at

  • In Improving Humanitarian Aid, an article in Foreign Affairs, David Miliband and Ravi Gurumurthy (International Rescue Committee) cite the repository as an important source for evidence on the impact of development interventions on tackling poverty in low- and middle-income countries and make a case for conducting more impact evaluations in fragile contexts.
  • Howard White (Campbell Collaboration) and David Raitzer (Asian Development Bank) highlight core concepts, methods and practical rules of thumb on how to implement successful evaluations for technical and non-economist audiences. They praise and point to 3ie’s public goods, including the Registry for International Development Impact Evaluations the Impact Evaluation Repository and our Microsoft Excel-based power calculator and guide as some of the resources evaluators can use. Download the full book, Impact evaluation of development interventions: a practical guide.
  • Anthony Swan, a researcher at Australia National University, uses 3ie’s impact evaluation repository to look at impact evaluations focused on Papua New Guinea and the Pacific Islands. He highlights low demand, issues of external validity and cost of impact evaluations as reasons behind the scarcity of studies. Read the full post: Evaluating impact evaluation in PNG and the Pacific.