Research Resources | The Abdul Latif Jameel Poverty Action Lab

Our library of practical resources is available for researchers undertaking randomized evaluations and those teaching the technique to others. The resources presented here are curated by J-PAL in partnership with Innovations for Poverty Action (IPA).

Guide to Publishing Research Data

J-PAL promotes the publication of de-identified data from randomized evaluations. This resource provides guidance on doing so in the form of a...

Guide to De-Identifying Data

Researchers who plan to publish data on human subjects should take careful steps to protect the confidentiality of study participants through data de...

Guide: Writing Randomization Code in Stata

This step-by-step guide uses data and an annotated Stata do-file to illustrate how a simple randomization can be carried out using Stata.

Using administrative data for randomized evaluations

This document provides practical guidance on how to obtain and use nonpublic administrative data for a randomized evaluation.

Real-world challenges to randomization and their solutions

This tool is primarily intended for policymakers and practitioners who have a general understanding of randomized evaluations and want to learn how to address six common challenges.

Brief: What is the Risk of an Underpowered Evaluation?

Randomized evaluations can provide credible, transparent, and easy-to-explain evidence of a program’s impact. However, a randomized evaluation requires sufficient statistical power to yield meaningful results. Budgetary, program, and timing constraints may create pressure to conduct an “underpowered...

Common Questions and Concerns about RCTs

Brief: Impact Evaluation Methods

Introduction to Evaluations

Why randomize

The Lessons of Administrative Data

Data gathered by government agencies, school systems, police departments, and hospitals for purposes other than evaluation has powered some of the most important and policy-relevant research produced over the last several years. Compared to new data collected through surveys, administrative data can...