Our library of practical resources is intended for researchers and research staff undertaking randomized evaluations, as well as those teaching the technique to others, and anyone interested in how randomized evaluations are conducted.
Incorporating lessons learned through our own experience and through guidance from researchers and research organizations, we provide practical advice for designing, implementing, and communicating about evaluations. These resources are a collaborative effort. We credit the authors of all the resources we post here, and link to their original work wherever possible.
Choose a View:
Introduction to Randomized Evaluations
A non-technical overview and step-by-step introduction for those who are new to randomized evaluations, as well as case studies and other teaching resources.
Before Starting a Project
Tips on successful field management and implementation partnerships for researchers who are new to fieldwork.
Highlights include: annotated checklist for designing an informed consent process, detailed advice on grant proposals and budgeting, and suggestions for proactive measures to help ensure ethical principles are followed in research design and implementation.
Those designing a survey for the first time may find Introduction to measurement and indicators and Survey design useful. We provide best practices for ensuring randomization is stable, verifiable, and replicable, with related Stata commands.
This section contains guidance specific to working with surveyors or survey companies, information about administrative data collection, and information applicable to all modes of data collection, such as on data security, data quality, and grant management.
Processing and Analysis
All the steps in a research project after the data was collected or assembled, from data cleaning to communicating results.
J-PAL North America’s Evaluation Toolkit is intended for researchers, research managers, research assistants, and students trained in economic theory and research design who are preparing to launch a...
Researchers who plan to publish data on human subjects should take careful steps to protect the confidentiality of study participants through data de-identification—a process that reduces the risk of...
This resource outlines steps to establish and build a strong working relationship with an implementing partner at the beginning of a randomized evaluation. Topics include questions to consider when...
Several challenges arise from the length and complexity of randomized evaluations, including the management of multiple data sources or multiple rounds of surveys, sensitive or personally identifiable...