Get Updates via EmailGet Updates Get our RSS Feed
  Follow Mathematica on Twitter  Share/Save/Bookmark
Mathematica Policy Research - Home

At a Glance


U.S. Agency for International Development

Project Time Frame:


Topic Areas:


Service Areas:

Impact Evaluation
Systematic Reviews
Data Collection
Technical Assistance and Capacity Building

School Dropout Prevention Pilot Program

This project is designing, implementing, and rigorously evaluating evidence-based school dropout prevention pilot interventions in four Asian countries—India, Tajikistan, Cambodia, and Timor Leste. During the initial stages of the project, Mathematica worked with Creative Associates and other participating organizations to develop a methodology to review existing U. S. and international evidence on interventions designed to prevent student drop out, based on the What Works Clearinghouse (WWC) model. Mathematica also participated in the review of the studies included in the resulting report. By drawing on the WWC model, Mathematica is also advising Creative Associates on the development of the project website.

In its work with Creative Associates, Mathematica helped to conceptualize an approach to assess dropout trends and issues in the four pilot countries. Mathematica also participated in the design of in-country assessment instruments and protocols to more closely examine dropout in each of the four countries, as well as the design of a data entry system and table shells for the resulting data. The review of the international evidence and the country assessments informed the design of the interventions.

Working with the other partners, Mathematica helped design rigorous impact evaluations of the pilot programs in each of the four countries. Mathematica developed an evaluation design report documenting the design parameters for each country evaluation. Mathematica also developed survey instruments, is overseeing data collection and data entry, analyzing data, and producing baseline reports for the impact evaluations of the interventions in each of the four countries. The data sources include both school administrative and teacher and student survey data. Several rounds of data collection will be conducted during the study.