Impact evaluation using Stata
A training package to guide researchers who wish to conduct rigorous experimental impact evaluations.
The "interactive" material consists of a PDF manual, with links to a series of audio podcasts, command files (executable with Stata) and datasets.
Download: Impact Evaluation Using Stata
This initiative was undertaken in the context of the PEP's Experimental Research Program - one of the three programs through which PEP resources contribute to building capacities for high quality research in developing countries, using different methods, techniques and approaches. It was prepared by two PEP-affiliated senior researchers (Scientific Mentor, Habiba Djebbari, and Maria Adelaida Lopera Baena).
This document aims to serve as a complement to an interactive textbook, entitled "Impact Evaluation in Practice", published by the World Bank in 2011 to provide an "accessible introduction to the topic of impact evaluation and its practice in development".
The new PEP training package "Impact Evaluation Using Stata" presents a series of examples illustrating the basic analysis required in a rigorous program evaluation report*, and is organized as follows:
- Chapter 1 is a quick introduction to Stata and its programming language
- Chapter 2 illustrates the randomization process and how to compute basic power calculations
- Chapter 3 shows how to estimate simple program effects
The document is also "interactive":
- data sets and Stata exercises are available by clicking on the links
- complementary videos are accessible by clicking on the blue icon
One easy click to access this new PEP training material, free of charge:
- Click here to download the training manual "Impact Evaluation Using Stata" (in PDF)
- For the *dta and *do, click:
* It is important to note that each "evaluation" has particularities that need to be addressed and researchers are expected to further explore the caveats of their own program. The required background is a basic knowledge of the software Stata (http://www.stata.com/) and some familiarity with statistical terminology.