Impacts of Treatment-on-the-Treated

Overview

In this module, we learn how we can use two-stage least squares (2SLS) to estimate the local average treatment effect of a program under imperfect compliance.

This module contains one reading, a set of slides for in-class discussion, and an empirical exercise.

Readings

Impact Evaluation in Practice: Chapter 5

Review Questions

1. What is an intent-to-treat estimate, and how does it differ from an estimate of the impact of treatment-on-the-treated?
2. What is a local average treatment effect?
3. What does imperfect compliance mean? Who are the compliers? Other than the compliers, what other groups might we expect to find in our impact evaluation sample?
4. How can we construct an estimate of the impact of treatment-on-the-treated?
5. What is randomized promotion? What three conditions must be satisfied for the randomized promotion approach to provide a valid estimate of program impacts?

Empirical Exercise

Empirical Exercise 8
In this exercise, we’ll be using data from the paper The Miracle of Microfinance? Evidence from a Randomized Evaluation by Abhijit Banerjee, Esther Duflo, Rachel Glennerster, and Cynthia Kinnan. The paper reports the results of one of the first randomized evaluations of a microcredit intervention. We’ll be using a small slice of the data from their paper to explore the use of instrumental variables techniques to estimate impacts of treatment on the treated - and to think about when such methods are appropriate.