Clustering

Overview

This is the second of three modules on power calculations. In this module, we’ll learn how to determine the appropriate sample size or cluster-randomized trials. Researchers often assign treatment at the cluster level (e.g. at the school, community, or market level) when we are worried about potential spillovers across individuals within a cluster. However, if outcomes within a cluster are correlated (as they usually are), we need to modify our approach to calculating statistical power and sample size.

Readings

Running Randomized Evaluations: Chapter 6

Optional: Using Randomization in Development Economics Research: A Toolkit

Lecture Slides

A handout version of the lecture slides is available here.

Empirical Exercise

The in-class activity is available as a do file or a pdf.

The empirical exercise is available as a pdf.

A web version of the empirical exercise is available here.