Using Causal Forests To Predict Treatment Heterogeneity: An Application to Summer Jobs

• NCJ Number: 252089
• Author(s): Jonathan M.v. Davis; Sara B. Heller
• Date Published: May 2017
• Length: 5 pages
• Grant Number(s): 2014-IJ-CX-0011
• Agencies: NIJ-Sponsored
• Publication Type: Research (Applied/Empirical); Report (Study/Research); Report (Grant Sponsored); Program/Project Description
• Award: Re-engaging Justice-Involved Youth

Annotation

In order to estimate treatment heterogeneity in two randomized controlled trials of a youth summer jobs program, this study used Wager and Athey's (2015) causal forest algorithm.

Abstract

This article presents a step-by-step explanation for applied researchers regarding how the algorithm predicts treatment effects based on observables. It then explores how useful the predicted heterogeneity is in practice by testing whether youth with larger predicted treatment effects actually respond more in a hold-out sample. The application highlights some limitations of the causal forest, but it also suggests that the method can identify treatment heterogeneity for some outcomes that more standard interaction approaches would have missed. (Publisher abstract modified)