An estimand of interest in empirical studies with observational data is the average treatment effect of a multi-valued treatment in the treated subpopulation. We demonstrate three estimation approaches: outcome regression, inverse probability weighting and inverse probability weighted...

This chapter describes the main impact evaluation methods, both experimental and quasi-experimental, and the statistical model underlying them. Some of the most important methodological advances to have recently been put forward in this field of research are presented. We focus not only on the...

Researchers regularly use synthetic control methods for estimating causal effects when a sub-set of units receive a single persistent treatment, and the rest are unaffected by the change. In many applications, however, units not assigned to treatment are nevertheless impacted by the intervention...

The goal of this paper is to leverage household-level data to improve food-related policies aimed atincreasing the consumption of fruits and vegetables (FVs) among low-income households. Currently, several interventions target areas where residents have limited physical access to grocery stores,...

Proximity augments homeless students' educational outcomes. Homeless K-8 graders whose families are placed in shelters near their schools have 8 percent (2.4 days) better attendance, are a third (18 percentage points) less likely to change schools, and exhibit higher rates of proficiency and...

Using an original administrative dataset in the context of a scarcity induced-natural experi-ment in New York City, I find that families placed in shelters in their neighborhoods of origin remain there considerably longer than those assigned to distant shelters. Locally-placed families also...

The paper presents a modification of the matching and difference-in-differences approach of Heckman et al. (1998) for the staggered treatment adoption design and a Stata tool that implements the approach. This flexible conditional difference-in-differences approach is particularly useful for...