Tools of the Trade
In numerous discussions with colleagues I am struck by the varied views and confusion around whether to use sample weights in regression analysis (a confusion that I share at times). A recent working paper by Gary Solon, Steven Haider, and Jeffrey Wooldridge aims at the heart of this topic. It is short and comprehensive, and I recommend it to all practitioners confronted by this question.
On October 3rd, I sent out a survey asking people what was the biggest, most embarrassing, dramatic, funny, or other oops mistake they made in an impact evaluation. Within a few hours, a former manager came into my office to warn me: “Christel, I tried this 10 years ago, and I got exactly two responses.”
In clustered randomized experiments, random assignment occurs at the group level, with multiple units observed within each group. For example, education interventions might be assigned at the school level, with outcomes measured at the student level, or microfinance interventions might be assigned at the savings group level, with outcomes measured for individual clients.
As our impact evaluations broaden to consider more and more possible outcomes of economic interventions (an extreme example being the 334 unique outcome variables considered by Casey et al. in their CDD evaluation) and increasingly investigate the channels of impact through subgroup heterogeneity analysis, the issue of multiple hypothesis testing is gaining increasing prominence.
The primary goal of an impact evaluation study is to estimate the causal effect of a program, policy, or intervention. Randomized assignment of treatment enables the researcher to draw causal inference in a relatively assumption free manner. If randomization is not feasible there are more assumption driven methods, termed quasi-experimental, such as regression discontinuity or propensity score matching. For many of our readers this summary is nothing new. But fortunately in our “community of practice” new statistical tools are developed at a rapid rate.
Attrition is a bugbear for most impact evaluations, and can cause even the best designed experiments to be subject to potential bias. In a new paper, Luc Behaghel, Bruno Crépon, Marc Gurgand and Thomas Le Barbanchon describe a clever new way to deal with this problem using information on the number of attempts it takes to get someone to respond to a survey.
Last week David linked to a virtual discussion involving Dave Giles and Steffen Pischke on the merits or demerits of the Linear Probability Model (LPM).