Academic journal article Cityscape

Use of Genetic Matching in Program Evaluation: The Case of RAD

Academic journal article Cityscape

Use of Genetic Matching in Program Evaluation: The Case of RAD

Article excerpt

(ProQuest: ... denotes formulae omitted.)

In this article, we describe the use of genetic matching in program evaluation, define cases in which this approach would be appropriate, and detail the value that this approach can provide. In particular, we focus on how the researchers used genetic matching in the ongoing evaluation of the Rental Assistance Demonstration (RAD) program, the results they obtained, and how they assessed its success. Clinical researchers and social scientists have developed genetic matching as a sampling technique for conducting nonrandomized observational studies in a quasi-experimental fashion. The method matches each member of the treatment group with one or more members of the control group. The match uses a set of key covariates, which the analyst selects based on prior expectations about possible treatment group participation factors. In the RAD evaluation, the research staff used stratified random sampling to select the RAD project sample (treatment group) from the participating RAD population. For the non-RAD sample (control group), researchers used a genetic matching algorithm to select a matched group of non-RAD public housing projects from the nonparticipating public housing population. Postsampling analysis confirmed that, on covariates likely to impact participation in RAD, the control group and the treatment group were similarly distributed. This matching technique can be a useful tool in program evaluation when membership in the treatment or control group is not random; for instance, if participation is voluntary, as is the case in the RAD program.

Overview of RAD

RAD was authorized in 20121 as a pilot program for converting public housing projects that are subsidized through public housing programs to assisted housing projects that are subsidized through project-based Section 8 Housing Assistance Payment (HAP) contracts. Participation in the program is voluntary for public housing authorities (PHAs). For a PHA to participate, RAD requires that it submit a project application with supporting documentation and analysis. Over a period of several months, the U.S. Department of Housing and Urban Development (HUD) reviews and approves the RAD application, grants a Comprehensive Housing Assistance Plan (CHAP), and issues a RAD Conditional Commitment (RCC). During this process, a project can be withdrawn by the PHA or have its CHAP revoked by HUD. At the end of this approval process, the PHA and HUD agree to convert the project to a project-based Section 8 HAP contract. After conversion, the former public housing project will receive its program funding from the project-based Section 8 program instead of from public housing programs. The primary intent of the RAD program is to preserve and improve the quality of subsidized housing by enabling PHAs to use their long-term Section 8 HAP contracts to leverage external capital for rehabilitation or new construction and financial stabilization.

Congress requires HUD to assess how the RAD program has been implemented and its impact on the physical and financial condition of converted housing and tenants. The core research questions revolve around whether RAD has produced better-quality housing and put that housing on a firmer financial foundation while continuing to serve low-income tenants. The evaluation began in 2014 and will continue through 2018. An interim report on the evaluation was released in September 2016.

Genetic Matching in Observational Studies

Few program evaluations can replicate the research design used in typical clinical experiments to test the efficacy of drugs and other medical treatments. In such experiments, the treatment group is administered the test drug, while the control group is given a placebo. Such studies are double blind in the sense that the assignment of each participant to the treatment or control group is random, and neither the research scientists nor the subjects of the experiment know to which group each subject has been assigned. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.