Publications & Resources
Using Multivariate Matched Sampling That Incorporates the Propensity Score to Establish a Comparison Group
Denise D. Quigley
Policymakers and educators aim to increase the competitive eligibility of high school students applying to the University of California through the School/University Partnership Program and academic development student programs. Along with this challenge, policymakers and educators want to know whether and to what extent the investment in these programs is making a difference in the choices and lives of students. The University of California Office of the President (UCOP) has funded, as part of its overall evaluation, a research project to identify program effects of outreach work on student achievement using a comparison group design and statistical controls. This report summarizes the literature on matching methods and outlines the strategy used for establishing comparison groups. This methodology is the first step in the evaluation analysis strategy used to measure the systemwide effect(s) of School/University Partnerships and academic development student programs on individual student-level achievement and student-level eligibility.
Quigley, D. D. (2003). Using multivariate matched sampling that incorporates the propensity score to establish a comparison group (CSE Report 596). Los Angeles: University of California, Los Angeles, National Center for Research on Evaluation, Standards, and Student Testing (CRESST).