Matching Methods for the Evaluation of Section 1115 Demonstrations
U.S. Department of Health and Human Services, Centers for Medicare & Medicaid Services, Center for Medicaid and CHIP Services
States evaluating their section 1115 demonstration often use a comparison group design to estimate demonstration impacts. Ideally, these comparison groups have the same characteristics as the demonstration group, but this is usually not the case in practice. This white paper describes several matching methods states can use to make the comparison group more similar to the group affected by demonstration policies. These methods include propensity score matching and weighting which can be easily implemented using statistical software. It also provides a summary of best practices and guidelines for how states can explain their use of matching methods in evaluation designs and reports.