This paper describes the common data source and methods used in this study. Data come from the Community Tracking Study Household Survey, a nationally representative survey of individuals conducted in 1996-1997. Focusing on the privately insured, nonelderly population, the study examines the effect of health maintenance organizations (HMOs) on access, service use, and consumer assessments, as well as how these effects differ across population subgroups. Multivariate models control for population characteristics and location differences between HMO and non-HMO enrollees. Tests for endogeneity of plan type (selection bias) indicated that this did not pose a threat to the analysis.