Using Medicaid Participant Data in the Estimation of County Poverty Levels

摘要

The U.S. Census Bureau's Small Area Income and Poverty Estimates (SAIPE) program produces income and poverty estimates using decennial census data, household survey data, administrative records and population estimates. This research focuses on augmenting the SAIPE programs county poverty model with an additional predictor, Medicaid participant data, from the Centers for Medicare and Medicaid Services. Since Medicaid eligibility is means tested based on income and assets, many Medicaid participants may live in families that have income below the federal poverty threshold. Medicaid participant totals might then have utility in estimating county poverty levels. Model diagnostics are presented under both the current SAIPE model and the expanded model for year 2000 to gauge possible model improvement.