Department of Statistics Seminar
North Carolina State University

presents

Dr. William Bell

U.S. Bureau of the Census

"Small Area Poverty Estimation for U.S. States and Counties"

ABSTRACT

In response to growing demand for small area estimates for public policy purposes, the Census Bureau has developed the Small Area Income and Poverty Estimates (SAIPE) program. The talk will give an overview of the SAIPE program, and then discuss the statistical models used to produce state and county estimates of poverty for school-age children. The estimates are used by the Department of Education in allocations of over $7 billion of funds under Title I of the Elementary and Secondary Education Act, and by Health and Human Services in monitoring the effects of welfare reform legislation. The models can be thought of as empirical Bayes or hierarchical linear models with dependent variables obtained from direct poverty estimates from March Current Population Survey (CPS) data, and predictor variables formed from IRS tax return files, food stamp program data, 1990 census data, and updated Census Bureau population estimates. The models also allow for sampling error in the direct CPS estimates. Some statistical issues arising with these models are discussed, including sampling error modeling, methods for using 1990 census data and previous year's CPS data in the models, and comparisons of Bayesian versus likelihood inference about the variance components.

Friday, February 04, 2000

3:35 - 4:35 pm

206 Cox Hall

Refreshments will be served on the second floor of Dabney Hall (left of Room 222) at 3:00 pm.