|Thursday, February 21|
|PS1 Poster Session 1 & Opening Mixer||
Thu, Feb 21, 6:30 PM - 8:00 PM
Explaining Small Area Estimation Methodology to Non-Statistician End Users (302569)
Scott Gilkeson, U.S. Census Bureau
Walter Holmes, U.S. Census Bureau
Keywords: small area estimates, data visualization, Census Bureau, SAIPE, SAHIE
The US Census Bureau has two sponsored small area estimates programs: SAIPE and SAHIE. Small Area Income and Poverty Estimates (SAIPE) is produced for the National Center for Education Statistics and used in allocating Title I funds, the largest source of federal education funding. SAIPE uses a Fay-Herriot model-based approach. The Small Area Health Insurance Estimates (SAHIE) program produces estimates of health insurance coverage by demographic and economic characteristics for use by the Centers for Disease Control and Prevention in assessing cancer-screening outreach efforts. SAHIE uses a Bayesian hierarchical approach.
The Census Bureau provides technical descriptions of the methodologies with enough detail that external statisticians can assess model validity. However, many key consumers of the data are not statisticians and need a simpler description.
This e-poster presentation will illustrate data visualization techniques that make the methodologies more understandable and useful to non-statistician audiences. The presentation will include short, animated sequences describing the SAIPE and SAHIE production processes and application of the estimates in one domain as an example.