Health Insurance Estimates from the ACS: The Ability to Detect Difference at Small Levels of Geography

• Activity Number: 331
• Type: Topic Contributed
• Date/Time: Tuesday, August 3, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Social Statistics Section
• Author(s): Jeanette Ziegenfuss*+ and Michel Boudreaux
• Companies: Mayo Clinic and University of Minnesota, SHADAC
• Keywords: Health insurance measurement ; Federal surveys ; ACS ; CPS ; Policy analysis

Abstract

Health insurance estimates became available from the ACS in the fall of 2009. There has been substantial interest in the use of the ACS compared to the CPS-ASEC for health policy analysis, primarily because of the ACS's substantial sample size (25-30 times larger than the CPS). This is important for evaluating local and state programs designed to increase health insurance coverage. While previous researchers have described the relative sample size of the ACS compared to the CPS, to our knowledge none have formally examined the statistical power of the ACS to detect change in health insurance coverage status. Here we present results of a power analysis that will determine the size of detectable differences for critical geographic and sub-population segments in the ACS and CPS-ASEC. The policy implications of our results will be discussed in terms of both federal and state needs.