Usability of the American Community Survey Estimates of the Group Quarters Population for Substate Geographies

• Activity Number: 155
• Type: Topic Contributed
• Date/Time: Monday, August 3, 2009 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Michael Beaghen*+ and Sharon Stern
• Companies: U.S. Census Bureau and U.S. Census Bureau
• Keywords: sample design ; estimation ; ACS ; GQ

Abstract

The American Community Survey (ACS) provides estimates of detailed characteristics of persons both in housing units (HU) and group quarters (GQ). However, the GQ sample frame, sample design and estimation methodology differ from those of the HU. Importantly, the sample and estimation methodology for the GQ population are designed to be optimal for state level estimates, whereas those of the HU population are designed for smaller geographies. These differences present challenges for the usability of estimates of GQ and total population at substate geographic levels, particularly for one-year estimates. For example, for some substate geographies there may be anomalies in the estimates of poverty. This paper describes these challenges and reviews proposed and implemented methodological changes to enhance the usability of GQ and total population estimates for smaller geographies.