Analysis of the Variances of American Community Survey Estimates of the Group Quarters Population

• Activity Number: 20
• Type: Topic Contributed
• Date/Time: Sunday, August 1, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): John Matthew Jordan*+ and Michael Beaghen
• Companies: U.S. Census Bureau and U.S. Census Bureau
• Keywords: Sample design ; controls

Abstract

There are two distinct types of places in which people live, housing units (HU) and group quarters (GQ). People who live in HUs live in houses and apartments, whereas GQ residents live in facilities including but not limited to college dormitories, correctional facilities, and nursing homes. It has been noted that variances are higher for American Community Survey (ACS) estimates of the GQ population of substate geographies than they are for the HU population. The research described in this paper has two goals. The first is descriptive, to determine for which estimates the GQ variances are highest and what the effects of the higher GQ variances are on the estimates of the total resident population (people in GQs and HUs). The second is analytic, to assess the relative contribution of certain previously identified reasons why the GQ population estimates have higher variances.