Improving the American Community Survey Margins of Error Through Data-Driven Regionalization

• Activity Number: 18
• Type: Topic Contributed
• Date/Time: Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Seth Spielman*+ and David Folch
• Companies: University of Colorado and University of Colorado
• Keywords: American Community Survey ; GIS ; Spatial

Abstract

This talk describes a method to improve the efficiency of small area demographic and economic estimates from the American Community Survey (ACS). Small area estimates from the ACS are terribly imprecise however it is possible to improve the precision of estimates via spatial (or temporal) aggregation. However, geographic aggregation of tracts can lead to significant information loss and a dilution of spatial patterns. The method presented here is a novel spatial optimization algorithm that improves small area estimates by intelligently grouping small areas while minimizing information loss and preserving spatial patterns. The algorithm allows the post-processing of public use data to to a user specified level of precision.