The Accuracy of a National Generalized Variance Function for Subnational Estimation

• Activity Number: 445
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Philip Lee* and Bonnie Shook-Sa and Marcus Berzofsky and Lynn Langton and Michael Planty
• Companies: RTI International and RTI International and RTI International and Bureau of Justice Statistics and Bureau of Justice Statistics
• Keywords: GVF ; Generalized Variance Functions ; NCVS ; National Crime Victimization Survey ; Taylor Series Linearization ; TSL

Abstract

The National Crime Victimization Survey (NCVS), sponsored by the Bureau of Justice Statistics (BJS) and conducted by the U.S. Census Bureau, is a multi-mode, rotating panel design survey of households that produces nationally-representative criminal victimization estimates for major types of crime in the United States. Two broad methods exist for calculating design-consistent variance estimates for surveys with complex sample designs like the NCVS - Generalized Variance Functions (GVFs) and direct variance estimation techniques such as Taylor Series Linearization (TSL). For the NCVS, GVFs created by the Census Bureau were designed to produce variance estimates at the national level, but their accuracy at the subnational level has not been evaluated. We assess the accuracy of GVF estimates within subnational areas based on geographic identifiers on the NCVS Public Use Files (i.e. Census region, population size, and urbanicity) by comparing them with TSL estimates. Our analysis found that TSL and GVFs do not provide consistent variance estimates within these subnational areas and thus GVFs should not be applied below the national level.