Using Multiple Data Sources To Identify Types and Sources of Coverage Errors on an American Indian Reservation

• Activity Number: 313
• Type: Contributed
• Date/Time: Tuesday, August 5, 2008 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Laurie Schwede*+
• Companies: U.S. Census Bureau
• Keywords: 2006 Census Test ; Coverage ; American Indian reservation ; Sioux ; methodology ; living situations

Abstract

This paper reports results of a study to improve understanding of household coverage on the Cheyenne River Indian Reservation in the 2006 Census Test. Matched data sources from 31 reservation households are used to identify and explore types and sources of coverage error to suggest improvements to enumeration methods, not to estimate coverage rates. Matched data include: 1)the actual Update/Enumerate household roster; 2) the Census Coverage Measurement (CCM) household roster collected 3 months later; 3) the CCM interview audiotape; and 4) an immediate ad-hoc debriefing to resolve any coverage anomalies. We identify types of coverage errors (e.g., omissions, erroneous enumerations); types of persons missed; and sources of error (e.g. mobility, living situations, cultural differences, non-city addresses, questions, recall issues). Suggestions are made to improve reservation enumeration.