Replicating the Race-bridging Models Using the Census Quality Survey

• Activity Number: 1
• Type: Invited
• Date/Time: Sunday, August 8, 2004 : 2:00 PM to 3:50 PM
• Sponsor: Social Statistics Section
• Author(s): Deborah D. Ingram*+
• Companies: National Center for Health Statistics
• Keywords: bridging models ; vital rates ; logistic regression

Abstract

The Census Quality Survey to Evaluate Responses to the Census 2000 Question on Race (CQS) is used to evaluate the bridging models developed by NCHS to bridge the Census 2000 multiple-race population to single-race categories (American Indian or Alaska Native, Asian or Pacific Islander, Black, and White). The Census Quality Survey used a split panel design to collect race data using both the 1990 census race question and the Census 2000 race question from about 50,000 households. To assess the robustness of the bridging models obtained using the 1997-2000 NHIS, the models are fit to the CQS data. The resulting parameter estimates are compared to those obtained using the NHIS data. Further, county-age-sex-Hispanic origin-specific probabilities of selecting each of the possible single-race categories are obtained from the CQS parameter estimates and applied to the Census 2000 Summary Data file to obtain bridged-race population counts. These population counts are compared to those obtained using the NHIS bridging proportions to determine how close the agreement is. The impact of using additional Census 2000 contextual variables (i.e. segregation indexes) is examined.