Bridging Between Two Standards for Collecting Information on Race and Ethnicity: An Application to Census 2000 and Vital Rates

• Activity Number: 277
• Type: Invited
• Date/Time: Tuesday, August 5, 2003 : 2:00 PM to 3:50 PM
• Sponsor: General Methodology
• Author(s): Nathaniel Schenker*+ and Jennifer D. Parker and Deborah D. Ingram and James Weed and Katherine Heck and Jennifer H. Madans
• Companies: National Center for Health Statistics and Centers for Disease Control and Prevention and Center for Disease Control and Prevention and National Center for Health Statistics and University of California, Davis and NCHS
• Keywords: comparability ; imputation ; logistic regression ; missing data ; vital rates

Abstract

In the 2000 decennial census, data on race were collected under the 1997 standards for Federal data collections, which allow multiple-race reporting. To make the 2000 census data compatible with other data systems that have not implemented these standards, the National Center for Health Statistics has bridged the multiple-race data from the census to single-race categories. Building on the approach of Schenker and Parker, bridging models were fitted to data from the National Health Interview Survey, which allows multiple-race reporting but also collects a "primary" race for multiple-race reporters. Under an agreement with the Census Bureau, the fitted models were then applied to the 2000 census to create single-race population estimates (as under the prior standards for race data); the estimates are available at www.cdc.gov/nchs. The bridging methodology will be discussed, and results will be illustrated using (1) death rates for the year 2000 (for which data for the numerators were collected under the prior single-race reporting system, whereas the denominators are counts from the 2000 census), and (2) data for multiple-race mothers in California.