Assessing if Mode of Data Collection Impacts the Reporting of Race and Hispanic Origin

David Raglin, U.S. Census Bureau
Deborah H. Griffin, U.S. Census Bureau

It is critical that questions designed to collect data on race and Hispanic origin result in the collection of consistent data. If a respondent is asked to report their race at two different points in time, their response should be the same. This paper examines the issue of consistency in the reporting of race and Hispanic origin by mode of data collection to determine if mode effects consistency of response. This study is based on data from two U. S. Census Bureau surveys--the 2000 Decennial Census and the 2000 American Community Survey. Race and Hispanic origin responses for a sample of persons interviewed in both of these surveys were compared, taking into account the mode of data collection. Reliability measures produced from these data provide evidence that in this application, mode had an effect on consistency in the reporting of race and Hispanic origin.