Analyzing the Residential Status of Demographic Frame Addresses

Abstract
The Census Bureau’s Demographic Frame is a comprehensive, person-level frame consisting of geographic, demographic, social, and economic characteristics. It could operate as a sampling frame for household surveys, reducing respondent burden by using information already available to the federal government, or improving data quality by drawing from the frame for data editing and imputation. It can be used to identify addresses associated with each person and potentially to identify their residence. Person-address records on the Demographic Frame are derived from administrative, third-party, census and survey data records. A person found on the Demographic Frame may have multiple address records, as they are often associated with several addresses across the various data sources. These multiple address-records create difficulty in placing a person at their correct residential address according to a given reference day. The Demographic Frame has a person-place model process that assigns probabilities to each person-address record and can be used to determine a person’s residence on a particular reference date. These models learn from person-place pairs on existing data within the Census Bureau to make predictions about other person-place pairs derived from administrative records. This analysis will evaluate the accuracy of addresses for people from these models on the Demographic Frame based on a reference date of July 1, 2021. As part of this analysis, we will compare these Demographic Frame addresses to addresses in the 2020 Census and the 2021 American Community Survey frames. This comparison will allow us to examine the Demographic Frame addresses that are not found within these other Census products, which may provide information to help identify whether subsets of these addresses may be more likely to be residential or non-residential addresses. This analysis can potentially improve the quality of the Demographic Frame and its person-place records by increasing the chance of placing a person at their correct residential address.
Type
Publication
ASA Joint Statistical Meetings