Abstract:
|
Administrative records have become a hot topic in survey research and are increasingly used to replace or supplement survey data. The U.S. Census Bureau is researching ways to use administrative records to improve survey response and reduce respondent burden in the American Community Survey (ACS). Previous research found high match rates for some items when linking administrative records to households in the ACS sample. Year structure built, acreage, property value, and property taxes were found to be the best candidates for further study. This research simulates the use of publically-available administrative records to replace responses to the 2015 ACS for these four survey items. The test file was subject to all ACS edit and allocation procedures and used to create a test set of estimates to compare with the 2015 ACS estimates. Improvements in missing data rates are measured by calculating the number of households that left the item blank but have data from administrative records. The study also assesses the potential reduction in respondent burden by estimating how many ACS sample addresses could use data from administrative records in lieu of being asked the survey question.
|