Online Program Home
My Program

Abstract Details

Activity Number: 242 - Issues in Frame Quality and Accuracy Assessments
Type: Contributed
Date/Time: Monday, July 29, 2019 : 2:00 PM to 3:50 PM
Sponsor: Government Statistics Section
Abstract #304518
Title: Imputation Models Using Computer Matching Results
Author(s): Glenn Reisch*
Companies: United States Census Bureau
Keywords: Census Coverage Measurement ; post-enumeration survey; dual system estimation; matching; Logistic regression ; imputation

The major goal of the 2010 Census Coverage Measurement (CCM) program was to measure coverage for housing units and people in the 2010 Census using dual-system estimation. As part of this process, person interviews were conducted at a sample of housing units collected independently of Census operations. After computer matching of this data to Census results, extensive clerical review and field followup were performed on nonmatches and cases with unresolved statuses. The 2020 Post-Enumeration Survey (PES) has the same goals and approaches as the 2010 CCM, which includes clerical matching and field followup. However, in this research, we investigated using models instead of clerical review or followup operations. Not including clerical review or followup operations would lead to larger numbers of nonmatches and unresolved statuses. We used models and information from the 2010 CCM to impute the match status and unresolved statuses. We then used the final 2010 CCM results to evaluate coverage estimates based on computer matching output.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program