|
Activity Number:
|
198
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2003 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #301295 |
|
Title:
|
Accuracy and Coverage Evaluation Revision II Missing Data Methodology and Results
|
|
Author(s):
|
Michael A. Beaghen*+ and Robert D. Sands
|
|
Companies:
|
U.S. Census Bureau and U.S. Census Bureau
|
|
Address:
|
#718, Alexandria, VA, 22307-6877,
|
|
Keywords:
|
evaluation followup ; E sample ; P sample ; noninterview adjustment ; unresolved ; imputation
|
|
Abstract:
|
This paper presents an overview of the Accuracy and Coverage Evaluation Revision II missing data methodology and a summary of its results, with attention given to how it differed from the Accuracy and Coverage Evaluation (ACE) methodology. The ACE Revision II corrected for errors identified in the ACE. A key part of this correction was the recoding of a sample of the ACE data, integrating the results of the evaluation followup. The evaluation followup identified ACE measurement errors in the determination of correct enumeration status in the census sample (E sample) and residency status in the population sample (P sample). The challenge for the ACE Revision II missing data operation was how to deal with the complexities of additional followup information, while making improvements based on what we learned from the ACE experience. We discuss in particular the non-interview adjustment, designed to account for new non-interviews identified by the evaluation followup, and the plan for the assignment of probabilities of correct enumeration and census day residency to those with incomplete or ambiguous interviews.
|
