JSM 2012 Home

JSM 2012 Online Program

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

Online Program Home

Abstract Details

Activity Number: 397
Type: Topic Contributed
Date/Time: Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
Sponsor: Section on Survey Research Methods
Abstract - #305755
Title: Dealing with 'Incompletely Linked' Data in Linked Survey/Administrative Databases: An Empirical Comparison of Alternative Methods
Author(s): Dean Judson*+ and Jennifer D Parker
Companies: National Center for Health Statistics and National Center for Health Statistics
Address: 3311 Toledo Drive, Hyattsville, MD, 20782, United States
Keywords: administrative records ; information integration ; reweighting ; imputation ; record linkage ; incompletely-linked data

The past twenty years has seen a notable increase in the technology for linking data across databases, in particular between survey data and administrative records. However, public concern about the use of linked data, about personal identifiers, etc., has resulted in "incompletely linked" databases. "Incomplete linkage" can be a result of record linkage error or of survey respondents withholding permission to link, and it is known that "incomplete linkage" varies across important subgroups. In theory, these "incompletely linked" databases can be treated as missing data. In practice, the choice of missing data technique is not obvious to the empirical researcher. This paper will explore multiple techniques for handling missing data, in particular reweighting and multiple imputation, in the context of the National Health Interview Survey data linked to Centers for Medicare and Medicaid Services data. The goal will be to examine some typical estimands from the administrative data and determine: 1) Do point estimates derived from the three imputation techniques differ substantively; and 2) What can be said about variance/uncertainty estimation in the context of these techniques?

The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2012 program

2012 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.