|
Activity Number:
|
152
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Government Statistics
|
| Abstract - #304021 |
|
Title:
|
Using Linked Survey and Administrative Data to Build Imputational Models to Adjust Survey Estimates of Medicaid Coverage
|
|
Author(s):
|
Michael Davern*+ and Jacob A. Klerman and Jeanette K. Ziegenfuss
|
|
Companies:
|
University of Minnesota and Abt Associates Inc. and Mayo Clinic
|
|
Address:
|
2221 University Ave. SE Suite 345, Minneapolis, MN, 55414,
|
|
Keywords:
|
Data Linkage ; Medicaid ; MSIS ; CPS ; NHIS ; Imputation
|
|
Abstract:
|
Linking survey data to program administrative data is a powerful approach for building imputational models that can be used to adjust fallible survey data on program participation to produce timely policy research. Currently, data linkage is often done after the survey data have been disseminated. Also linked data cannot be widely disseminated (if at all) because the agreements that facilitated the linkage forbid it. Our research objective is to leverage older vintage data linkage activities to improve policy analysis of Medicaid using the most recent public use survey data currently available. We develop imputational regression models that use only public use data elements as predictors of administrative Medicaid enrollment in a linked data set to partially correct Current Population Survey and National Health Interview Survey estimates of Medicaid enrollment and uninsurance.
|
