|
Activity Number:
|
317
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #304489 |
|
Title:
|
Propensity Score Methodology for Nonignorable Nonresponse
|
|
Author(s):
|
Leigh Ann H. Starcevich*+ and Virginia Lesser
|
|
Companies:
|
Oregon State University and Oregon State University
|
|
Address:
|
P.O. Box 1032, Corvallis, OR, 97339,
|
|
Keywords:
|
Missing data ; Nonignorable missingness ; Not-missing-at-random (NMAR) ; Propensity score ; Post-stratification adjustment
|
|
Abstract:
|
When data are not missing at random, approaches to reduce nonresponse bias include subsampling nonresponding units and modeling. Information from nonrespondent subsamples may be used to develop models for bias adjustment. When information from a nonrespondent subsample is available, weighting methods for missing-at-random data may be modified to reduce bias in estimates of population totals. Propensity score methodology for nonignorable missingness is used with the post-stratification adjustment and with the Horvitz-Thompson estimator to account for the dependence between the outcome of interest and the response mechanism. The proposed techniques are applied to a binary outcome subject to nonignorable missingness from a complex survey of elk hunters.
|
