|
Activity Number:
|
501
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #308918 |
|
Title:
|
Parametric and Semiparametric Model-Based Estimates of the Finite Population Mean for Two-Stage Cluster Samples with Item Nonresponse
|
|
Author(s):
|
Roderick J. Little*+ and Ying Yuan
|
|
Companies:
|
University of Michigan and The University of Texas M.D. Anderson Cancer Center
|
|
Address:
|
Department of Biostatistics, School of Public Health, Ann Arbor, MI, 48109,
|
|
Keywords:
|
multiple imputation ; missing data ; survey data ; Bayesian methods
|
|
Abstract:
|
This talk concerns item nonresponse adjustment for two-stage cluster samples. Specifically, we focus on Bayesian multiple imputation for two types of nonignorable nonresponse: nonresponse depending on covariates and underlying cluster characteristics, and depending on covariates and the missing outcome. In these circumstances, standard weighting and imputation adjustments are liable to be biased. To obtain consistent estimates, we extend the standard random-effects model by modeling these two types of missing data mechanism. We also propose semiparametric approaches based on fitting a spline on the propensity score, to weaken assumptions about the relationship between the outcome and covariates. These new methods are compared with existing approaches by simulation. The Behavioral Risk Factor Surveillance System (BRFSS) data are used to illustrate these approaches.
|
