|
Activity Number:
|
410
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #306793 |
|
Title:
|
Hierarchical Bayes Estimation of Response Rates with Spatial Correlations
|
|
Author(s):
|
Xiaoming Gao*+ and Chong He and Dongchu Sun
|
|
Companies:
|
Missouri Department of Conservation and University of Missouri-Columbia and Virginia Polytechnic Institute and State University/University of Missouri-Columbia
|
|
Address:
|
1110 S. College Ave., Columbia, MO, 65203,
|
|
Keywords:
|
attitude survey ; response rate ; Bayesian hierarchical model ; spatial correlation ; Gibbs sampling
|
|
Abstract:
|
A Bayesian hierarchical logistic linear mixed model is used to estimate response rates at the sub-area level for hunter attitude surveys. The model includes fixed mailing effects, hunter's characteristics, random geographic effects, and spatial correlations between neighboring sub-areas. The computation is done by Gibbs sampling and adaptive rejection sampling techniques. The method is illustrated using data from the 2001 Deer Hunter Attitude Survey in Missouri. The result shows that not only hunter's age and date of hunting permit purchase significantly influence response rates, but there are significant mailing differences and strong spatial correlations of response rates among counties.
|
