|
Activity Number:
|
137
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #309175 |
|
Title:
|
Estimate Response Rates and Satisfaction Given Response Jointly Using Hierarchical Bayesian Modeling Approach
|
|
Author(s):
|
Xiaoming Gao*+ and Dongchu Sun and Zhuoqiong (Chong) He
|
|
Companies:
|
Missouri Department of Conservation and University of Missouri-Columbia and University of Missouri-Columbia
|
|
Address:
|
1110 South College Ave, Columbia, MO, 65203,
|
|
Keywords:
|
Bayesian Hierarchical model ; Spatial corrrelation ; nonresponse ; 2-fold CAR model ; conditional probabilities
|
|
Abstract:
|
Human survey is an important tool in wildlife management. Missouri Department of Conservation (MDC) conducts Hunter Attitude Survey periodically to estimate hunters' satisfaction with MDC's deer management program. In an effort to increase response, the survey was sent out three mailings, but the state wide response rate is only 55%. In this study, we're interested in estimating response rate and satisfaction rate given response at county level. We lay out it as a multinomial problem based on conditional probabilities. We propose a Bayesian hierarchical model to characterize the mailing and county effects on the response and satisfaction rates. This model also includes a 2-fold CAR model which allows us to estimate not only the spatial correlations in response rates and satisfaction rates given response among counties, but also a correlation estimate between response and satisfaction.
|
