|
Activity Number:
|
70
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #308831 |
|
Title:
|
Hierarchical Bayes Modeling of Survey-Weighted, Small-Area Proportions
|
|
Author(s):
|
Benmei Liu*+ and Graham Kalton and Partha Lahiri
|
|
Companies:
|
Westat and Westat and University of Maryland
|
|
Address:
|
1650 Research Blvd, Rockville, MD, 20850,
|
|
Keywords:
|
weighted proportions ; Hierarchical Bayes modeling; ; beta distribution
|
|
Abstract:
|
When a Hierarchical Bayes area level model is used to produce estimates of proportions of units with a given characteristic for small areas, it is commonly assumed that the survey weighted proportion for each sampled small area has a normal distribution with known sampling variance. However, the assumptions of known sampling variance and normality are problematic when the small area sample size is small or when the true proportion is near 0 or 1. In an effort to overcome these problems, we propose an alternative modeling of the survey weighted proportion based on the beta distribution. We compare the results obtained from this alternative modeling with those obtained from a few commonly used modeling approaches using a Monte Carlo simulation study in which samples are generated from fixed finite population using both equal probability of selection (epsem) and non-epsem sampling designs.
|
