|
Activity Number:
|
153
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #310333 |
|
Title:
|
Testing for Informative Weights and Weights Truncation in Multivariate and Multilevel Modeling with Survey Data
|
|
Author(s):
|
Tihomir Asparouhov*+ and Bengt Muthen
|
|
Companies:
|
Muthen & Muthen and University of California, Los Angeles
|
|
Address:
|
2723 Sherwood Dr, Salt Lake City, UT, 84108,
|
|
Keywords:
|
Weights Ignorability ; Weights Truncation ; Survey data modeling ; Multivariate Models ; Multilevel Models
|
|
Abstract:
|
Analyzing the informativeness of the sampling weights can lead to significant improvement in the precision of model estimation with survey data. Testing for weights ignorability and weights truncation are two techniques that can be utilized in the sampling weights analysis. We evaluate the performance of these techniques in simulation studies based on multivariate factor analysis model as well as multilevel linear and logistic regression models. To test the weights ignorability we use Pfeffermann's (1993) test. We also compare different weights truncation techniques and discuss strategies for determining the optimal level of truncation. We describe the implementation of these techniques in the software package Mplus.
|
