|
Activity Number:
|
264
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #307755 |
|
Title:
|
Preservation of Skip Patterns and Covariance Structure through Semiparametric Whole-Questionnaire Imputation
|
|
Author(s):
|
David R. Judkins*+ and Thomas Krenzke and Zizhong Fan and Andrea R. Piesse
|
|
Companies:
|
Westat and Westat and MedImmune, Inc. and Westat
|
|
Address:
|
1650 Research Blvd., Rockville, MD, 20850,
|
|
Keywords:
|
Predictive mean matching ; Unordered polytomous models
|
|
Abstract:
|
Hot-deck imputation schemes are attractive because of how well they preserve complex features of marginal distributions, such as heaping of income reports at round figures, in addition to marginal means and variances. Historically, they have been less successful at preserving multivariate structure. The authors have previously reported on recent methodology for preserving important features of the multivariate structure of an entire questionnaire, including skip rules and the strongest bivariate associations among ordered variables. This methodology is based on predictive mean matching with recursive unsupervised modeling of ordered and binary variables to be imputed. They have recently extended the methodology to include preservation of associations among unordered variables. They will report on this methodology, including a simulation study.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2007 program |
