The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
583
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 1, 2012 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Survey Research Methods
|
Abstract - #305587 |
Title:
|
Multiple Imputation for Hierarchical Data Sets
|
Author(s):
|
Joerg Drechsler*+ and Joe Sakshaug and Matthias Speidel
|
Companies:
|
German Institute for Employment Research (IAB) and German Institute for Employment Research (IAB) and Ludwig-Maxilians-Universität München
|
Address:
|
Regensburger Str. 104 , Nuremberg , , Germany
|
Keywords:
|
fixed effects ;
hierarchical ;
multilevel ;
multiple imputation ;
random effects ;
missing data
|
Abstract:
|
Although multiple imputation is widely accepted as the preferable approach to deal with item nonresponse in surveys, research on imputation involving multilevel databases is still limited. Nevertheless, it has been shown that a naïve imputation ignoring cluster effects leads to biased results for hierarchical data. This is especially true if the aim of the analysis is to estimate intra cluster correlations or other quantities that combine variables from different levels of hierarchy. Imputation strategies that account for the hierarchical data structures are typically based on fixed or random effects models. While the fixed effects approach is easy to implement it can still lead to biased results if the analysis of interest involves a random effects model. On the other hand, random-effects models are more difficult to fit and might be subject to misspecification, especially if the normality assumption of the random effects is violated. In this talk we evaluate different approaches for incorporating the hierarchical structure of the data in the imputation models.
|
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 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.