|
Activity Number:
|
36
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #306749 |
|
Title:
|
Sample Reweighting To Reflect an Initial Population
|
|
Author(s):
|
Julia Bienias*+ and Phillip S. Kott and Todd L. Beck and Denis A. Evans
|
|
Companies:
|
Rush University Medical Center and National Agricultural Statistics Service and Rush University Medical Center and Rush University Medical Center
|
|
Address:
|
1645 W. Jackson Blvd., Suite 675, Chicago, IL, 60612,
|
|
Keywords:
|
complex surveys ; Poisson sampling ; finite-population inference ; model-based inference ; longitudinal
|
|
Abstract:
|
The Chicago Health and Aging Project is a complex longitudinal community-based study which examines risk factors for chronic health problems of older adults. Every three years we draw a stratified Poisson sample for detailed clinical evaluation. We wish to combine these samples and represent the complete population as it existed at the baseline of the study. We propose creating model-dependent sample weights, treating these later sampled observations as though they were actually obtained at baseline but the data were "delayed." Such an approach has an implicit non-response adjustment, and we discuss the ramifications of this as well as using alternative adjustments for non-participation and mortality when creating our weights. These methods are illustrated with several predictors of interest in aging research and for outcomes such as mortality. Supported by NIH grant AG11101.
|
