|
Activity Number:
|
455
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 7, 2003 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #301384 |
|
Title:
|
Weight Calibration for Surveys with Nonresponse Follow-Up
|
|
Author(s):
|
Vincent G. Iannacchione*+ and Avinash C. Singh and Jill A. Dever
|
|
Companies:
|
RTI International and RTI International and RTI International
|
|
Address:
|
1615 M St., NW, Washington, DC, 20036-3209,
|
|
Keywords:
|
dual-frame estimation ; small-area estimation ; extreme weights ; nonresponse bias ; mean squre error ; weight calibration
|
|
Abstract:
|
In surveys where response rates are low, a follow-up survey of nonrespondents may be used to reduce the nonresponse bias associated with survey estimates. However, when cost considerations require that the follow-up sample size be small, the reduction in bias obtained from the follow-up may be negated by the increase in sampling variance that results from the increased variability in the sampling weights of the combined sample. In this situation, extreme weights may be trimmed to reduce the mean square error (MSE) associated with key survey estimates. We present a weight calibration method that can reduce the MSEs of multiple survey estimates while preserving the known population totals for key variables. Our method is motivated by small area estimation which balances the variance of an unbiased but unstable design-based estimator with a biased but stable model-based estimator. We show that the ideas underlying dual-frame estimation together with sampling weight calibration can be used to develop composite weights to produce estimates that strike a balance between variance and bias. We also apply the method to a nonresponse follow-up for a survey of Gulf War veterans.
|
