|
Activity Number:
|
252
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #309494 |
|
Title:
|
Variance Estimation for Calibration with Estimated Control Totals
|
|
Author(s):
|
Jill Dever*+ and Richard Valliant
|
|
Companies:
|
University of Maryland and University of Maryland
|
|
Address:
|
Joint Program in Survey Methdology, College Park, MD, 20742,
|
|
Keywords:
|
estimated-control calibration ; survey-estimated control totals ; variance estimation ; simulation
|
|
Abstract:
|
Calibration estimators use auxiliary information to improve the efficiency of survey estimates. The control totals, to which sample weights are calibrated, are assumed to be population values; often, however, the controls are estimated from other surveys. Many researchers apply traditional calibration variance estimators to cases where the control totals are estimated, thus assuming that any additional sampling variance associated with these controls is negligible. The goal of our research is to develop and evaluate variance estimators for two-stage designs with estimated-control (EC) calibration using design-unbiased controls. We compare linearization and replication variance estimators via simulation for point estimators of totals and means. Illustrations are given of the effects on variances of different levels of precision in the estimated controls.
|
