|
Activity Number:
|
279
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #305749 |
|
Title:
|
Estimation for Two-Phase Panel Surveys
|
|
Author(s):
|
Jason Legg*+ and Wayne Fuller and Sarah M. Nusser
|
|
Companies:
|
Iowa State University and Iowa State University and Iowa State University
|
|
Address:
|
Department of Statistics, Ames, IA, 50011,
|
|
Keywords:
|
longitudinal ; panel survey ; two-phase sampling
|
|
Abstract:
|
Two-phase panel surveys are conducted to study trends over time. The outcome of these studies is often a dataset containing characteristic values and weights for a set of observations. Replication weights are often included in the dataset to allow variance estimation for nonlinear functions. We propose a general linear model as a basis for estimating means and totals at each survey time point. A consistent replication variance estimator is provided as well as a central limit theorem for use in constructing confidence intervals for functions of means or totals.
|
