|
Activity Number:
|
439
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #301418 |
|
Title:
|
Empirical Evaluation of Raking Ratio Adjustments for Nonresponse
|
|
Author(s):
|
Ismael Flores Cervantes*+ and J. Michael Brick
|
|
Companies:
|
Westat, Inc. and Westat, Inc.
|
|
Address:
|
1650 Research Blvd., Rockville, MD, 20850,
|
|
Keywords:
|
Raking ; nonresponse adjustment ; weighting
|
|
Abstract:
|
Raking ratio adjustments are used to benchmark sampling weights to known control totals for a variety of reasons. Raking is used to reduce sampling error through the use of auxiliary variables correlated to survey response. In recent years, raking has been used to reduce non-response bias. It also produces weights that have face validity. The use of raking can expedite the creation of analysis weights by simplifying the number of weighting adjustments. This paper is an empirical evaluation of analysis weights created by raking sampling weights without nonresponse adjustments. The raked based weights are compared with analysis weights created by applying sequential weighting adjustments at each stage of nonresponse in a telephone survey (i.e., screener interview nonresponse, extended interview nonresponse). As part of the evaluation of these weights, estimates and their standard errors ar
|
