|
Activity Number:
|
341
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #309160 |
|
Title:
|
Evaluation of the Current Weighting Methodology for the Behavioral Risk Factor Surveillance System (BRFSS) and Alternatives for Improvement
|
|
Author(s):
|
Mansour Fahimi*+ and Darryl Creel and Paul S. Levy and Michael Link
|
|
Companies:
|
RTI International and RTI International and RTI International and Centers for Disease Control and Prevention
|
|
Address:
|
6110 Executive Boulevard Suite 902, Rockville, MD, 20852-3907,
|
|
Keywords:
|
RDD ; Poststratification ; Raking ; Variance Inflation.
|
|
Abstract:
|
The 2006 Expert Panel evaluating the BRFSS suggested that the current weighting procedure for BRFSS be improved. Currently, this world's largest random digit dialing telephone survey relies on a simplistic weighting methodology to partially compensate for the bias that results from differential nonresponse and undercoverage. This work examines the current weighting methodology of BRFSS, identifies opportunities for improvements, and introduces an alternative weighting procedure that has greater potential for reducing nonresponse and undercoverage biases without introducing any appreciable inefficiency vis-à-vis variance inflation. Moreover, a general evaluation methodology is presented for comparing the alternative method against the current weighting procedure, as well as other weighting enhancements designed to compensate for exclusion of nontelephone households.
|
