|
Activity Number:
|
115
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #309011 |
|
Title:
|
A Comparison of Level of Effort and Benchmarking Approaches for Nonresponse Bias Analysis of an RDD Survey
|
|
Author(s):
|
Daifeng Han*+ and David Cantor
|
|
Companies:
|
Westat and Westat
|
|
Address:
|
1650 Research Blvd, Rockville, MD, 20850,
|
|
Keywords:
|
response rate ; refusal conversion ; nonresponse adjustment
|
|
Abstract:
|
One method for studying nonresponse bias is to analyze estimates by level of effort needed to complete an interview. Level of effort (LOE) is typically measured using the number of times needed to contact a respondent and/or whether the respondent had initially refused to do the interview. LOE analyses are dependent on the assumption that nonrespondents are similar to those who respond with more effort. A second method to assess nonresponse bias is to compare estimates to an external benchmark. In this paper we compare the results of a nonresponse bias analysis for an RDD survey using a LOE approach to one that compares estimates to an external benchmark of a survey with a higher response rate. Important differences are found in the results between the two approaches. One conclusion is that the assumptions behind the level of effort model may not be correct for this survey.
|
