|
Activity Number:
|
414
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #309147 |
|
Title:
|
Using Proxy Measures of the Survey Variables in Post-Survey Adjustments in a Transportation Survey
|
|
Author(s):
|
Ting Yan*+ and Trivellore E. Raghunathan
|
|
Companies:
|
University of Michigan and University of Michigan
|
|
Address:
|
Institute for Social Research, Ann Arbor, MI, 48104,
|
|
Keywords:
|
post-survey adjustments ; nonresponse error ; proxy measures of survey variables ; response propensity
|
|
Abstract:
|
Nonresponse bias and the associated sampling error are reduced when the post-survey adjusted estimate employs variables that are related to both the survey variable of interest and propensity to respond. Using the survey conducted by the University of Michigan Transportation Research Institute, this paper will take advantage of the driving history data obtained from the State of Michigan and develop a new weighting scheme drawing on proxy measures of the survey variables, design variables, and process measures that are highly correlated with both response propensity and the survey variable of interest. This paper will then compare this new weighting scheme against a more traditional one employing only demographic variables. Both weighting methods will be assessed on their effectiveness in reducing nonresponse bias.
|
