|
Activity Number:
|
106
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #306451 |
|
Title:
|
Nonresponse Bias Analysis in the 2003 National Survey of Recent College Graduates
|
|
Author(s):
|
Aref Dajani*+ and Jerry J. Maples and Ronald S. Fecso
|
|
Companies:
|
U.S. Census Bureau and U.S. Census Bureau and National Science Foundation
|
|
Address:
|
6101 Sebring Drive, Columbia, MD, 21044,
|
|
Keywords:
|
nonresponse bias ; logistic regression ; reweighting
|
|
Abstract:
|
The National Survey of Recent College Graduates recently experienced an increase in nonresponse rates. A nonresponse bias analysis was indicated to assess any impact on key estimates. Logistic modeling was used to explore possible relationships between early and late responders and variables not currently used in the creation of nonresponse adjustment cells. To analyze estimates by level of effort, late respondents and late ineligibles served as proxies for nonrespondents. This augmented set of nonrespondents was used to reweight the sample. Key estimates were compared. After reweighting the sample, there was no pattern of differences between weighted and reweighted estimates. Further, the readjusted counts had the same distribution as the original adjusted counts. Future analysis should test whether recommended modifications will yield favorable results.
|
