|
Activity Number:
|
153
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #308449 |
|
Title:
|
Improving the Unit Nonresponse Adjustment in the NLSCY Using Logistic Regression Modeling and Calibration
|
|
Author(s):
|
Mike Tam*+ and Marcelle Tremblay and Sarah Franklin and Claude Girard
|
|
Companies:
|
Statistics Canada and Statistics Canada and Statistics Canada and Statistics Canada
|
|
Address:
|
Main Building 2500 K, Ottawa, ON, K1A 0T6, Canada
|
|
Keywords:
|
Nonresponse ; Logistic Regression ; Calibration
|
|
Abstract:
|
The National Longitudinal Survey of Children and Youth (NLSCY) is a longitudinal survey conducted by Statistics Canada. For the first five cycles, unit nonresponse for the original cohort of 0--11 year-olds was treated by reweighting within RHGs formed via segmentation modeling. In an effort to improve the longitudinal consistency of the estimates, the use of logistic regression to form the RHGs was investigated and implemented at cycle 6 of the NLSCY. Since the model that was used in production included only frame and cooperation variables, further development of the logistic regression model for cycle 7 of the NLSCY is planned. The potential improvements, including incorporating more questionnaire variables from all cycles and the use of paradata, are discussed. We also present the results of using calibration as another way to achieve longitudinal consistency.
|
