|
Activity Number:
|
309
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 14, 2002 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Survey Research Methods*
|
| Abstract - #301454 |
|
Title:
|
Redesign of Current Population Survey Raking to Control Totals
|
|
Author(s):
|
Brandon Schneider*+ and Edwin Robison+ and Martha Duff
|
|
Affiliation(s):
|
U.S. Bureau of Labor Statistics and U.S. Bureau of Labor Statistics and U.S. Bureau of Labor Statistics
|
|
Address:
|
2 Massachusetts Ave NE, Washington, District of Columbia, 20212, US 2 Massachusetts Avenue NE, Washington , District of Columbia, 20212-0001,
|
|
Keywords:
|
Current Population Survey ; Raking ; Weighting
|
|
Abstract:
|
Weighting for the monthly Current Population Survey includes an iterative proportional fitting process ('second-stage ratio estimation') that forces CPS estimates to match three sets of population control totals. The procedure, now in place, successively computes ratio adjustments to weights so that estimates are forced to match: 1.) the civilian noninstitutional population of each state; 2.) ethnicity x gender x age controls; and 3.) race x gender x age controls. The monthly control totals are treated as fixed constants and are derived from models that update decennial demographic census totals using administrative data sources. A new process to be implemented in January 2003 retains the three-way iteration but is designed based on a reevaluation of data analysis needs, convergence properties, and survey coverage. Six gender x age controls are used for each state, Asian is added as a new race, and the age categories are harmonized to improve convergence. Non-iterated coverage adjustment steps are added to avoid harmful interactions that were found among the successive ratio adjustments for state, ethnicity, and race.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2002 program |
