|
Activity Number:
|
36
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #308561 |
|
Title:
|
Reduction of Sample Areas in the Consumer Price Index and Consumer Expenditure Survey Designs
|
|
Author(s):
|
Lawrence Ernst*+ and William H. Johnson and William E. Larson
|
|
Companies:
|
Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics
|
|
Address:
|
2 Massachusetts Ave NE Rm 1950, Washington, DC, 20212,
|
|
Keywords:
|
PSUs ; maximization of overlap ; PSU weights
|
|
Abstract:
|
The Consumer Price Index (CPI) and the Consumer Expenditure Survey (CE) are surveys with multistage designs, revised every 10 years. The first stage CPI and CE samples include a set of areas (PSUs) selected from the set of U.S Core Based Statistical Areas. CE additionally selects a set of PSUs to represent the rest of the nation. After selecting the original sample of PSUs, a reduction was considered for budgetary reasons, a reduction implemented in CE only. In this paper we describe: the details of the reduction process used and alternative approaches; the adjustment of the PSU weights resulting from the reduction, which was complicated by the use of a maximization of overlap procedure in the original selection of the new sample PSUs; possible improvements to the overlap procedure in the next redesign; and the impact of having different sets of CPI and CE sample PSUs.
|
