|
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 - #309487 |
|
Title:
|
Multiple Matrix Sampling: A Review
|
|
Author(s):
|
Jeffrey Gonzalez*+ and John L. Eltinge
|
|
Companies:
|
Bureau of Labor Statistics and Bureau of Labor Statistics
|
|
Address:
|
2 Massachusetts Avenue NE, Washington, DC, 20212,
|
|
Keywords:
|
Split questionnaire ; Respondent burden ; Nonresponse ; Sample survey ; Selection probability ; Variance estimation
|
|
Abstract:
|
The Consumer Expenditure Quarterly Interview Survey (CEQ) is an ongoing panel survey of US households in which detailed information on an estimated 60 to 70 percent of total expenditures for a consumer unit is collected. The CEQ is generally administered face-to-face and takes about 70 minutes to complete. One proposed method to decrease the length of interview is to use multiple matrix sampling. This involves dividing the questionnaire into sections of questions and then administering these sections to subsamples of the main sample. We provide an overview of the current research on multiple matrix sampling. We review its origins, highlight the fields in which it receives the most application and discuss how it has been applied to problems in surveys. Finally, we discuss the statistical methodology that has been developed for dividing the questionnaire and analyzing the resulting data.
|
