|
Activity Number:
|
387
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 11, 2004 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #301209 |
|
Title:
|
An Example of Using Two-phase Sampling Techniques to Integrate Disjoint Surveys and Data Sources into One Survey
|
|
Author(s):
|
Jack R. Lothian*+ and Tracy Tabuchi
|
|
Companies:
|
Statistics Canada and Statistics Canada
|
|
Address:
|
R.H. Coats Bldg. 11-O, Ottawa, ON, K1A 0T6, Canada
|
|
Keywords:
|
two-phase sampling ; survey redesign ; administrative data
|
|
Abstract:
|
In recent years a number of coverage, lack of consistency and data gap issues has led Statistics Canada to consider re-engineering the survey program of motor freight carriers. First, deregulation of the industry has increasingly blurred the lines between what were formerly considered to be two disjoint industries and surveys. Second, electronic reporting of administrative data is now the norm in the Canadian business community. Statistics Canada is increasingly using this data to reduce the burden on respondents, decrease survey costs, and improve the quality of small-area estimation. The re-engineering will integrate the two surveys and incorporate administrative taxation data into the survey. These changes present unique challenges because the original two surveys measured different characteristics in the two populations and the administrative data does not provide sufficient detailed data to estimate all the required characteristics. The paper presents a review of the problems encountered and their solutions including a two-phase sampling procedure that permits estimation of the original characteristics covered by the two surveys.
|
