|
Activity Number:
|
33
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #304309 |
|
Title:
|
Variance Estimation in Complex Surveys with One PSU per Stratum
|
|
Author(s):
|
Harold Mantel*+ and Suzelle Giroux
|
|
Companies:
|
Statistics Canada and Statistics Canada
|
|
Address:
|
SRID, 16th floor, R.H. Coats Building, Ottawa, ON, K1A 0T6, Canada
|
|
Keywords:
|
variance estimation ; sample surveys ; one PSU per stratum ; collapsed strata
|
|
Abstract:
|
In this talk we will discuss variance estimation for surveys where the number of first-stage sample units in one or more strata is one. Unbiased variance estimation in such cases is not possible. We will review relevant ideas that have been discussed in the literature, particularly collapsing of strata. We then propose a new approach based on components of variance from different stages of sampling. The motivation and illustrations come from the Canadian Health Measures Survey, in which the number of PSUs per stratum is very small (in fact, just one for the Atlantic region) due to operational and financial constraints.
|
