Month-in-sample effects in the Labour Force Survey / Gopa Ray and Diane Stukel. : CS11-619/2000-5E-PDF

"The Canadian Labour Force Survey (LFS) uses a rotating panel design in which each selected dwelling remains in the sample for six consecutive months, and is then replaced by a new dwelling so that one-sixth of the overall sample is replaced each month. Each panel is designed to be representative of the population. However, studies have shown that the LFS experiences month-in-sample bias, particularly for the first month, i.e., the estimate of a variable based on the panel of dwellings which are in the sample for the first month differs significantly from the overall estimate. Measures of the month-in-sample-bias used in this study are the Bailar and modified Mohi month-in-sample indices, for estimated counts and rates of key LFS variables. The work of Brisebois and Mantel (1996) and Kennedy, Drew and Lorenz (1994) on the month-in-sample bias in the Labour Force Survey have been compared with those obtained in this follow-up, from the perspective of improved nonresponse and poststratification adjustments, using these month-in-sample indices"--Abstract.

Lien permanent pour cette publication :
publications.gc.ca/pub?id=9.841596&sl=1

• Format MARC XML
• Format MARC HTML

Demander des formats alternatifs