Journal of Official Statistics, Vol.18, No.2, 2002. pp. 217–231
A Design-based Analysis Procedure for Two-treatment Experiments Embedded in Sample Surveys. An Application in the Dutch Labor Force Survey
Jan A. van den Brakel and C.A.M. van Berkel
Abstract:Recently in the Dutch Labor Force Survey (LFS), the questionnaire had to be changed. Before the new questionnaire was implemented as standard, its effects on the outcomes of the LFS as compared with the effects of the regular one were investigated by means of an experiment embedded in the regular LFS. To test hypotheses about possible effects on estimates of finite population parameters estimated by the sample survey a two-treatment randomized block design was analyzed using a design-based analysis procedure. The results from this analysis are compared with the results obtained with a model-based analysis.
Keywords:Completely randomized designs; complex sampling designs; embedded field experiments; questionnaire effects; randomized block designs.
Copyright © Statistics Sweden 1996-2018. Open AccessISSN 0282-423XCreated and Maintained by OKS Group