Vignette Equivalence and Response Consistency: The Case of Job Satisfaction
Institut d'Anàlisi Econòmica (IAE-CSIC), Barcelona
B.M.S. Van Praag
University of Amsterdam - Faculty of Economics and Business (FEB); Institute for the Study of Labor (IZA); Tinbergen Institute in Amsterdam; CESifo (Center for Economic Studies and Ifo Institute for Economic Research)
University of Aberdeen - Business School
November 24, 2011
Tinbergen Institute Discussion Paper No. 2011-167/3
We compare reported job satisfaction with vignette evaluations of hypothetical jobs by using a British, Greek and Dutch data set, containing 95 randomly assigned vignettes. In order to test comparability of international data sets recently the method of anchoring vignettes has been introduced by King et al. (2004). This intuitively and attractive idea requires the properties of vignette equivalence and response consistency. In our data set both job satisfaction and vignettes are numerically evaluated on a 0-10-scale. This fact allows us to interpret the evaluations as cardinal satisfaction values and to estimate satisfaction functions for vignettes and for the own job situation. We find that both functions differ significantly: vignette evaluations appear to depend on the own job situation and other individual characteristics. Hence, without correction for those differences in background characteristics, vignette evaluations are not comparable between individuals. Similar conclusions are reached for response consistency.
Number of Pages in PDF File: 33
Keywords: vignettes, vignette equivalence, response consistency, job satisfaction, subjective well-being
JEL Classification: J28, D6, J24, C25working papers series
Date posted: November 25, 2011
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo7 in 0.469 seconds