Methodological aspects of the SAVE data set

This paper describes the general design of the SAVE survey: the design of the questionnaire, inter-viewer and interviewee motivation, and the sampling designs of the various subsamples collected in 2001 and 2003. It discusses the representativeness of the data, explains the construction of weights, and provides probit regressions to analyze potential selectivity problems. The paper finishes by discussing implications for the use of the SAVE data in various estimation procedures.