General purpose multi-country survey project collecting data on push and pull factors of international migration in selected origin and destination countries. Project was funded by the European Commission, executed by Eurostat and implemented by NIDI in collaboration and consultation with country teams in Ghana, Morocco, Senegal, Egypt, Turkey, Italy, Spain. Both households with and without emigrated or returned households members were sampled and interviewed between 1996-1998 in these countries. In origin countries households were sampled. In destination countries individual respondents were sampled.Before using the data, read the methodology section and technical appendix of the Comparative Research Report. Analyze the data using country-specific questionnaires, and consult country-specific research reports about design, implementation, and analytical results regarding push and pull factors of migration. Country reports and the comparative report follow the same structure.Consult the enumerator-instruction manual to familiarize with concepts and -definitions that are specific to this international migration survey project. For each country, there are always two main data-files available: one household-level file and one so-called join-match file. The former file only comprises characteristics of the household, the latter comprises individual-respondent characteristics as well as all characteristics of the household to which a respondent belongs. Data file variable names correspond with questionnaire numbers. Data file variable and value labels correspond with numbers and answer categories in the questionnaires. Code 9,99,999, etc. is missing value, 8,98,998 is don't know, 7, 97,997 is refuse to answer question. SPSS system missing code (.) is used in case a question is not applicable to respondent. Standardized weights are included in files.Additional files have been added comprising specific information about particular country-data files.

The reports and questionnaires pertaining to Morocco and Senegal are in French, all others are in English.Some of the variables are not explained, or not fully explained. This includes the derivative values. The meaning of variables directly related to the surveys should become clear by comparing them to the to the questions in the questionnaires.For more information about this research please contact Drs. George Groenewold via groenewold@nidi.nl.