9.2 Sample

An operational definition of migrants in the context of the project was prepared in collaboration with investigators and project leaders. The definition was designed to specifically include non-EC migrants and refugees.

A mix of sampling techniques were used in order to meet the data and information realities of the contexts in which the projects were being undertaken. The obvious shortcoming of this included potential lack of data comparability, and potential bias in any non-randomised samples. It was felt that these shortcomings would nevertheless be acceptable in light of the pilot nature of the project, and the fact that one of the purposes of the pilot project was to explore optional sampling techniques for any future use. The decision was also predicated on the fact that in some locations it was simply not possible to use migrant records for sampling because these do not exist, while in other situations accessing migrants for interviews had to be done using convenience samples and street techniques.

Random sampling of migrants using municipal statistics was done in Arnhem, while in Roubaix, Antwerp, Arhus and Rome it was more appropriate to recruit respondents using street samples. In general, migrant groups who are thought to be potentially at risk for drug abuse were contacted through street corner workers. No new data collection took place in Athens but an assessment of previously gathered data was carried out. In the case of Barcelona no new data collection was undertaken either but will be at a later stage.

Initial findings indicate that the cities included in the project have a broad range of "migrants" and the more significantly represented ones are referred to below. In Greece, the project involved Albanians; In Roubaix, Arnhem, and Antwerp second and third generation North Africans; in Arhus refugees from Somalia, the Middle East; and in Rome African or Asian refugees and migrants.

The cultural diversity of the migrants covered by the study has given the project an added value component which had not been anticipated, but which now provides a strong opportunity for inter-cultural comparison. The fact that a range of countries and backgrounds are reflected in the samples is also an opportunity for different health care responses to be considered in light of the contexts from which people have originated. It also provides an insight into the relative impact of uprooting, migration and resettlement on people from different backgrounds and vulnerability.