For each Candidate Countries Eurobarometer new and independent samples were drawn. The basic sampling design in all member states was a multi-stage, random (probability) one. It was developed by GALLUP and uniformly used in each country. For this purpose the local agencies received detailed and uniform instructions for interviewer training.
The sampling was based on a random selection of sampling points after stratification by the distribution of the national, resident population in terms of metropolitan, urban and rural areas, i.e. proportional to the population size (for a total coverage of the country) and to the population density. These primary sampling units (PSU) were selected from each of the administrative regions in every country, at least 100 except for Malta and Cyprus. NUTS 2 regions were applied for the Czech Republic, Hungary, Poland, Romania and Slovakia, NUTS 3 regions for the Baltic countries and Slovenia Estonia, other equivalent regional subdivisions for Bulgaria and Turkey [NUTS - Nomenclature of territorial units for statistics] [classification server] [classification history]. No regional subdivisions are applied for Malta and Cyprus.
In the second stage, a cluster of addresses was selected from each sampled PSU. Addresses were chosen systematically using standard random route procedures, beginning with an initial address selected at random, based on address listings whenever available.
A basic English master questionnaire was developed based on the specifications of and in agreement with the European Commission's Eurobarometer unit, and translated into the national languages. Back-translation control was applied. A second language version was used in countries with significant minorities, in Estonia, Latvia, and Lithuania in Russian.
All interviews were face-to-face in respondents' homes and in the appropriate language.