The basic sample design applied in all states is a multi-stage, random (probability) one. In each country, a
number of sampling points was drawn with probability proportional to population size (for a total
coverage of the country) and to population density.
In order to do so, the sampling points were drawn systematically from each of the "administrative regional
units", after stratification by individual unit and type of area. They thus represent the whole territory of
the countries surveyed according to the EUROSTAT NUTS II 1 (or equivalent) and according to the
distribution of the resident population of the respective nationalities in terms of metropolitan, urban and
rural areas.
In each of the selected sampling points, a starting address was drawn, at random. Further addresses
(every Nth address) were selected by standard "random route" procedures, from the initial address. In each
household, the respondent was drawn, at random (following the "closest birthday rule"). All interviews
were conducted face-to-face in people's homes and in the appropriate national language. As far as the
data capture is concerned, CAPI (Computer Assisted Personal Interview) was used in those countries where
this technique was available.
For each country a comparison between the sample and the universe was carried out. The Universe
description was derived from Eurostat population data or from national statistics offices. For all countries
surveyed, a national weighting procedure, using marginal and intercellular weighting, was carried out
based on this Universe description. In all countries, gender, age, region and size of locality were introduced
in the iteration procedure. For international weighting (i.e. EU averages), TNS opinion & social applies the
official population figures as provided by EUROSTAT or national statistic offices. The total population
figures for input in this post-weighting procedure are listed here.
Readers are reminded that survey results are estimations, the accuracy of which, everything being equal,
rests upon the sample size and upon the observed percentage. With samples of about 1,000 interviews,
the real percentages vary within the following confidence limits:
