For each Central and Eastern Eurobarometer new and independent samples were drawn on a nationally representative basis. Starting with CEEB 2 (1991) a multi-stage random probability sample design has been applied with slight variations for all countries, except Hungary which adopted a quota sampling technique for the second wave. The first round consisted of a set of polls initiated by the European Commission, but carried out relative independently. In this case specific information on sampling and fieldwork is not available.
From the second wave onwards, the sampling was based on a random selection of sampling points from each of the major socio-economic areas in every country, after stratification by the geographical spread of the national, resident population and the distribution in terms of types of locality. Electoral lists and available national census statistics were used for drawing the primary sampling units (PSU), in most cases at least about 100. 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. As an alternative contacts were randomly selected from specially commissioned lists or from a list of the electorate. In the latter case quotas were imposed for the population below the voting age. The second wave in Latvia generally combined the random walk with a quota regarding region, nationality, sex and age groupings.
In Albania, CEEB 7 was carried out distributing the interviews by districts in accordance with their total population share and following a quota sampling with regard to rural vs. urban, sex, age groupings and education. In Slovakia, CEEB 8 and previous waves were carried out with quota for size of community groupings, sex, age groupings and education groupings. Interviews were initially distributed by districts in accordance with their total population share.
All interviews were conducted face-to-face in respondents' homes by fully trained interviewers and in the appropriate language, with a maximum of one interview in any individual household. As of CEEB 5 a back-translation procedure was applied whenever a new question was introduced or a new language required. The English master questionnaire was translated into the the national languages by the local institutes and then back-translated into English by professional translators.
The first round of relative independent surveys was carried out by local academies or universities, as listed in the report. The next three waves (1991-1993) where co-ordinated by GALLUP UK, London. Starting with CEEB 5 (1994) the fieldwork and data integration were coordinated by GfK EUROPE Ad hoc Research together with regional partners.
