Sampling Procedure
The metropolitan, urban and rural population and all iadministrative regional unitsi as defined in Official Europe Union Statistics (NUTS 2) covered proportionately the respective population aged 18 and above. The country was divided into an appropriate number of areas, grouping NUTS regions at whatever level appropriately. The NUTS covered in Germany were the following; Arnsberg, Berlin-Ost, Berlin-West, Bremen, Chemnitz, Cottbus, Darmstadt, Detmold, Dreden, Leipzig, D Csseldorf, Frankfurt/Oder, Gera, Suhl, Giessen, Hall, Erfurt, Hamburg, Kassel, Koblenz, K?ln, Magdeburg, Mittelfranken, MCnster, Neubrandenburg, Niederbayern, Nordbaden-Karlsruhe, NordwCrttemberg-Stuttgart, Oberbayern, Oberfranken, Oberpfalz, Potsdam, RB LCneburg, RB Braunsweig, RB Weser-EMS, RB Hannover, Rheinhessen-Pfalz, Rostock, Saarland, Schleswig Holstein, Schwaben, Schwerin, SCdbaden-Freiburg, SCdwCrttemberg-TCbingen, Trier, Unterfranken.
The basic sample design was a multi-stage, random probability sample. 100 sampling points were drawn with probability proportional to population size, for a total coverage of the country. The sampling points were drawn separately in West Germany and East Germany. The sampling points were drawn after stratification by NUTS 2 region and by degree of urbanisation. They represented the whole territory of the country surveyed and are selected proportionally to the distribution of the population in terms of metropolitan, urban and rural areas. In each of the selected sampling points, one address was drawn at random. This starting address forms the first address of a cluster of a maximum of 20 addresses. The remainder of the cluster was selected as every Nth address by standard random route procedure from the initial address. In theory, there is no maximum number of addresses issued per country. Procedures for random household selection and random respondent selection are independent of the intervieweris decision and controlled by the institute responsible. They should be as identical as possible from to country, full functional equivalence being a must. At every address up to 4 recalls are made to attempt to achieve an interview with the selected respondent. There was only one interview per household. The final sample size is 1,123 completed interviews.