To ensure that the descriptive part of the study was representative, the dataset was weighted. The weighting factor was determined in cooperation with ZUMA (Center for Survey Research and Methodology, Mannheim, Germany). Weighting was done prior to the statistical analyses and was completed in two steps. First, the design was weighted using the number of phone connections and target persons per household. Second, adaptation weighting was done on the basis of the German Microcensus 2004 data according to the variables age, gender and education. Following the weighting of the data, the frequency distributions for age, gender and education corresponded to the Microcensus data for this age group in Baden-Wuerttemberg.