In order to ensure that the results are representative, the European Working Conditions Survey data is weighted, in a three-step process:
- First, the data is weighted in order to account for the unequal probability of selection of respondents according to household composition (this compensates for the fact that, as only one member in the household is interviewed, those in one-person households have a higher probability of being selected).
- Second, a non-response weight is constructed, comparing for a few selected variables (gender, age, occupation, sector and region) the distribution of the sample in the survey with the distribution in the European Labour Force Survey (LFS). The results are weighted so that the marginals for these variables coincide with the marginals of the EU LFS.
- Third, a cross-national weight is applied, which assigns to each country an importance in the overall sample which corresponds to its proportion of the overall EU employed population.
Page last updated: 03 January, 2011