EWCS 2015 – Weighting

The target number of interviews was 1,000 per country – except in Poland (1,200), Spain (1,300), Italy (1,400), France (1,500), UK (1,600), Germany (2,000) and Turkey (2,000). The governments of Belgium, Slovenia and Spain made use of the possibility offered by Eurofound to pay for an addition to the initial sample size, which resulted in sample sizes of 2,500, 1,600 and 3,300 respectively in these countries. 

As in previous surveys, three types of weights needed to be applied to ensure that results based on the data from the sixth survey could be considered representative for workers in Europe.

Design weights: Because of the way the sampling process is designed, people in households with fewer workers have a greater chance of being selected into the sample than people in households with more workers. For example, in a household with one person in employment, the probability of this person being selected is 100%, whereas it drops to 50% for persons in a household with two people in employment. Selection probability weights (or design weights) are constructed to correct for this. Also, they correct for the differences in sizes of primary sampling units (PSUs) and the share of people in employment.

Post-stratification weights: Because of differences in individual’s willingness and availability to participate in the survey, certain groups are overrepresented – and others underrepresented – in the EWCS sample. To ensure that the results accurately reflect the population of workers in each country, post-stratification weighting is needed. The design weights are calibrated by comparing the EWCS with Eurostat’s Labour Force Survey with regard to respondents’ gender, age, region, occupation and sector of economic activity.

Supranational weights: The differences between countries in the size of their workforce are not (fully) reflected in the sample size in each country. To ensure that larger countries weigh heavier in the EU-level results, supranational weights need to be applied when performing analyses on the European level.

Achieved sample sizes by country:

For further information please see the Weighting report.