EWCS 2015 – Methodology

In its European Working Conditions Survey (EWCS), Eurofound aims to collect comparable and reliable data on working conditions across Europe. The methodology is incrementally improved. based on lessons learned as well as to reflect the state-of-the-art in survey methodology.

For more information please see the Technical report.

The sample used in the European Working Conditions Survey (EWCS) is representative of those aged 15 years and over (16 and over in Bulgaria, Norway, Spain and the UK) who are in employment and are resident in the country that is being surveyed. In each of the 35 countries, a multi-stage, stratified clustered sampling design was used.

In each country, primary sampling units (PSUs) were sampled, stratified according to geographic regions (NUTS 2 level or below) and three levels of urbanisation (DEGURBA). Subsequently, in each PSU, households were sampled. In countries where an up-to-date, high-quality address or population register was available, this was used as the sampling frame. If such a register was not available, enumeration was used to generate addresses using the random-walk method. Enumeration was separated from the interviewing stage. Finally, a screening procedure was applied to select the eligible respondent within each household.

For the target and achieved sample sizes per country see Weighting.

For further information please see the Sampling implementation report.

The sixth EWCS includes a limited number of open questions to record respondent’s occupation and the sector of economic activity of the organisation they work for.

• Occupations were coded at the 4-digit level according to the International Standard Classification of Occupations (ISCO) for occupations ISCO 08 and ISCO 88.
• Sectors were coded at the 3-digit according to the Statistical Classification of Economic Activities in the European Community – NACE rev 1.1 and NACE 2.0.

Coding in both versions of the occupations and activity nomenclature facilitates comparisons with previous surveys in the EWCS series.

Information on the level of education or training attained by the respondent is also available in the EWCS. Country­-specific questions regarding education allowed for automatic coding in order to obtain the internationally comparable categories in the International Standard Classification of Education (ISCED) at 1- digit level.

Data on work-related earnings is available.

Where respondents did not provide a precise answer regarding their earnings, they were asked which category on an income scale best matched their earnings. The income scale was built using information from the Structural Earnings Survey (SES) in combination with the results of the fifth EWCS. In case where the information was not available, alternative high-quality local sources were used to make the income. The obtained data were then converted into euro figures. The exchange rates used for the conversion were those valid on the median date of fieldwork for each country. 

For more information, please see the Coding report.

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.

Eurofound places a strong emphasis on quality and aims for precision of estimates for the  European Union as a whole as well as for individual Member States. With its wide geographical coverage and numerous waves, EWCS data enables comparisons across countries and over time. Therefore, quality assurance focuses on consistency in survey implementation and methodology across each wave of the EWCS as well as comparability across countries.

To ensure high quality in the sixth EWCS, Eurofound applies state-of-the-art practices in cross-cultural survey methodology. Each stage of the survey was carefully planned, closely monitored and documented, and specific controls were put in place building on recommendations outlined in the Quality assurance report of the fifth EWCS, the External quality assessment report of the fifth EWCS, as well as Eurofound’s experiences with previous waves of the EWCS and other surveys. Quality control mechanisms were applied to ensure quality in terms of relevance, accuracy, timeliness and punctuality, accessibility and clarity, and coherence and comparability.

At the start of the preparation phase, the contractor and Eurofound established quality assurance indicators which were outlined in an agreed Quality Assurance Plan. Each stage of the EWCS was documented in detail in order to promote transparency and to be able to draw lessons for future reference. The Quality control report, compiled by the contractor, provides a summary of the quality control activities performed before, during and after the fieldwork. The report provides an objective quality assessment of the outputs and processes against the Quality Assurance Framework. It also provides a critical reflection on the quality indicators and targets as identified in the Quality Control Plan.

As a further measure, Eurofound contracted an independent third party to assess the survey quality in relation to comparable surveys and international standards, such as the quality criteria outlined in the European Statistical System, and to provide suggestions for improvement of future waves of the EWCS. The external Survey quality assessment report concludes that the quality of the sixth EWCS is very high, compared to similar surveys and complies with international quality standards.