Methodology EVS 2017
The EVS methodology has been greatly strengthened for the current wave to bring the data quality to a high level. Each national survey conforms to guidelines designed to ensure quality and consistency.
- Respondents are interviewed face-to-face for approximately one hour. However, in the 5th wave of EVS, there is the possibility to carry out a mixed-mode data collection.
- The Countries translated the Master Questionnaire following a team approach. The process was managed, implemented and documented on a web-based platform, the Translation Management Tool (TMT).
- In each Country, the Master Questionnaire was translated in each language spoken by at least 5% of the population.
- Countries sharing a language (e.g., French, German, Russian) cooperated for the harmonization between the same-language translations of the EVS2017 questionnaire. Some countries cooperated in the whole translation process, in other cases one country’s translation was used as a basis for the other countries’ questionnaires.
- The national effective sample size is 1200 [countries with a population over 2 million] and 1000 [countries with a population above 2 million]
- Random samples give full coverage of the target population (persons 18 years and older resident in private households, regardless of nationality or language).
- Information is provided for weighting purposes, including the complete hierarchical structure of the sampling design, the size of the stages/clusters at each step, population distributions, and so on.
- Repeated revisits are made to interview the person randomly selected.
- The questionnaire contains pre-classified variables for education, income, political parties, religion, and region, and open questions for occupation, nationality, and language, which are coded according to pre-classified country-comparable classifications.
- Trial interviews are conducted, and interviewers are thoroughly trained before going into the field.
- Quality control checks are carried out and documented on a sample of respondents, refusals, and non-contacts.