ESENER 2019 Methodology

Universe and unit of analysis

The statistical population comprises all establishments that have five or more employees in the 33 participating countries, covering all sectors of economic activity except for private households (NACE T) and extraterritorial organisations (NACE U). The relevant statistical unit of analysis is the establishment, defined as comprising the activities of a single employer at a single set of premises (e.g. a single branch of a bank, a car factory or a school).

The 33 participating countries comprise all 27 European Member States, as well as Iceland, North Macedonia, Norway, Serbia, Switzerland and the United Kingdom.

Respondents

In each establishment surveyed, the person who knows best about the way safety and health risks are managed at their workplace is interviewed. This is the same target respondent as in ESENER 2014 and part of the effort to ensure comparability between waves and start building time series.

Sampling strategy

In order to ensure that the survey results are cross-nationally comparable, it is essential that the sampling strategy result in the same type of units being surveyed in each country. The quality of the available address registers varies across the participating countries in terms of coverage (especially the sectors of activity included) and in terms of the availability and accuracy of the necessary background information (such as the sector of activity and number of employees). Existing address registers are not comparable cross-nationally, therefore considerable efforts have been made to build samples that provide the necessary quality and ensure cross-national comparability. These efforts have required a screening procedure in those countries where the available address registers provide information at company rather than at establishment level. 

Stratification of the sample is based on a matrix of four size classes and eight groups of sectors. This 32-cell sampling matrix was used in all countries.

ESENER 2019 Sampling matrix

Questionnaire development

The ESENER 2019 questionnaire is largely the same to the one used in the previous wave (ESENER 2014) in order to try to ensure comparability across waves.  The development of the ESENER 2019 questionnaire was done in close cooperation between EU-OSHA, the main contractor and a group of researchers on occupational health and safety from Germany, Latvia and the Netherlands.

The questionnaire has been subject to several tests:

a)     A cognitive pre-test in Germany, Latvia and the Netherlands, with 36 in-depth face-to-face interviews. The main aim was to check the understanding and clarity of the questions, particularly the new or modified ones with respect to ESENER 2014. There were also a number of cognitive questions aimed at testing whether the standardized questions were interpreted in the intended way by respondents from different countries and different types of organisations and whether they were relevant to them. The cognitive test led to a number of modifications to the questionnaire.

b)     A translatability assessment of the English master questionnaire version. Experienced translators from three different language families (Swedish for Germanic languages; French for Latin languages; and Polish for Slavic languages) elaborated rough translations of the new and modified questionnaire items in order to identify any ambiguities or other difficulties for translation. Where such difficulties were identified, translators made proposals for alternative formulations for the master version, which was revised accordingly. Trend items (i.e. items that are unchanged from previous ESENER waves) did not undergo the translatability assessment, given that for most of the items such an assessment was done already in the previous ESENER wave.

c)     A pilot field test was carried out in all 33 countries, with at least 30 interviews per country. This pilot test was done in CATI (and CAWI) and used the same infrastructure as the main survey. The pilot field test resulted in a number of changes to the master questionnaire and to the individual national versions of the questionnaire.

Translation

43 different national versions of the questionnaire have been developed for the countries covered by the survey (access the questionnaires in the table below). For languages shared by two or more countries, different national versions have been developed that take into account work-related practices such as worker representation or labour inspection, which require different questions and terminology. More details on the translation process can be found in the Translation Report –see below.

An optimal translation strategy is essential to ensure that each national version of the questionnaire comprises high-quality questions that can be directed at all types of enterprise within the country concerned and generates information that can be compared cross-nationally. The translation of the ESENER 2019 questionnaire follows the TRAPD approach, an acronym for Translation, Review, Adjudication, Pre-testing and Documentation (Harkness, J. A. (2003). Questionnaire translation. In J. A. Harkness, F. van de Vijver, & P. Ph. Mohler (Eds.), Cross-cultural survey methods (pp. 35-56). Hoboken, NJ: John Wiley & Sons)

The basic steps included in the translation strategy adopted in ESENER 2019 are:

1. Two independent translations into each of the languages carried out by professional translators -native speakers of the target language- who were not in contact with each other while producing their translation.
2. Review of the independent translations by an adjudicator, who drafted a new version.
3. Check of the adjudicated versions by EU-OSHA’s network of Focal Points, with a specific focus on the national OSH terminology, including the questions related to bodies of employee representation, and the new and modified questions.
4. The adjudicated version, with comments from EU-OSHA’s focal points, was then sent to both translators and the adjudicator for an Adjudication Meeting, where both translators and the adjudicator discussed the choices made during the Adjudication phase, and implemented the changes suggested by EU-OSHA’s focal points, when necessary.
5. The updated questionnaire was proofread by another independent linguist (not yet involved in the translation of the ESENER 2019 questionnaire), whose task consisted in identifying minor mistakes and commenting on them.
6. Lastly, the adjudicator was responsible for checking the comments written by the final proofreader, and for implementing their changes, when necessary.

Sample sizes and questionnaire language versions

Documents

Further information for researcher

The full ESENER 2019 dataset is accessible via the UK Data Service of the University of Essex and GESIS.

Please note that all works using or referring to ESENER should acknowledge its source by means of bibliographic citation in footnotes or in the reference section of publications. Similarly, any publication, whether printed, electronic, or broadcast, based wholly or partly on ESENER should acknowledge the European Agency for Safety and Health at Work (EU-OSHA). It should also carry a statement that EU-OSHA bear no responsibility for their further analysis or interpretation.