Annex II: Methodology

• Author: European Union Agency for Fundamental Rights (EU body or agency)
• Pages: 62-67



This annex summarises the implementation and data collection outcomes

of the Fundamental Rights Survey. Adetailed description of the survey

methodology and f‌ieldwork outcomes will be available in atechnical report,

which will be published in the second half of 2020. The technical report will

also describe the pre-test study FRA carried out in 2015–2016 to develop the

survey questionnaire, and the piloting that took place in 2017 to help inform

the design of the f‌inal survey, which was conducted in 2019.

POPULATION AND REPRESENTATIVENESS

People who were 16 years of age and older and had their usual place of

residence in the survey country where eligible to take part in the survey.

The results are representative of this population at the EU-level as well as

for each individual country included in the survey.

GEOGRAPHIC COVERAGE AND SAMPLING

The survey was carried out in the 27 EU Member States, the United Kingdom

(an EU Member State at the time the data collection took place), and North

Macedonia (as the only non-EU country with an observer status to FRA at

the time the survey was designed).

In each country, available sampling frames were assessed before mainstage

data collection. The frames that offered nationwide, close to 100% coverage

of the population were selected for use in the survey, to draw arandom

probability sample of respondents and to contact them. These sample frames

could include sources such as population registers or registers of addresses.

In countries where such sample frames do not exist, or national authorities

did not give access to the frames, the sample was selected in amulti-stage

selection procedure, as commonly used in surveys. The addresses of the

population were enumerated in arandom selection of areas within the

countries (partially including random route methods). This method allowed

for aselection of arepresentative sample of people based on the enumerated

addresses.

In the context of the survey, it was not possible to access existing sample

frames or to use enumeration of addresses in Germany and France. In these

two countries, the data collection used existing online panels.

In all cases, irrespective of the sample frame used, the published results have

been adjusted through weighting so that the key respondent characteristics

of the sample replicate the distribution of key population characteristics in

accordance with off‌icial statistics (see the section on ‘Weighting’ in this annex).

Annex II: Methodology