Eurobarometer Representation Dataset
The Eurobarometer Representation Dataset is a dataset I created with Robert Rohrschneider that harmonizes and merges variables from five decades of Eurobarometer surveys, from Eurobarometer 2 (1974) through Eurobarometer 105.2 (2026) — 113 waves and roughly 2.5 million respondents in a single file. Rather than treating each wave as a stand-alone survey with its own question numbering, the dataset assigns a single, consistent variable name to each substantive question and tracks where that question does (or does not) appear across the full series of surveys, making it possible to study change over time in EU public opinion using one longitudinally consistent file. Substantive coverage includes attitudes toward EU institutions and the European Parliament, support for EU membership and the single currency, national and European identity, political knowledge of the EU, trust in EU and national institutions, economic perceptions during the financial crisis, attitudes toward immigration, and a standard battery of demographic measures. The dataset also carries a standardized composite measure of EU support, built so that respondents from waves that asked different combinations of the underlying questions can be placed on one common scale. The accompanying codebook documents, for each variable, its coding scheme, exact question wording, and Eurobarometer source wave, along with the composite measure and the limits on comparing it over time.
- Codebook (Word) — full documentation of the dataset: how it was built, coverage wave by wave, every variable with its coding and source question, the composite EU-support measure, the two identity scores, the twelve harmonized continuous series, and an honest account of where the data are thin (edition of 15 September 2026)
- Variable Tracker (Excel) — the variable-by-wave crosswalk, showing which Eurobarometer question supplies each variable in each of the 113 waves
Anyone interested in the Stata file should contact me at clarkn@susqu.edu.
