Methodology & transparency
This project studies regional, national and European identity through repeated regional surveys. This page explains how the data are collected, cleaned and reported, so anyone can judge the results for themselves.
The surveys
Each region is studied with an online questionnaire hosted on Unipark/EFS. Surveys run on a yearly, rolling calendar by category (micronations from January through to Spain in August), so every region is re-measured over time.
Questionnaires are offered in the relevant regional and national languages, and cover strength of regional identity, national and European attachment, and preferences on autonomy.
Participants & sampling
Participation is open and voluntary, recruited through regional networks and social media. Samples are therefore not probability samples of the whole population; results describe the people who took part, and the number of respondents (N) is always shown alongside every result.
Cleaning & missing values
Raw exports are decoded from numeric codes to readable labels. Technical missing-value codes (0, −66, −77, −99 — “not shown”, “filtered”, “no answer”, “missing”) are excluded from every chart and calculation.
All percentages are “valid percent”: the denominator is the number of substantive answers, not the total number of respondents. The number excluded as missing is reported for transparency.
Aggregation & privacy
Only aggregates — counts and percentages per answer — are ever published or exported. Individual responses never leave the project. Results are published region by region, and only once a region’s analysis is complete; anything still shown as a preview is clearly labelled as sample data.
Reuse, citation & DOIs
Published datasets are open data under CC BY 4.0: free to reuse with attribution. Each dataset can be downloaded as CSV or JSON and embedded as a live chart. As datasets are archived, a citable DOI is minted via Zenodo.
Limitations
Because samples are self-selected, comparisons between regions and over time should be read as indicative rather than precise population estimates. We publish the full aggregates and sample sizes so readers can weigh the evidence themselves.