Methodology
Transparent, honest, comparable
An open, embedded survey is not the same as a representative poll. We don't hide this; we describe it precisely and handle the limits as well as we can.
The same question everywhere
The core questionnaire is identical in every country. Translations are reviewed by native speakers and back-translated to make sure the questions mean the same everywhere.
Weighting
Volunteers differ from the population. We therefore weight results by country, age group, gender and education to Eurostat data, and always publish both raw and weighted figures.
Filtering abuse
- ✦One response per device, bot filtering.
- ✦Automatic flagging of abnormal patterns (mass or too-fast responses).
- ✦We publish the share of filtered responses too.
What this is not
It is not a referendum or an election forecast. It is a huge, open conversation that shows what participating Europeans think, and where they agree.