A European achievement in listening
Europe built a shared way to listen to its societies. The European Social Survey (ESS) turns questions about trust, wellbeing, migration, inequality and democratic life into comparable research data across borders. Its achievement is an enduring public research infrastructure: national teams work to common standards, publish documentation and let other people examine the evidence. Europeans can be proud of making social understanding something to build together, rather than leaving each country to interpret its neighbours through anecdotes.
The ESS is a research survey, distinct from an election, a census or a continuously updated opinion poll. It repeatedly samples populations to study attitudes and social conditions. It does not ask every resident, and successive rounds do not ordinarily follow the same individuals. That distinction shapes the conclusion: a difference between rounds can describe a changing population pattern, but cannot by itself show that particular people changed their minds. Start with the question the design can answer.
Who builds the shared evidence?
As UK Research and Innovation explains, the project became a European Research Infrastructure Consortium (ERIC) in 2013 and is hosted at City St George’s, University of London. The consortium’s General Assembly governs its operations, supported by scientific and methods advice, a director, a core scientific team and national coordinators. This is European infrastructure extending beyond EU membership. Neither its legal form under an EU framework nor its location in Britain makes it a Commission opinion-polling department.
The shared method is delivered through national expertise. Sciences Po describes the French team’s responsibilities as questionnaire preparation and translation, sampling arrangements, fieldwork commissioning and monitoring, and harmonised data production. Switzerland’s FORS likewise describes national collection under common methodological guidelines. The scientific asset is therefore distributed: central coordination makes standards comparable, while national partners make those standards work in actual languages, households and administrative settings.
Comparable questions require more than fluent translation. A term for an institution, qualification or household situation can carry different associations across countries. The ESS prospectus describes its translation, review, adjudication, pretesting and documentation approach. The point is to preserve meaning and record decisions, rather than mechanically replace English words. Readers should still inspect the question text and national documentation before treating two variables with similar labels as identical measurements.
Whose experiences are measured?
Probability sampling matters because the people easiest to recruit are not necessarily the people whose circumstances matter most. The University of Bergen’s methods account explains Norway’s probability sample of residents aged 15 and above in private households, irrespective of citizenship. That resident-population scope is especially valuable in a mobile Europe. It is also a reminder to check each study’s coverage: a survey of residents is a different population from citizens eligible to vote or people living in institutions.
The question programme combines repeated core subjects with rotating modules. Bergen documents trust, politics, identity, wellbeing and social background alongside changing areas of deeper investigation. Round 11’s modules include health inequalities and gender; Round 12 returns to personal and social wellbeing and attitudes to immigrants and refugees. Repetition supplies a historical frame, while modules give researchers room to examine a topic more closely. A headline should identify which question and round actually support it.
There is practical democratic value in this architecture. A student can examine whether an apparent national difference remains after comparing similar age groups. A journalist can check whether a claim concerns trust in people or trust in parliament. A researcher can ask how income, health and attitudes relate within countries before invoking a national stereotype. These are examples of useful questions, not findings reported here. The achievement is making such questions answerable with a common evidence base.
How to read a cross-country claim
The ESS Data Portal is the starting point for data and documentation; national partners explain access and registration conditions. Record the exact dataset edition, countries, fieldwork dates, variable wording and missing-value treatment. A round labelled by two years does not mean every interview happened on the same day or that every participating country appears in the first integrated release. Publishing the edition and scope gives a reader a way to reproduce or challenge the claim.
Weights help account for how a sample relates to its population. The ESS weighting guide explains analysis weights and sample-design information, including the analysis weight supplied in integrated files from Round 9 onwards. A serious comparison also needs uncertainty appropriate to the survey design. Do not turn a small numerical difference into a league-table verdict simply because software can sort a column. Check the release-specific guidance before combining countries or comparing subgroups.
A worked editorial test makes the distinction tangible. Suppose a draft says 'Europe trusts parliament less now'. Ask first which countries were measured in both periods and which parliament the question names. Then inspect fieldwork dates, question wording, collection modes, weights, non-response and uncertainty. Compare the matched scope before interpreting the direction. If the evidence covers only selected countries, name those countries. If it measures a single period, remove 'now'. No numerical result is invented by this example.
The 2026–2027 transition
The next methodological test is already under way. UKRI describes Round 12 as a bridge combining face-to-face and self-completion collection, ahead of a planned full move to self-completion in Round 13. Online and paper responses can help sustain coverage as fieldwork changes, but a different mode can affect how people answer. A change in a published attitude measure could therefore mix social change with a measurement change. Treat the bridge as a research task, not a reason to discard the series.
Status checked 7 November 2026: the Finnish ESS team’s 23 September notice moves the first Round 12 release to mid-January 2027, allowing post-stratification weights and mixed-mode guidance to be prepared. This is a scheduled release, not available final results. Meanwhile, Greece’s EKKE reports national Round 12 fieldwork from 4 September to 30 November 2026. Those dates illustrate why a programme year, national fieldwork completion and a quality-controlled public dataset must remain separate milestones.
Evidence for a shared democracy
The limits deserve proportionate attention. Non-response, translation choices, excluded populations and changing modes need documentation; a survey also cannot make an observational association prove causation. But Europe’s response is a shared infrastructure that makes methods visible and improvement possible. Its value resembles the open evidence built by Euro-Argo in another domain: continuity depends on national contributions and common standards, not just a clever instrument or one successful round.
The European Research Council supports investigator-led research; the ESS supplies a reusable measurement resource. Both belong in Proud Europe, because collective capability includes understanding how people live and what they believe. A democratic federal Europe needs the ability to hear differences without inventing unanimity. The ESS gives that public conversation evidence that can be shared, checked and improved across borders.