Shadows: xenophobia, disinformation, polarisation

The sharpest turn in the whole CBOS series: in three years sympathy for Ukrainians fell 22 points and antipathy rose 26, exceeding its pre-war level. The lines crossed in 2024 and keep diverging.

A counterpoint, so as not to dramatise: declared antipathy is rising alongside rising real contact, not in a vacuum. More than half of Poles know Ukrainians personally — as neighbours, colleagues, friends — and 27% deal with them daily. This is tension within coexistence, not rejection of an unknown group.

Xenophobia in Poland is not uniform. Towards most nations of Western and Central Europe one in ten Poles or fewer declares antipathy; the spike concerns a few specific groups — Russians (since 2022), Ukrainians and Belarusians (since 2024), and Jews, towards whom antipathy rose 8 points in a single year to the highest level in two decades.

For the third year running attitudes towards the nations that gained after the war broke out — Americans and the English — are also worsening. It is a chapter in which the data allow no comfortable generalisation in either direction.

The strongest single number of the chapter: support down from 94% to 42% in four years. In July 2026, for the first time in CBOS's measurements, opponents of taking in refugees outnumbered supporters.

A nuance from the same survey: attitudes are not uniformly hard. A majority supports limiting benefits, but a majority opposes taking the roof from over the heads of mothers with children. Poland still hosts about 950,000 people under temporary protection — second in the EU after Germany (chapter 16).

The dividing line runs exactly through the middle of the chart: uniformed and local institutions enjoy trust, political and judicial ones do not. Four in five Poles trust the army, one in five trusts political parties; 70% declare distrust of parties, 61% of the government.

This is the measurable face of polarisation: the political dispute dents trust in every institution that takes part in it (government, parliament, courts, the Tribunal, media) and spares those standing outside it. After the 2023 election trust in government and parliament rose, only to return by 2026 to its pre-change level.

A flat line for 24 years says more than any spike: in successive CBOS measurements since 2002 the share who trust others has fluctuated between 17% and 26%, and in 2026 stands at 22%. It is a structural deficit of social capital, attributable to no particular government. The gap to Denmark is 50 percentage points.

This chart explains why Poland does not climb higher in rankings of resilience to disinformation despite good educational results — trust is the weakest component of its score. At the same time Poland is no European exception: Czechia, Italy and France trust little more, and Greece and Romania far less.

Poland is mid-table in Europe for resilience to disinformation: clearly above Hungary, Romania or Bulgaria, but 17 points behind the Scandinavian leaders. The components explain the structure: PISA results pull the index up (63–64 points), interpersonal trust (42) pulls it down.

The exposure channel is plain: three-quarters of Poles learn about current events from social platforms, trust in news fell 8 points in a year — one of the largest drops among the markets surveyed, which the Reuters Institute links to polarisation during the 2025 presidential election — and one in nine pays for news.

The most dramatic curve in the whole material: a fall of 48 places in seven years, then the fastest improvement in this ranking anywhere in Europe. In 2026 Poland is 27th of 180 countries — higher than in any year since 2015.

The contrast with the next chart is instructive: press freedom rebounded spectacularly after the 2023 change of government; the perception of corruption — not at all.

The hardest chart for an optimistic narrative: unlike press freedom, perceived corruption did not rebound after 2023 but settled at the bottom — ten points below the 2015 peak. The CPI measures the perceptions of experts and business, not the number of crimes, but it is precisely that perception which builds a state's credibility.

In the rule-of-law index Poland is improving very slowly (+0.4% in a year) and remains in the bottom quarter of its region — clearly behind Czechia (0.74) and Lithuania (0.77), clearly ahead of Hungary (0.50).

A perfect close to the chapter and a bracket for the whole title: asked without prompting, Poles themselves name quarrelsomeness and polarisation as the second-biggest reason for shame — and the ability to unite in a crisis as the second reason for pride. Three of the five most frequent reasons for shame concern politicians directly.

Polarisation also has its numbers from the ballot box: a record 74.4% turnout in the 2023 election and 71.6% in the second round of the 2025 presidential election, in which the country split almost exactly in half — 50.89% to 49.11%, 1.78 points out of nearly 21 million votes. High mobilisation is here at once the strength of democracy and the measure of division.

The most often shown map of Poland of the last twenty years: the east and south-east vote for the right, the west and north for the centre and the left, and the line runs surprisingly close to the partition borders of a century ago. The 2025 result confirms the pattern, but with exceptions worth seeing: the Łódź region breaks the western belt, Warmia-Masuria and the Opole region — despite lying in the western and northern lands — split almost evenly, and Mazovia as a whole was decided by half a point: Warsaw and its ring vote differently from the rest of the voivodeship.

This chapter shows the map not as a comment on the result but as a measure of polarisation: a 30-point difference between the extreme voivodeships at a record turnout of 72% means a country deeply but evenly divided — and mobilised on both sides.