Where left and right go without saying — and where they don’t
Share of respondents who decline to place themselves on the 0–10 left–right scale (“don't know” or refused), by country, pooled ESS rounds 1–11.
Takeaway: in the most engaged Western publics fewer than 1 in 20 decline — Norway 2.8 percent. In much of the post-communist east it is 1 in 3 or more — Ukraine 40 percent. Pooled, non-placement is 11 percent in the West against 28 in the post-communist east.
lr_scale missing; the item is fielded in every round, so missingness is genuine “don't know”/refusal, not an unasked question. Countries with fewer than 500 respondents are dropped. Weighted by anweight. Source: European Social Survey rounds 1–11, 2002–2023.Who leans left, who leans right — Western Europe, 2002–2023
Mean left–right self-placement (0 = Left, 10 = Right) by group, across ESS rounds. 5 is the exact midpoint. Choose a lens.
anweight. Source: European Social Survey, Western Europe.Brahmin left, Merchant right
Mean left–right self-placement by education × household-income tercile, Western Europe, ESS rounds 5–11 pooled. Blue = left, red = right.
Takeaway: the most left-leaning Europeans are the highly educated on low incomes (4.30); the most right-leaning are the least educated on high incomes (5.09). Read down any column, more education means more left; read across any row, more income means more right.
lr_scale for that education×income combination, pooled over rounds 5–11 (where both items are fielded); cell effective Ns range from about 1,900 to 12,200. Weighted by anweight. Source: European Social Survey, Western Europe.