Citizenship and Education in Twenty-Eight Countries: Civic Knowledge and Engagement at Age Fourteen

Representative samples of about 3,000 students per country in 28 countries

14-year-olds in 28 countries, 1999.

Cross-sectional international assessment: a keyed cognitive civic-knowledge test plus unkeyed attitudinal and behavioural items. Descriptive and correlational throughout - no counterfactual anywhere. Arthur Lupia's book Uninformed argues the CivEd items are poorly designed and produce more noise than signal, a criticism Campbell 2019 takes seriously enough to devote a section to. Widely used as the source for 'open classroom climate' correlations, which is exactly the claim Persson et al. 2020 tested experimentally and could not confirm.

Establishes cross-national descriptive benchmarks for 14-year-olds' civic knowledge and engagement; US students scored highest on the civic SKILLS component (e.g. reading a mock election leaflet, distinguishing fact from opinion). Its correlational finding that open classroom climate predicts civic outcomes has driven twenty years of policy advocacy on the strength of no causal design.

Not assessed in detail; see exclusion note.

The CivEd survey was succeeded by the IEA International Civic and Citizenship Education Study (ICCS) cycles in 2009, 2016 and 2022, which reproduce its descriptive patterns. The United States stopped participating after 1999, so the US data are frozen in the 1990s.