Robustness of the lockdown-civic outcome associations varies by outcome, with legal protest and trust/closure results most consistent¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r3 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Results for legal protest activities with stringency and trust in civic institutions with closure duration were consistent across samples; other coefficients were susceptible to outliers. → Kennedy 2026

Evidence¶

Kennedy 2026¶

Kennedy, A. I., Mejía-Rodríguez, A. M., Strietholt, R. (2026). Civic Education Outcomes After Emergency Lockdowns: International Evidence from the COVID-19 Pandemic. Preprint submitted to Learning and Instruction. https://github.com/OxCGRT/covid-policy-dataset

q2 · i? · associational · r3

Leave-one-out robustness checks removing one country at a time (and Denmark with Germany/NRW together) from the fixed-effects models. The article reports these two relationships "appear to be consistent across different data samples".

"results for the relationships between changes in legal protest activities and average stringency as well as changes in trust in civic institutions and school closure duration appear to be consistent across different data samples."

Kennedy 2026 (2)¶

Kennedy, A. I., Mejía-Rodríguez, A. M., Strietholt, R. (2026). Civic Education Outcomes After Emergency Lockdowns: International Evidence from the COVID-19 Pandemic. Preprint submitted to Learning and Instruction. https://github.com/OxCGRT/covid-policy-dataset

q2 · i? · associational · r3

The same robustness exercise found coefficients for closure duration no longer significant after removing particular countries for several outcomes (e.g., civic knowledge after removing Colombia).

"Less consistency is observed when looking at the results for the relationship between school closure duration and changes in civic outcomes."

Discussion¶

Learner Variables¶

• Civic Mindedness — outcome: instruction changes it

Related Claims¶

• Stricter, longer lockdowns are associated with increases (or smaller declines) in students' expected protest participation, strongest for legal protest — related
• Lockdown stringency and school closure duration are associated with greater declines in student trust in civic institutions — related
• Lockdown measures are positively associated with expected political participation but show null or negative associations with expected electoral participation — related
• Longer school closures are associated with greater declines in grade 8 civic knowledge — related
• Item-level declines in trust under lockdown stringency are driven by local government and parliament, with no significant relationship for trust in political parties — related