Political Socialization and the High School Civics Curriculum in the United States

Nationally representative 1965 survey of American high-school seniors

US high-school seniors, 1965, from the Youth-Parent Socialization Study.

Cross-sectional analysis of a nationally representative survey, correlating civics-course taking with civic outcomes. No causal design at all: course-taking is self-selected and the outcomes are contemporaneous self-report. Included in the record because it is the single most consequential paper in this literature's history - Campbell 2019 argues it froze research on civic education for a generation.

Civics courses had little to no effect on political knowledge, interest, media consumption, political discourse, efficacy, cynicism, tolerance, or anticipated participation - EXCEPT among African American students, for whom courses did have a positive effect. The subgroup result, largely forgotten for decades, is the origin of the compensation hypothesis that dominates the modern literature.

Not assessed in detail; see exclusion note.

Its overall null was later contested by Niemi and Junn 1998 on NAEP data (civics courses raise knowledge by ~4 percentage points). Its SUBGROUP finding - civics courses mattered for African American students, who had least political socialisation at home - has been repeatedly reproduced in compensation form (Campbell and Niemi 2016; Neundorf, Niemi and Smets 2016; Lindgren et al. 2019).