In a new article in the Journal of Politics (gated, I am afraid) a group of Yale scholars (including Donald Green) finds that randomly exposing students to an enhanced civics curriculum increases the students’ knowledge of the constitution and civil liberties but it does not increase their support for civil liberties. This raises the possibility that knowledge and attitudes about civil liberties are causally disconnected. One caveat is that the study does not compare students with “no knowledge” to students with some education. What the evidence shows is that the average marginal effect of going beyond the standard curriculum is zero if we care about support for civil liberties (if we care about knowledge the marginal effect of more education is positive).
The abstract is below. The authors are: Donald P. Green, Peter M. Aronow, Daniel E. Bergan, Pamela Greene, Celia Paris, and Beth I. Weinberger.
For decades, scholars have argued that education causes greater support for civil liberties by increasing students’ exposure to political knowledge and constitutional norms, such as due process and freedom of expression. Support for this claim comes exclusively from observational evidence, principally from cross-sectional surveys. This paper presents the first large-scale experimental test of this proposition. More than 1000 students in 59 high school classrooms were randomly assigned to an enhanced civics curriculum designed to promote awareness and understanding of constitutional rights and civil liberties. The results show that students in the enhanced curriculum classes displayed significantly more knowledge in this domain than students in conventional civics classes. However, we find no corresponding change in the treatment group’s support for civil liberties, a finding that calls into question the hypothesis that knowledge and attitudes are causally connected.
Update: The authors have made a non-gated version available here.