Civic Education at Scale in Mexico

How to cite

Seira, Enrique; Simpser, Alberto. 2025. "Civic Education at Scale in Mexico." AEA RCT Registry. AEARCTR-0015115. https://doi.org/10.1257/rct.15115-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Enrique Seira* , Alberto Simpser

* Primary investigator

Registry ID
AEARCTR-0015115
DOI
https://doi.org/10.1257/rct.15115-1.0
First registered
Last updated
Status
In development
Trial dates
2025-01-06 – 2027-03-28
Countries
Private
JEL codes
I2, O10, P00, Z10
Registry record
http://www.socialscienceregistry.org/trials/15115

Abstract

We propose to conduct a two-year randomized field experiment to evaluate the effectiveness of a curricular civic education intervention on college-age students and the middle school students they mentor in Mexico. Evaluating which policies can support democracy in a period of democratic backsliding is extremely important, but rigorous evidence on the causal effects of civic education on democratic behaviors is scarce. We will implement such a course in Mexico, a large country of great strategic importance to the US, where democratic institutions are being rapidly undermined (Krauze, 2023). Specifically, the project aims to test hypotheses on the following questions: 1. To what extent can civic education strengthen democratic values and behaviors such as attitudes towards democracy, tolerant behaviors, and participation in democratic processes? 2. Can civic education be effectively delivered remotely in a scalable way?

Intervention

The randomized controlled trial will deliver a civic education course on JANN’s platform on a group of about 2,000 college-student tutors and about 6000 to 10,000 of their middle- and primary school students. Tutors and their corresponding students will be randomized into a treatment arm or a placebo control. The college tutors assigned to the treatment arm will take 5 civic education activities that require 20 hours of commitment. Once they master the content, they will adapt it to teach it to the middle- and primary-school students assigned to them. The control arm will work similarly, but the course will focus on artificial intelligence.

Experimental design

The randomized controlled trial will deliver a civic education course on JANN’s platform on a group of about 2,000 college-student tutors and about 6000 to 10,000 of their middle- and primary school students. Tutors and their corresponding students will be randomized into a treatment arm or a placebo control. The college tutors assigned to the treatment arm will take 5 civic education activities that require 20 hours of commitment. Once they master the content, they will adapt it to teach it to the middle- and primary-school students assigned to them. The control arm will work similarly, but the course will focus on artificial intelligence. We will primarily focus on outcomes for tutors who must learn and then teach the course.

Design details

Randomization method
Randomization in Stata, in batches.
Randomization unit
Individual tutors will be randomized first into treatment or control (individual randomization) Then, individual students will be randomly assigned to tutors (individual randomization) Outcomes for tutors are not clustered (outcomes for students, which are secondary, will be clustered at the tutor level)
Clusters
2000 tutors (clusters)
Observations
2000 total tutors (and their 10,000 students)
Arms
1000 tutors to civic education course 1000 tutors to AI course
Primary outcomes
Survey measures: 1. Factual knowledge of course content. 2. Factual knowledge of current political events and policies in Mexico not be taught in the intervention 3. Attitudes towards democracy and authoritarianism measured with classic survey questions 4. Affective polarization measured with classic feeling thermometer questions REVEALED PREFERENCES / INCENTIVIZED BEHAVIORS 5. Experiment on choice of decision method. Within small groups charged with making a decision or choice, a randomly appointed group leader will select whether to make the choice herself or put it to a vote. 6. Experiments on respect for majority decisions. Participants will make a choice about what food to provide for the group. Participants who are in the minority of a collective decision will be given the choice to either impose their personal preferences on the rest of the group or to respect the majority’s choice. 7. Willingness to post a pro-democracy message on social media. 8. Willingness to attend online meeting with out-partisans. 9. Conjoint choice experiment on candidate selection, probing democratic vs. authoritarian preferences and tradeoffs, as in Graham and Svolik (2020). 10. Sharing game (Cappelen et al 2005) with in-partisans and out-partisans. 11. Implicit association test. We will implement an implicit association test focusing on attitudes towards democracy vs autocracy. (POTENTIALLY): ADMIN DATA ON VOTING 12. Obtaining or renewing the voter ID registration card 13. Voter turnout in the 2027 midterm federal elections Hypotheses: Effect of civic education intervention: o H1. Factual Knowledge. Treatment will increase knowledge in comparison with the control group. The null hypothesis is no effect. (Measures 1-2) o H2. Attitudes. Treatment will enhance attitudes in support of democracy and tolerance for the outgroup in comparison with the control group. (Measures 3-4) o H3. Democratic behaviors. Tutors in the treatment group will have a larger tendency to support voting and respect voting decisions versus imposing their individual will (measures 5-6) and democratic behaviors (measures 7-13) in comparison with the control group (12 and 13 pending availability of data). Because our hypothesis are directional, we plan to use one-sided t-tests
Secondary outcomes
Secondary outcomes will be a simplified version of the democratic attitudes questions for students and their parents.

Cited by

No citation data available for this registration yet.

Keywords