Tackling Biased Beliefs to Foster Women’s Entry into Economics: Experimental Evidence from Mexico

Last registered on October 27, 2025
View Trial History

Pre-Trial

Trial Information

General Information

Title
Tackling Biased Beliefs to Foster Women’s Entry into Economics: Experimental Evidence from Mexico
RCT ID
AEARCTR-0017053
Initial registration date
October 25, 2025

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
October 27, 2025, 9:14 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Name
Stanislao Maldonado
Affiliation
Tecnologico de Monterrey
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Elvira Naranjo
PI Affiliation
Tecnologico de Monterrey
Contact Investigator
PI Name
Joana Chapa
PI Affiliation
Universidad Autónoma de Nuevo León
Contact Investigator

Additional Trial Information

Status
In development
Start date
2026-02-02
End date
2026-12-31
Keywords
Behavior, Education, Gender
Additional Keywords
Gender gaps; role models; motivated beliefs; economics education; randomized controlled trial; Mexico.
JEL code(s)
C93, I23, J16, D83
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Economics remains one of the most gender-imbalanced academic fields, with women systematically underrepresented among students and professionals. In Mexico, as in many other countries, female students make up less than one-third of economics majors, reflecting persistent stereotypes about the discipline’s technical nature and limited social relevance. Recent behavioral economics research suggests that such disparities are not solely the result of structural barriers but also stem from motivated beliefs—psychologically grounded misperceptions about one’s own abilities and the character of the field. Women often underestimate their quantitative competence and perceive economics as disconnected from social impact, even when evidence contradicts these views.

This project evaluates the causal impact of an in-person female role model intervention designed to correct these misperceptions and foster greater interest in economics among high-school students in Nuevo León, Mexico. The intervention consists of interactive classroom sessions in which trained female economists share their professional experiences, highlight the social dimensions of economics, and address common stereotypes related to gender and quantitative ability.

The study leverages the networks of the Universidad Autónoma de Nuevo León (UANL) and the Tecnológico de Monterrey, encompassing 31 affiliated high schools and over 45,000 students. Within each participating school, six classroom groups enrolled in the Proyecto de Vida course are selected, and treatment is randomized at the group level (three treatment, three control). This within-school randomization design allows identification of causal effects while controlling for institutional and contextual heterogeneity.

By integrating psychological insights about belief formation with field-experimental methods, the study contributes to understanding how exposure to relatable female role models can reshape gendered beliefs about economics. Outcomes include students’ perceptions of the field, self-assessed quantitative abilities, and stated intentions to pursue economics or related majors. The findings will shed light on the mechanisms—belief updating, identity alignment, and stereotype reduction—through which early interventions can reduce gender disparities in economics education and inspire broader participation in the discipline.
External Link(s)

Registration Citation

Citation
Chapa, Joana, Stanislao Maldonado and Elvira Naranjo. 2025. "Tackling Biased Beliefs to Foster Women’s Entry into Economics: Experimental Evidence from Mexico." AEA RCT Registry. October 27. https://doi.org/10.1257/rct.17053-1.0
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
We propose an in-person role model intervention designed to address women's misperceptions and beliefs regarding economics and their quantitative abilities. By engaging students at a critical stage in their educational trajectory—just before university application decisions—the intervention aims to reshape beliefs about economics as a field and strengthen students' confidence in their own quantitative skills.

The intervention consists of a single, interactive classroom session led by trained female economists. During these in-person visits, role models will deliver a comprehensive 45–60 minute session that combines the sharing of personal narratives, the projection of informative videos, guided reflection exercises, and structured information delivery. Each session will take place during the Proyecto de Vida course, which focuses on students' personal development and career planning. This format ensures high participation rates and curricular relevance, while allowing discussions to unfold in a psychologically safe and familiar classroom environment.
Intervention Start Date
2026-02-02
Intervention End Date
2026-05-01

Primary Outcomes

Primary Outcomes (end points)
Primary Outcomes (End Points)

The primary outcome is the proportion of students who apply to an undergraduate economics program during the 2026 admissions cycle (April–June 2026).
This outcome is measured using administrative records from the participating universities (Universidad Autónoma de Nuevo León and Tecnológico de Monterrey), complemented where necessary with application-system logs (file creation and submission flags).

This behavioral endpoint directly captures the intervention’s main objective: whether exposure to female role models increases the likelihood that female students apply to study economics.
It will be analyzed as a binary indicator (1 = applied to economics program; 0 = did not apply) using a difference-in-differences framework comparing treated and control schools across cohorts
Primary Outcomes (explanation)
The primary outcome—whether a student applies to an undergraduate economics program—will be constructed from administrative data provided by the Universidad Autónoma de Nuevo León (UANL) and the Tecnológico de Monterrey (Tec).

Each student in the final-year cohort (2026) will be matched to the universities’ application records. The variable will take the value 1 if the student submits an application to an economics program during the 2026 admissions cycle (April–June 2026), and 0 otherwise.

Where available, application logs (e.g., file creation, submission status) will be used to validate participation and timing. The analysis will aggregate these binary indicators at the school and classroom levels as needed for cluster-based inference.

To strengthen causal identification, the analysis will use a difference-in-differences estimator, comparing the change in the proportion of applicants between 2025 (pre-intervention) and 2026 (post-intervention) cohorts across treatment and control schools.

This outcome is not an index; it represents a behavioral measure of revealed academic choice, reflecting the intervention’s core goal of increasing women’s participation in economics

Secondary Outcomes

Secondary Outcomes (end points)
Secondary outcomes capture the belief- and identity-based mechanisms through which exposure to female role models may influence students’ aspirations and application decisions. These outcomes are measured using standardized survey instruments administered at three points in time:
(i) baseline (pre-visit),
(ii) short-term endline (immediately after the classroom session), and
(iii) long-term endline (8–10 weeks later).

Each domain is summarized as a standardized z-score index to improve statistical power and reduce multiple-testing concerns. The pre-specified secondary outcome domains are:

Intentions and Aspirations – students’ academic intentions, likelihood of applying to economics, perceived family support, and perceived affordability of studying economics.

Beliefs about Economics – understanding of what economists do, perceived social relevance of the field, and rejection of the belief that economics is only about money or finance.

Identity Alignment and Self-Concept – perceived fit between self and the discipline of economics, identification with economists, comfort with studying economics, and self-assessed analytical ability relative to peers.

Gender Stereotypes and Perceived Barriers – endorsement of gender stereotypes (“men are naturally better at economics”), perceived discrimination in economics-related careers, and ability to name female economists.

All indices are computed so that higher scores indicate more inclusive or pro-equality attitudes, greater confidence, or stronger alignment with economics as a field
Secondary Outcomes (explanation)
Each secondary outcome domain will be constructed as a standardized index following the inverse-covariance-weighted (ICW) procedure proposed by Anderson (2008). Within each domain, survey items are oriented so that higher values represent more inclusive beliefs, higher confidence, or greater identification with economics. Items are standardized using the control-group baseline mean and standard deviation, then aggregated into indices weighted by the inverse of their covariance matrix to improve precision and account for inter-item correlation.

The resulting indices are:

Intentions and Aspirations Index: combines items on likelihood of applying to economics, preferred field of study, perceived family support, and affordability.

Economics Beliefs Index: combines measures of understanding of what economists do, perceived social relevance of the field, and rejection of the belief that economics is mainly about money or finance (reverse-coded).

Identity and Self-Concept Index: aggregates items on perceived fit of economics with self-concept, identification with economists, comfort level studying economics (reverse-coded), and self-assessed quantitative ability relative to peers.

Gender Stereotypes and Barriers Index: includes items on belief that men are better at economics (reverse-coded), perceived discrimination, perceived barriers to studying economics, and recall of female economists.

All items use 0–10 Likert-type scales, and higher index values indicate more positive or inclusive outcomes.
Missing data below 10% will be imputed within domain means; for larger gaps, available items will be used with missing-indicator flags in robustness checks.
Each index is re-standardized to have mean 0 and standard deviation 1 in the baseline control group to facilitate interpretation across survey waves and domains

Experimental Design

Experimental Design
The study implements a clustered randomized controlled trial (RCT) to evaluate the impact of an in-person female role-model intervention on high-school students’ beliefs, aspirations, and application decisions regarding economics.

The experiment will take place in 29 upper-secondary schools in the Monterrey metropolitan area: 24 affiliated with the Universidad Autónoma de Nuevo León (UANL) and 5 with the Tecnológico de Monterrey (Tec).
Randomization occurs at two levels:

School-level assignment: half of the schools are randomly allocated to the treatment group (receiving the intervention) and half to the control group (business-as-usual).

Classroom-level assignment within treated schools: in each treated school, six Proyecto de Vida classroom groups are selected; three are randomly assigned to receive the intervention and three remain untreated to allow measurement of within-school spillovers.

The intervention is delivered during the Proyecto de Vida course in February–April 2026. Each 45–60 minute session is led by trained female economists and combines personal narratives, interactive discussion, and information about economics as a socially relevant discipline.

Data will be collected at baseline, immediately after the intervention, and approximately two months later.
Primary outcomes relate to actual applications to economics programs, while secondary outcomes measure beliefs, aspirations, and identity alignment.

Analyses will follow the intention-to-treat (ITT) principle, clustering standard errors at the school level, and will use ANCOVA specifications controlling for baseline outcomes
Experimental Design Details
Not available
Randomization Method
Randomization was conducted centrally by computer using a blocked and re-randomized procedure.

At the first stage, schools were randomly assigned to treatment or control using computer-generated random numbers, stratified by university network (UANL vs. Tecnológico de Monterrey).
At the second stage, within each treated school, three of six classroom groups in the Proyecto de Vida course were randomly selected to receive the intervention.

The algorithm followed the Morgan and Rubin (2012) re-randomization approach, repeating the assignment until standardized mean differences across treatment arms in key baseline covariates (gender composition, GPA, and baseline beliefs) were below 0.25 standard deviations.

All randomization was implemented by the research team at the coordinating university, using Stata’s random-number generator (set seed recorded) to ensure reproducibility
Randomization Unit
The experiment uses a two-level randomization design:

School level (primary unit): Schools are the main clusters for treatment assignment. Half of the participating schools are randomly assigned to the treatment condition and half to the control group.

Classroom level (secondary unit within treated schools): In each treated school, six classroom groups from the Proyecto de Vida course are selected, with three classrooms randomly assigned to receive the intervention and three left untreated to allow measurement of within-school spillover effects.

Accordingly, the study’s units of randomization are schools and classroom groups, with clustering of standard errors at the school level during analysis
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
The study includes a total of 29 schools (clusters):

24 schools affiliated with the Universidad Autónoma de Nuevo León (UANL)

5 schools affiliated with the Tecnológico de Monterrey (Tec)

Within each treated school, 6 classroom groups from the Proyecto de Vida course will participate in the study (3 treated and 3 untreated), yielding a total of 174 classroom groups across all schools
Sample size: planned number of observations
The main field experiment will include approximately 5,220 students across 29 schools. This estimate assumes an average of 6 classroom groups per school, each with roughly 30 students (29 × 6 × 30 ≈ 5,220). In addition, a laboratory subsample of around 200 students (10 sessions with 20 participants each, balanced by treatment and gender) will participate in incentive-compatible behavioral tasks designed to measure psychological mechanisms such as belief updating, competitiveness, and aspirations. Thus, the total planned number of individual observations across both components is approximately 5,400 students
Sample size (or number of clusters) by treatment arms
School-level assignment:

14 schools assigned to the treatment arm (schools that implement the in-person female role model intervention).

15 schools assigned to the control arm (schools that continue with regular curriculum and receive no intervention during the study period).

Classroom-level assignment within treated schools:
In each treated school, 6 classroom groups in the Proyecto de Vida course are selected. Of these:

3 classrooms per treated school are assigned to receive the in-person intervention session.

3 classrooms per treated school are not treated (to measure within-school spillovers).

This yields approximately 42 treated classrooms and 42 untreated classrooms within treated schools (14 schools × 3 classrooms per status)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
The design is powered to detect small but policy-relevant effects. Belief-, aspiration-, and identity-related outcomes (survey measures; secondary outcomes): These outcomes are analyzed at the student/classroom level using ANCOVA with baseline adjustment and clustering. Under the planned sample (≈29 schools, 6 classrooms per school, ~30 students per classroom), the minimum detectable effect size (MDE) for these standardized indices is in the range of 0.14 to 0.18 standard deviations. This calculation accounts for school and classroom clustering, intra-class correlation at the classroom level (ICC_class ≈ 0.10), potential school-level correlation (ICC_school ≈ 0.03), pre/post correlation of individual outcomes (ρ ≈ 0.6–0.7), α = 0.05, and 80% power. Interpretation: An effect of 0.2 SD is commonly considered “small” in education research, so detecting effects as low as 0.14–0.18 SD means the study is powered to capture relatively small shifts in beliefs, aspirations, or identity alignment that follow from exposure to the female role-model intervention. Application to economics programs (primary outcome; administrative outcome): The main primary outcome is a binary indicator equal to 1 if the student applies to an undergraduate economics program. This outcome will be analyzed using a difference-in-differences specification comparing the 2025 (pre-intervention) and 2026 (post-intervention) graduation cohorts across treated and control schools. This approach increases precision by differencing out baseline school-level differences and exploiting temporal variation. The detectable effect size for this outcome is evaluated in standard deviation (SD) units rather than percentage points because administrative application rates are modeled at the school level and then differenced across cohorts. The study is powered to detect effects in the same approximate range (about 0.15–0.20 SD) once baseline application rates and school fixed effects are controlled for.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number
Analysis Plan

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information