Breaking the Aspiration Trap

Last registered on August 27, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Breaking the Aspiration Trap
RCT ID
AEARCTR-0019476
Initial registration date
August 21, 2026

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
August 27, 2026, 11:57 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
Yuki Takahashi
Affiliation
University of Osaka
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Saisawat Samutpradit
PI Affiliation
University of Osaka
Contact Investigator
PI Name
Weerachart T. Kilenthong
PI Affiliation
University of the Thai Chamber of Commerce
Contact Investigator
PI Name
Monticha Uraipong
PI Affiliation
Kasetsart University
Contact Investigator

Additional Trial Information

Status
In development
Start date
2026-09-01
End date
2029-12-31
Keywords
Education
Additional Keywords
low-income students, aspirations, role model, mentoring
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Students from low-income households tend to achieve lower levels of education than their counterparts from high-income households. This is not only due to a lack of financial resources; a lack of aspirations may also hold them back from pursuing higher levels of education.

Role models who share students' background can raise aspirations, because the evidence they provide is about people like the student. However, despite their scalability and cost-effectiveness, role models alone may not be enough to translate heightened aspirations into actionable steps. Mentors typically have richer educational backgrounds and can support students in taking those steps, but that same distance from students’ backgrounds may make them weaker evidence of what is attainable for a student from a low-income household. The two functions are therefore filled by different people, and combining them may do what neither does alone.

We study whether raising and sustaining educational aspirations improves educational and behavioral outcomes. We study this in Thailand, where financial support for post-lower-secondary education is widely available to students from low-income households but underused. We conduct a three-arm, school-level experiment with second-year lower-secondary students from low-income households, many of them receiving conditional cash transfers from the Equitable Education Fund, in four northeastern provinces. Schools are randomly assigned to a control group, a role-model webinar, or the webinar plus one school year of online mentoring. The design identifies the effect of the role-model intervention alone and the additional effect of mentoring on top of it.
External Link(s)

Registration Citation

Citation
Kilenthong, Weerachart T. et al. 2026. "Breaking the Aspiration Trap." AEA RCT Registry. August 27. https://doi.org/10.1257/rct.19476-1.0
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)
Intervention Start Date
2026-11-01
Intervention End Date
2027-10-31

Primary Outcomes

Primary Outcomes (end points)
(1) Researcher-administered test scores in Thai language and mathematics; (2) national standardized exam scores (O-NET); (3) post-lower-secondary applications and educational pathways; (4) applications for and receipt of scholarships and student loans.
Primary Outcomes (explanation)
1. Researcher-administered test scores. Students complete a test covering Thai language and mathematics at baseline and endline. Items were selected from past O-NET test questions, with each subject containing 10 questions for a total of 20.
2. O-NET scores: National standardized exam scores at the end of lower secondary school, obtained from administrative records through the Equitable Education Fund.
3. Post-lower-secondary applications and educational pathways. Whether the student applied to continue after lower secondary school, and the number and type of schools applied to, together with the student's realized status after graduating from lower secondary school. We also consider whether the student completed lower secondary education and, if not, the timing of the dropout.
4. Scholarship and student loan applications and approvals. Whether the student applied for scholarships or student loans, the number of applications submitted, and whether the applications were approved.

Secondary Outcomes

Secondary Outcomes (end points)
(1) aspirations, (2) self-efficacy, (3) required support, (4) perceived barriers, (5) interest in upper secondary school tracks, (6) time use, (7) absenteeism, and (8) knowledge about scholarships and student loans, all collected from a researcher-administered survey.
Secondary Outcomes (explanation)
1. Aspirations: ranking of the five educational goals by how much students want to achieve them: completing lower secondary school, vocational upper secondary school, academic upper secondary school, higher vocational school, and university or above.
2. Self-efficacy: how capable they think they are of achieving each of the five educational goals.
3. Required support: how important each of the following is to achieve their educational goals: information about educational options and application processes, financial support and understanding from family, and support from teachers and peers.
4. Perceived barriers: how much each of the same factors hinders them from achieving their educational goals.
5. Interest in upper secondary school tracks: how interested they are in each of the popular academic and vocational upper secondary school tracks.
6. Time use: days per week spent on nine activities covering family work, study, enrichment, screen use, and leisure.
7. Absenteeism: days absent from school in the past month and their reasons.
8. Knowledge about scholarships and student loans: true/false questions about scholarship and student loan requirements and interest rates.

We will also investigate heterogeneity by students’ gender, CCT status, growth mindset, pre-intervention academic performance (baseline test scores and O-NET score at the end of primary school); by parents’ and teachers’ aspirations for students’ education; and by attendance and behavior during the mentoring sessions.

Experimental Design

Experimental Design
Students are from low-income households, and many are recipients of conditional cash transfers under Thailand's Equitable Education Fund, which supports students from low-income households throughout compulsory education through conditional cash transfers and provides scholarships for upper secondary schools and beyond. Recipients continue to receive these transfers regardless of assignment; the interventions are additional to them. Participating schools are located in four provinces in northeastern Thailand, and the study sample consists of second-year lower-secondary students in these schools. Schools are randomized across three arms: a control group, a role-model webinar, and the webinar plus mentoring.

The baseline survey will take place in September 2026, before the webinar. The webinar will be delivered in November 2026, at the start of the second semester. Mentoring will run from November 2026 for one school year, ending before the period in which students apply to upper secondary schools. The endline survey will take place from October 2026 onwards. Administrative data on O-NET scores and post-lower-secondary pathways will be obtained after students complete lower secondary school.
Experimental Design Details
Not available
Randomization Method
randomization done in office by a computer
Randomization Unit
School. Randomization is at the school level only; there is no second level of randomization. Assignment is stratified by province, fraction of conditional cash transfer recipients (median split), and school size (median split).
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
67 schools
Sample size: planned number of observations
1500 students
Sample size (or number of clusters) by treatment arms
22 schools in the control group, 24 schools in the role model only group, 21 schools in the role model and mentoring group.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
With 67 schools randomized equally across three arms and an average of 22 students per school, we calculate minimum detectable effects for 80% power at the 5% significance level, two-sided, for a pairwise comparison between two arms (approximately 45 schools). We assume an intracluster correlation of 0.10, 10% attrition, and a baseline-endline correlation of 0.5 at the individual level and 0.8 at the school level, so that baseline controls explain approximately 25% of individual-level and 64% of school-level variance in endline outcomes. Degrees of freedom are adjusted for 15 stratification fixed effects under the ANCOVA specification; individual fixed effects absorb these, so no adjustment applies there. Under these assumptions, the MDE is approximately 0.229 standard deviations under an ANCOVA specification and 0.250 standard deviations under an individual fixed effects specification. The MDE for a pooled comparison of two arms against the third is 0.196 standard deviations under an ANCOVA specification and 0.215 standard deviations under an individual fixed effects specification.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Ethics Committee, Graduate School of Economics, the University of Osaka
IRB Approval Date
2026-05-26
IRB Approval Number
R80526-2
Analysis Plan