Reducing School Dropout in Honduras

How to cite

Koning, Bart de; Riehl, Evan; Hall, Matt. 2025. "Reducing School Dropout in Honduras." AEA RCT Registry. AEARCTR-0016030. https://doi.org/10.1257/rct.16030-1.0

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

Bart de Koning* , Evan Riehl , Matt Hall

* Primary investigator

Registry ID
AEARCTR-0016030
DOI
https://doi.org/10.1257/rct.16030-1.0
First registered
Last updated
Status
In development
Trial dates
2025-05-01 – 2026-05-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/16030

Abstract

We evaluate the impact of an after school program called "Atrevete Club". The goal of the program is to reduce the likelihood of dropping out among students who are at risk of doing so. We conduct a randomized field experiment in 10 schools involving 3,816 students who are predicted to be above the median level of dropout risk.

Experimental design

The experimental sample is drawn from 7,567 students in grades 7 to 11 at ten schools in Honduras and includes all students with a predicted dropout risk above the median in the whole sample. In addition, the experimental sample is topped up with the next most at-risk students to ensure that there are at least [2.2 times the Atrevete Capacity at the school] students from each school. The factor of 2.2 is chosen so that each school has a control group that is at least as large as the treatment group plus a waitlist that is 20 percent of the size of the treatment group. In total, 3,816 students are included in the experiment, which is a little over half of all students. The reason for not including all students is that the Atrevete Club is targeted at, and expected to be most effective for, at-risk students.

Within the experimental sample, students were randomly chosen for the Atrevete Club at each school. The sample is stratified based on students’ predicted dropout probability (above or below median in the experimental sample) and gender, which gives four strata. For each school and stratum, a proportional number of students is randomly assigned to the treatment group. Since the Atrevete capacity at the school is not always a multiple of four, it is typically not possible to perfectly match the intended assignment rates of the strata. Thus students were randomly assigned to the Atrevete Club up to the largest multiple of four below the capacity constraint, and beyond that students were admitted to the Atrevete Club from the strata that were furthest from the intended ratio one-by-one until capacity was reached.

Next, [0.2 times the Atrevete Capacity at the school] were randomized to the waitlist for each school. In case of non-compliance among students initially assigned to the treatment, students on the waitlist will be assigned to the treatment group in order to the treatment group.

Lastly, remaining students were assigned to the control group for each school.

Design details

Randomization method
Randomization done in office by a computer.
Randomization unit
Individual
Clusters
3,816 students
Observations
3,816
Arms
1,050 treatment, 210 waitlist, 2,556 control
Primary outcomes
Dropout
Secondary outcomes
School engagement Self-reported attendance Academic performance Friendships

Cited by

No citation data available for this registration yet.

Keywords