Referral information and persistence in multi-stage processes

Last registered on July 06, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Referral information and persistence in multi-stage processes
RCT ID
AEARCTR-0019061
Initial registration date
June 29, 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
July 06, 2026, 7:18 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
Manuel Munoz
Affiliation
New York University Abu Dhabi
Contact Primary Investigator

Other Primary Investigator(s)

Additional Trial Information

Status
In development
Start date
2026-08-03
End date
2026-11-30
Keywords
Behavior, Education, Labor
Additional Keywords
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
This online field experiment studies whether information about referral status encourages or discourages persistence in a multi-stage application process for a talent recognition program at a university in Colombia. Eligible students are invited to apply for the award. The application is a three-stage assessment process in which the top 25 percent of applicants receive a certificate of excellence and a subset of certificate recipients receive monetary prizes. Before invitations are sent, faculty members and eligible students nominate up to four eligible students through online surveys. Students are randomized within three mutually exclusive referral-status blocks: students nominated by faculty, students nominated by peers, and students who were not nominated. Within these blocks, the experiment varies whether students are informed about being nominated, not informed about being nominated, or, among non-nominated students, informed that referrals were collected but that they were not nominated. The primary outcomes measure persistence: a binary indicator for completing all three stages by the final deadline and a continuous measure of the number of stages completed. Secondary analyses use survey, assessment, and administrative data to study beliefs, confidence, assessment performance, and the composition of students who persist.
External Link(s)

Registration Citation

Citation
Munoz, Manuel. 2026. "Referral information and persistence in multi-stage processes." AEA RCT Registry. July 06. https://doi.org/10.1257/rct.19061-1.0
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)
Eligible students at a Colombian university are invited to apply to a talent recognition program. The application process consists of three weekly stages, each requiring approximately 20 minutes to complete. Before invitations are sent, faculty members and eligible students nominate up to four eligible students through online surveys. Students are assigned to treatment conditions based on their referral status. Students nominated by faculty are randomized into being informed or not informed that a faculty member nominated them. Students nominated only by peers are randomized into being informed or not informed that a peer nominated them. Students who were not nominated are randomized into receiving no referral information, being informed that faculty referrals were collected but that they were not nominated by faculty, or being informed that peer referrals were collected but that they were not nominated by peers.
Intervention Start Date
2026-08-03
Intervention End Date
2026-11-30

Primary Outcomes

Primary Outcomes (end points)
The primary outcomes measure persistence in the talent recognition program application process. I will use two measures: first, a binary indicator equal to one if the student completes the full three-stage application process by the final deadline, and zero otherwise; second, a continuous measure of the number of application stages completed, ranging from 0 to 3.
Primary Outcomes (explanation)
Persistence will be measured using administrative records from the online application platform. The binary persistence outcome is equal to one if the student completes all three stages of the application process by the final deadline, and zero otherwise. Students who do not start the application, start but do not complete all stages, or complete stages after the relevant deadline will be coded as zero. The continuous persistence outcome is the number of application stages completed by the final deadline, ranging from 0 to 3. A student who does not complete any stage is coded as 0; a student who completes only Stage 1 is coded as 1; a student who completes Stages 1 and 2 is coded as 2; and a student who completes all three stages is coded as 3.

Secondary Outcomes

Secondary Outcomes (end points)
- completion of Stage 1 and Stage 2
- perceived probability of receiving the certificate
- perceived reasons for being considered
- perceived academic ability
- confidence and expected performance
- assessment performance in the timed tasks
- composition of students who complete each stage
Secondary Outcomes (explanation)
Secondary outcomes include: completion of Stage 1 and Stage 2; perceived probability of receiving the certificate of excellence, measured in Stage 1 and at the end of Stage 3; perceived reasons for being considered for the award, measured through the Stage 1 open-ended question; perceived academic ability, measured using expected own percentile in the university GPA distribution; confidence and expected performance, measured after the timed task in each stage using expected rank among a hypothetical group of six participants; assessment performance in the three timed tasks; and the composition of students who complete each stage and the full process, measured using pre-treatment administrative characteristics.

Experimental Design

Experimental Design
This study is an individual-level randomized field experiment among students eligible to apply to a university talent recognition program in Colombia. Before invitations are sent, faculty members and eligible students nominate students whom they believe should be invited to participate. Eligible students are then randomized within referral-status blocks that depend on whether they were nominated by faculty, nominated only by peers, or not nominated. The intervention varies whether students are informed about their referral status when invited to apply. All students receive the same core information about the opportunity, including the three-stage application process, expected time cost, deadlines, selection criteria, the certificate awarded to the top 25 percent of applicants, and the possibility of receiving a monetary prize. The study measures whether referral-status information affects persistence in the application process.
Experimental Design Details
Not available
Randomization Method
Randomization done by statistical software.
Randomization Unit
Individual student.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
N/A - no clusters.
Sample size: planned number of observations
Planned observations - faculty nomination block: About 390 students. Planned observations - peer-only nomination block: About 1,700 students. Planned observations - no-nomination block About: 2,200 students. Planned observations - total: About 4,290 students.
Sample size (or number of clusters) by treatment arms
Faculty nomination block -- T1. Faculty aware: N=195
Faculty nomination block -- T1. Faculty unaware: N=195
Peer-only nomination block -- T3. Peer aware: N= 850
Peer-only nomination block -- T3. Peer unaware: N= 850
No-nomination block -- T5. No-faculty aware: N= 733
No-nomination block -- T6. No-peer aware: N= 733
No-nomination block -- T7. None unaware: N= 734
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Assuming equal allocation, no clustering, a two-sided test with alpha = 0.05, 80 percent power, and a baseline outcome rate of 50 percent, the approximate minimum detectable effects for binary outcomes are: T1 versus T2 in the faculty nomination block: 14.2 percentage points; T3 versus T4 in the peer-only nomination block: 6.8 percentage points; and T5 versus T7 or T6 versus T7 in the no-nomination block: 7.3 percentage points. For the continuous number-of-stages outcome, the corresponding approximate minimum detectable effects are 0.28 standard deviations in the faculty nomination block, 0.14 standard deviations in the peer-only nomination block, and 0.15 standard deviations in the no-nomination block.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
NYU Abu Dhabi
IRB Approval Date
2026-04-17
IRB Approval Number
HRPP-2026-8
Analysis Plan