Information and Stated Educational Migration Choices: Experimental Evidence from India

How to cite

Demirel, Merve. 2025. "Information and Stated Educational Migration Choices: Experimental Evidence from India." AEA RCT Registry. AEARCTR-0016378. https://doi.org/10.1257/rct.16378-1.0

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

Merve Demirel*

* Primary investigator

Registry ID
AEARCTR-0016378
DOI
https://doi.org/10.1257/rct.16378-1.0
First registered
Last updated
Status
In development
Trial dates
2025-08-04 – 2026-08-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/16378

Abstract

This study investigates how students' beliefs about college destinations influence their higher education choices, and whether testimonial information from previous students can shift both beliefs and preferences. Using data from secondary school students in India, I first estimate the baseline relationship between migration-related beliefs and stated college preferences. I then test the causal impact of peer testimonials on beliefs and preferences through a randomized information experiment, with particular attention to gender differences in these relationships.

The study comprises two complementary analyses. First, I estimate preference formation using discrete choice models that exploit within-subject variation across college destinations. Second, I use randomized assignment of testimonial information to generate exogenous variation in beliefs, allowing me to identify the causal impact of beliefs on preferences and test whether information interventions can influence educational migration decisions.

The intervention provides students with testimonial information from current university students who migrated from similar backgrounds to study in one out-of-state destination. Each testimonial focuses on one of three key concerns about educational migration: safety, parental approval, or social integration. Students are randomly assigned to receive one testimonial, or to a control group that receives no additional information. The testimonials feature same-gender migrants sharing their personal experiences and addressing common misconceptions about studying away from home.

Intervention

The information treatment consists of testimonials from migrant students who share the respondent's gender and have moved to the out-of-state destination, focusing on the experiences of the migrant student on one of three non-pecuniary dimensions; safety, parental approval, or social integration at the destination.

Experimental design

The pure control group will comprise 50% of the survey sample. The remaining 50% will be evenly divided across three treatment arms (receiving testimonies related to parental approval, safety or social integration), with each treatment group assigned 16.7% of the sample.

Design details

Randomization method
Two-stage randomization using random uniform numbers (0-1) generated automatically within the survey instrument: Stage 1: A random number determines treatment status (0.00-0.50 = control; 0.50-1.00 = treatment) Stage 2: For participants assigned to treatment, a second random number determines treatment arm (e.g., 0.00-0.33 = Treatment A; 0.33-0.67 = Treatment B; 0.67-1.00 = Treatment C)
Randomization unit
Individual
Clusters
15 schools
Observations
900 pupils across 15 schools
Arms
450 in treatment, 450 in control
Primary outcomes
Index of willingness to migrate Target non-pecuniary cost of migration
Secondary outcomes
Index of non-pecuniary costs of educational migration

Cited by

No citation data available for this registration yet.

Keywords