Recovering Risk Aversion After Disaster: The Role of Confidence in Elementary School Students

How to cite

Yamaguchi, Uta. 2025. "Recovering Risk Aversion After Disaster: The Role of Confidence in Elementary School Students." AEA RCT Registry. AEARCTR-0016638. https://doi.org/10.1257/rct.16638-1.0

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

Uta Yamaguchi*

* Primary investigator

Registry ID
AEARCTR-0016638
DOI
https://doi.org/10.1257/rct.16638-1.0
First registered
Last updated
Status
On going
Trial dates
2025-06-17 – 2025-09-17
Countries
Private
JEL codes
C33 C91 D91
Registry record
http://www.socialscienceregistry.org/trials/16638

Abstract

In many developing countries, natural disasters are a major factor hindering economic growth. Disasters cause not only physical damage but also profound psychological effects on survivors. People who experience large-scale disasters in particular tend to become risk-averse, a tendency observed in multiple countries, suggesting it is likely a universal phenomenon. Excessive risk aversion suppresses actions essential for economic development, such as adopting new technologies or starting businesses, thereby hindering the long-term growth of developing countries.

While this “increased post-disaster risk aversion” has garnered attention in recent years, research on effective intervention methods to restore it to pre-disaster levels remains scarce. Existing research indicates that confidence strongly influences risk-preference decisions, and gender differences in confidence have been shown to create significant variations in risk aversion (Niederle and Vesterlund, 2007). This finding suggests that interventions to boost confidence post-disaster could potentially mitigate excessively heightened risk aversion.

This study will conduct a randomized controlled trial (RCT) targeting elementary school students in Hmawbi Township, Myanmar, which suffered damage from a major earthquake in March 2025. The participants are 1,007 fourth-grade students enrolled in three schools, divided into three groups for a 12-week intervention: a control group with no intervention, a group receiving a success recall intervention, and a group receiving both a success recall intervention and a praise intervention. The intervention consists of activities designed to evoke confidence by describing past successful experiences, supplemented by standardized praise from teachers.

The intervention's effectiveness will be evaluated by measuring risk preferences and confidence before and after the intervention. Risk avoidance will be assessed using the “Pencil Risk Task” at baseline and the “Balloon Analogue Risk Task (BART)” at endline. Confidence will be measured through self-assessment using the “Retrospective Self-Appraisal Scale.”

The purpose of this study is to verify whether an intervention to enhance confidence in elementary school students who have experienced natural disasters leads to a reduction in risk aversion and to clarify its effectiveness. If effective, it could become a new policy tool to support psychological recovery after disasters.

Experimental design

The intervention was implemented across three schools, each with four cohorts, totaling 1,007 students. Twelve cohorts were randomly assigned to three groups: a control group, treatment group 1, and treatment group 2. The control group received no intervention. Treatment group 1 and 2 received the intervention twice a week for 12 weeks.Treatment group 1 received an intervention using worksheets to recall past successful experiences. Treatment group 2 received the same intervention as treatment group 1, plus an additional intervention where teachers praised the students for the successful experiences they recalled. The purpose of the added praise intervention in treatment group 2 was to suggest that having successful experiences praised by others may further enhance confidence.

Design details

Randomization method
randomization done in office by a computer
Randomization unit
cohorts
Clusters
12 cohorts from 3 schools
Observations
1,007 students
Arms
358 students from 4 cohorts in control, 308 students from 4 cohorts in treatment group 1, 341 students from 4 cohorts in treatment group 2
Primary outcomes
Risk preference, Confidence
Secondary outcomes
Grade

Cited by

No citation data available for this registration yet.

Keywords