A Lab-in-the-Field Experiment on the Legacies of Herbicidal Warfare in Vietnam

How to cite

Ito, Gaku; Khan, Ghulam Dastgir; Tran, Duc; Yoshida, Yuichiro. 2024. "A Lab-in-the-Field Experiment on the Legacies of Herbicidal Warfare in Vietnam." AEA RCT Registry. AEARCTR-0014279. https://doi.org/10.1257/rct.14279-1.0

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

Gaku Ito* , Ghulam Dastgir Khan , Duc Tran , Yuichiro Yoshida

* Primary investigator

Registry ID
AEARCTR-0014279
DOI
https://doi.org/10.1257/rct.14279-1.0
First registered
Last updated
Status
In development
Trial dates
2023-08-01 – 2024-12-31
Countries
Private
JEL codes
D74, O17, Z13
Registry record
http://www.socialscienceregistry.org/trials/14279

Abstract

How does exposure to wartime violence shape individuals' attitudes and behavior in a post-conflict society? This project employs a lab-in-the-field experiment with approximately 600 participants to explore the impact of wartime violence on the attitudes and behavior of survivors, their descendants, and non-descendants. We will randomly assign two treatments: (1) an information stimulus priming participants about herbicidal warfare during the Vietnam War, and (2) varying the recipient status in a dictator game. We will then assess the associations between these treatments and our key outcomes—stated empathy toward herbicide victims and the allocation of rewards. Additionally, we will analyze the conditional average treatment effect (CATE) to quantify how direct (survivor) and indirect (descendant) exposure to wartime violence, combined with the treatments, influence individuals' attitudes and behavior.

Intervention

We randomly assign participants to different conditions by varying two factors: (1) the order in which key outcome questions, measuring primary outcomes like empathy toward herbicide victims and the dictator game, are presented relative to other survey questions; and (2) the recipient's status in the dictator game. In the first treatment group, participants answer the key outcome questions after being primed with information about herbicidal warfare. In contrast, in the first control group, these questions are asked before the priming. In the second treatment group, participants play a dictator game with an anonymous recipient from a household affected by herbicide. In contrast, the corresponding control group plays the game with an anonymous recipient from a household not affected by herbicidal warfare.

Experimental design

We randomly assign participants to different conditions by varying two factors: (1) the order in which key outcome questions, measuring primary outcomes like empathy toward herbicide victims and the dictator game, are presented relative to other survey questions; and (2) the recipient's status in the dictator game. In the first treatment group, participants answer the key outcome questions after being primed with information about herbicidal warfare. In contrast, in the first control group, these questions are asked before the priming. In the second treatment group, participants play a dictator game with an anonymous recipient from a household affected by herbicide. In contrast, the corresponding control group plays the game with an anonymous recipient from a household not affected by herbicidal warfare.

Given the technical constraints in the field and the lack of detailed information about sample individuals and households outside the field, we independently and randomly assigned the two treatments to half of the sample individuals using simple randomization, without implementing any blocking

Our sample consists of 600 individuals, with one individual per household, in Vietnam: 300 from households affected by herbicidal warfare and 300 from households not affected by herbicidal warfare.

Our econometric analysis will compare the outcomes of interest across different treatment conditions and explore potential interactions between the treatments and individual- and family-level exposure to herbicide.

Design details

Randomization method
Randomization is done by a computer. Given the technical constraints in the field and the lack of detailed information about sample individuals and households outside the field, we independently and randomly assigned the two treatments to half of the sample individuals using simple randomization, without implementing any blocking.
Randomization unit
Individual (one per household)
Clusters
N/A.
Observations
600 individuals (one individual per household)
Arms
1. Treatment Group 300 individuals, with (approximately) 150 from households with herbicide victim(s) and (approximately) 150 from households without herbicide victim(s). 2. Control Group 300 individuals, with (approximately) 150 from households with herbicide victim(s) and (approximately) 150 from households without herbicide victim(s). Given the technical constraints in the field and the lack of detailed information about sample individuals and households outside the field, we independently and randomly assigned the two treatments to half of the sample individuals using simple randomization, without implementing any blocking.
Primary outcomes
1. Attitudinal: Stated empathy toward herbicide victims 2. Behavioral: Amount of the allocated rewards to another (anonymous) person in a dictator game

Cited by

No citation data available for this registration yet.

Keywords