Edutainment, social network, and ethnic discrimination in rural Bangladesh

How to cite

Siddique, Abu; Vlassopoulos, Michael; Zenou, Yves; Islam, Asad. 2023. "Edutainment, social network, and ethnic discrimination in rural Bangladesh." AEA RCT Registry. AEARCTR-0010730. https://doi.org/10.1257/rct.10730-1.1

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

Abu Siddique* , Michael Vlassopoulos , Yves Zenou , Asad Islam

* Primary investigator

Registry ID
AEARCTR-0010730
DOI
https://doi.org/10.1257/rct.10730-1.1
First registered
Last updated
Status
Completed
Trial dates
2022-09-01 – 2023-05-31
Countries
Bangladesh (Rajshahi Division)
JEL codes
C93, D9, I31, J15, O12
Registry record
http://www.socialscienceregistry.org/trials/10730

Abstract

In this document, we provide the research design and analysis plan for evaluating the impact of information dissemination through a documentary film that educates the ethnically dominant Bengalis about the ethnic minority Santals in Bangladeshi polyethnic villages. We want to evaluate whether exposure to information affects behavior and opinions of the ethnic majority towards minorities. More importantly, we want to test whether information diffusion is stronger when it is targeted to network-central ethnic majorities. We will measure outcomes using survey questions, administrative data, lab-in-the-field experiments, and a field experiment.

Intervention

The objective of this project is to evaluate an innovative intervention aimed at updating the ethnic majority people’s existing information regarding ethnic minority people, with a view to curb prejudice and foster interethnic harmony. More specifically, we propose to carry out a randomized experiment in rural Bangladesh to evaluate the effectiveness of local screenings of a documentary film that educates the ethnic majority Bengali about the lives and livelihoods of Santals (one of the largest ethnic minorities in Bangladesh) in reducing ethnic discrimination. We also leverage social relationships of the ethnic majority in villages and target information to network-central Bengalis to test if it reduces discrimination towards minorities further. Being exposed to entertainment media (edutainment) promises to change norms and behavior through the channels of providing both new information as well as making existing information attractive, as posited by social learning theory. However, information dissemination through central agents in a network can be more effective in spreading information than broadcasting information widely, such as via mass media. Therefore, in this study, we aim to provide the first empirical test of both theories in the context of ethnic discrimination.

Experimental design

Using a cluster randomized control trial in the Rajshahi and Naogaon districts of Bangladesh, we will evaluate the effectiveness of a documentary film on the lives and livelihoods of Santals of Northwestern Bangladesh. To carry out the fieldwork, we collaborated with two local NGOs, Ashrai and SARCH, and randomly selected 120 polyethnic villages for the experiment (where the fraction of minorities is between 30-70 percent). We divide these villages equally (40 villages in each of the three treatment arms) to provide the following treatments: (1) Treatment Random: 14 randomly selected households per village will watch the documentary film on Santals. (2) Treatment Central: 7 most network-central households and 7 randomly selected households per village will watch the documentary film on Santals. (3) Control: 14 randomly selected households per village will watch a documentary film on floriculture or flower farming in Bangladesh (same length as the documentary film on Santals). We will closely follow Banerjee et al. (2019) to indentify network-central residents in our Treatment Central villages. We rapidly surveyed 20 random people per village and asked them to name the 15 most suitable people (or households) in their villages who would be a good diffuser of information. Based on this list, 7 most consistent and highly ranked names will be selected for the Central treatment. The screening will take place at each participants house, individually. The documentary film will be screened via 10 inch tablet PCs connected to USB speakers. This way participants and other household members will be able to watch the documentary film together. In each village, we will collect baseline and endline data on the main participants that watch the documentaries, as well as on an equal number of randomly selected people per village (untreated) to check for spillover effects.

Design details

Randomization method
Stata
Randomization unit
Village level
Clusters
120 villages
Observations
Main participants: 1,680 Spillover: 1,680 Total: 3,360
Arms
560 main and 560 spillover in each of the three arms: total 1,120 (40 villages) per arm.
Primary outcomes
Survey Outcomes: Interethnic interaction index, Number of non-coethnic close friends, Water bill charge, Number of visits to non-coethnic neighbors, Number of non-coethnic visitors, Discriminatory opinions index, Mental health index, Subjective well-being index, Trust, Food security index (only applicable for Santals), New employment (only applicable for Santals), Monthly income (only applicable for Santals). Lab-in-the-field outcomes: Altrusim and solidarity. Field experiment: Productivity in a labor market task. Administrative data: Police complaints, Complaints made to village counselors.
Secondary outcomes
Potential mechanisms (intermediate outcomes): Emotions (sadness, angry, disgust, happiness), Willingness to help others, Stereotypes, intercultural competence.

Cited by

No citation data available for this registration yet.

Keywords