Incentivizing School Attendance in Bangladesh: A Loss Aversion Conditional Cash Transfer Design

How to cite

Fujii, Tomoki; Ho, Christine; Shonchoy, Abu; Ray, Rohan. 2017. "Incentivizing School Attendance in Bangladesh: A Loss Aversion Conditional Cash Transfer Design." AEA RCT Registry. AEARCTR-0002373. https://doi.org/10.1257/rct.2373-3.4

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

Tomoki Fujii* , Christine Ho , Abu Shonchoy , Rohan Ray

* Primary investigator

Registry ID
AEARCTR-0002373
DOI
https://doi.org/10.1257/rct.2373-3.4
First registered
Last updated
Status
Completed
Trial dates
2017-01-15 – 2021-06-30
Countries
Bangladesh (Gaibanda)
Registry record
http://www.socialscienceregistry.org/trials/2373

Abstract

This study aims at incentivizing secondary school attendance in Bangladesh. We conduct a randomized control trial (RCT) that would entail giving conditional cash transfers (CCT) using a loss aversion design. The design seeks to exploit the widely observed psychological trait that people tend to more strongly avert losses than the reduction of the same amount of gains. The main research question is whether CCT may have a larger impact on school attendance when combined with a loss aversion design. Given the high costs involved in CCT programs, a higher impact design may help further promote the policy goal of encouraging school attendance while keeping costs contained.

Intervention

We implement a conditional cash transfer program in northwestern part of Bangladesh in which the households of the students receive cash when the students attend school. We will see how it affects various outcomes of the students and their households through surveys and experiments. We will also explore different designs of conditional cash transfer programs.

Experimental design

400 grade 6 and 7 student participants were recruited from 3 secular secondary schools in Gaibanda district in Northern Bangladesh. From late 2017, additional 400 new grade 6 to 9 student participants (50% male and 50% female) will be recruited from the 3 secondary schools. They will be randomly assigned to one of the 4 treatment arms described below. During the last three months of intervention (from around March 2018), participants in CCT treatment arms will be randomized into two subgroups each. One of the subgroup within each treatment arm will receive compensation of 30 taka per school day attended while the other subgroup will still receive 10 taka per school day attended.

Design details

Randomization method
Randomization done in office by a computer.
Randomization unit
Individuals
Clusters
N/A
Observations
400 observations from the beginning of the study and 400 additional observations from late 2017.
Arms
1/4 of total number of observations in each of the following treatment arms. A) Standard CCT B) Loss-aversion CCT C) SMS only D) Control In the last three months of treatment, students in treatment arms A) and B) will be subdivided into two groups to vary the transfer amount.
Primary outcomes
school attendance, educational outcomes, attitudinal measures, and various socioeconomic outcomes of the household.

Papers from this trial

  1. FUJII, Tomoki; HO, Christine; RAY, Rohan; and SHONCHOY, Abu S.. School attendance information or conditional cash transfer? Evidence from a randomized field experiment in rural Bangladesh. (2023). 1-78. Available at: https://ink.library.smu.edu.sg/soe_research/2666 https://ink.library.smu.edu.sg/soe_research/2666

Cited by

No citation data available for this registration yet.

Keywords