Can Information Nudges Prevent Learning Decay During COVID-19? Evidence from a Randomized Field Experiment in Bangladesh
How to cite
Fujii, Tomoki; Shonchoy, Abu; Ho, Christine; Ray, Rohan. 2021. "Can Information Nudges Prevent Learning Decay During COVID-19? Evidence from a Randomized Field Experiment in Bangladesh." AEA RCT Registry. AEARCTR-0007674. https://doi.org/10.1257/rct.7674-1.2
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
COVID-19 has created unprecedented difficulties for the education sector in Bangladesh, as schools remain closed since March 2020. To tackle this crisis, Bangladesh has taken a multi-pronged approach—delivering education through the national terrestrial television. However, the uptake of these programs has been low among rural students, and the leading causes of the low uptake have been information inadequacy and lack of commitment. This study aims to test through a randomized field experiment the impact of regular and repetitive information nudges on televised programs' uptake—delivered by SMS and automated voice calls to households. Moreover, the staggered proposed reopening of schools—based on grades—will help us identify the impact on learning outcomes, which will be measured by exams and SMS-based spot quizzes with achievement rewards. This study will help policy-makers understand effective solutions to reduce learning decay during the pandemic, which is scalable for Bangladesh and other developing countries.
Intervention
We stratify the sample of students by school, grade, and gender. Within each stratum, we will randomly allocate each student to one of the following two treatment arms:
Information group: Those in the treatment group will receive weekly information through SMS text and voice calls. The content of the SMS and calls will relate to information that is potentially useful for the students' education.
Control group: Those in the control group will not receive the weekly information through SMS text and voice calls.
Experimental design
For this study, we plan to recruite 1200 secondary school students, stratified by school, grade, and gender. The main objectives of our intervention is to understand whether the information nudges can improve students' learning behavior and outcomes in the treatment groups. We consider the following econometric model to estimate the intention-to-treat effects:
𝑌𝑖 = 𝛼 + 𝛽𝑇𝑖 + 𝛾𝑋𝑖 + 𝜖𝑖 (i)
where Yi is the TV viewership or standardized test score in English/Mathematics of individual 'i'; Ti takes a value of 1 if the individual belongs to the treatment group, and 0 otherwise;. The reference category here is the control arm students who do not receive information nudges. Xi will include any other observable covariate that is not balanced at baseline across treatment arms. All test scores will be
normalized relative to the control mean and standard deviation for every school-grade combination.
Design details
Cited by
No citation data available for this registration yet.