Going Public: Breathing New Life into Learning
How to cite
Malik, Samreen; Blakeslee, David; Walter, Torsten Figueiredo; Bharti, Nitin Kumar; Asad, Khizer. 2025. "Going Public: Breathing New Life into Learning." AEA RCT Registry. AEARCTR-0017407. https://doi.org/10.1257/rct.17407-2.0
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
This project examines the effectiveness of air purifiers in improving indoor air quality and their impact on educational and behavioral outcomes in public school settings characterized by inadequate infrastructure and overcrowding. Conducted in collaboration with the Punjab Ministry of Education, our randomized intervention placed air purifiers in classrooms significantly more dilapidated than typical private schools. We assess the impacts on cognitive performance, classroom disruption (measured objectively through noise levels), and behavioral aspects such as moral values, cooperation, free-riding tendencies, and risk-taking behaviors—factors essential for effective teamwork and future labor-market success. Using baseline and follow-up assessments administered to grade 2 students at randomly assigned intervals of 1,2,4,5,8 or 9 weeks of exposure, our experimental design aims to provide causal evidence on the timing and cumulative effects of air purification. These findings offer critical insights for policymakers considering investments to improve air quality and educational outcomes in public schools.
Intervention
We provide purifiers in randomly selected grade 2 classrooms in public schools. We also place air monitors in treatment and control sections to track air quality.
Experimental design
We employed a stratified randomization design, assigning 240 schools into treatment and control groups within each stratum. Stratification was based on two dimensions: the gender composition of the schools and the exposure window, resulting in a total of 18 strata.
In total, 117 sections were randomly assigned to the treatment group, with the remaining sections placed in the control group. If a school had multiple sections, one section was randomly selected. Monitors were also randomly assigned within each stratum. To remotely collect noise levels (in decibels) and other data, mobile phones were placed in each selected classroom for the duration of the study.
Randomization maintained proportional representation across exposure durations and gendered statrum, with more weight given to 1 and 2 week exposures, second highest weight given to 4 and 5 week exposure and third highest weight to 8 and 9 week exposure.
Design details
Cited by
No citation data available for this registration yet.