Effect of Classroom Air Quality Monitoring on Child Health and Attendance

How to cite

Kerwin, Jason; Bellemare, Marc; Horsager, Kent. 2022. "Effect of Classroom Air Quality Monitoring on Child Health and Attendance." AEA RCT Registry. AEARCTR-0008783. https://doi.org/10.1257/rct.8783-2.2

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

Jason Kerwin* , Marc Bellemare , Kent Horsager

* Primary investigator

Registry ID
AEARCTR-0008783
DOI
https://doi.org/10.1257/rct.8783-2.2
First registered
Last updated
Status
Completed
Trial dates
2021-11-01 – 2022-09-09
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/8783

Abstract

This project will examine how classroom air quality monitoring affects the health and attendance of children at childcare centers, preschools, and elementary schools. We plan to conduct a pilot test of this study at the University of Minnesota Child Development Laboratory School (CDLS).

Intervention

We will randomly assign classrooms in the center to either receive an air quality monitor or not. The air quality monitors detect CO2 levels, which can have two harmful effects. First, CO2 has direct negative consequences for alertness and thus can reduce learning outcomes. Second, CO2 is strongly correlated with the concentration of airborne viruses in the air, including COVID-19 but also many other respiratory viruses.

Teachers in classrooms that are assigned an air quality monitor will also receive basic guidance in how to use it. The monitors have an indicator for high CO2 levels and are easily read and interpreted by users, and teachers will be advised to open a window when levels are high.

Experimental design

Classrooms will be randomly assigned to either be in a treatment group (which receives an AQM) or a control group. The randomization will be done in Stata by generating a pseudorandom number for each classroom and sorting the classrooms by this number. The first classroom in this randomly-sorted list will be assigned to the treatment group, the second to the control group, the third to the treatment group, and so on.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Classroom
Clusters
13 classrooms
Observations
Approximately 150 students
Arms
7 treatment, 6 control
Primary outcomes
This pilot study will collect two administrative variables at the level of a student-day that we will use as outcomes: 1) absences 2) recorded symptoms (for days when they were in school). We will also construct a third outcome, which is equal to 1 whenever the student is either sick or has symptoms (or both), and zero otherwise. Our data will have a panel structure, following students (indexed by i) over days of school (indexed by t). The outcomes will be defined only for days when school was in session. They will also be defined only for days when the student was actually enrolled in school, so the panel will be unbalanced.

Cited by

No citation data available for this registration yet.

Keywords