Blog

It’s Time To Say Goodbye to the Hot and Stuffy Classroom

Earlier this week, most students in the U.S. were attending school in temperatures above 90°F. While this type of heat has long been endured by students and teachers in the southern parts of the country, it is now also becoming more common in the Northeast as a result of climate change. Recent research (by me and others) suggests that hot classrooms not only feel uncomfortable for students and teachers, but the heat also seriously interferes with learning. The impact is even greater for Black and Hispanic students. So what can we do?

That hot classrooms lower learning outcomes really isn’t surprising to any of us who remember sweltering inside hot classrooms at the end of June, unable to focus and unmotivated to learn. But simply comparing learning outcomes in hot and low-scoring (southern) vs. cool and high-scoring (northern) parts of the country is not enough to make the case, given that other factors like school spending also vary by geography.

In “Heat and Learning,” which I co-authored with Jisung Park, Michael Hurwitz and Jonathan Smith, we make a more rigorous argument using data that includes millions of U.S. students who took the PSAT exam in 10th grade and again in 11th grade. We show that these students, when compared to themselves, tended to score lower in whichever year preceding the exam was hotter.

In other words, the randomness of 10th grade being hotter than 11th grade, or vice versa, turns out to be clearly connected to a student’s performance on the exam. And, days of extreme heat are even worse than days of milder heat, so that days in the 90s are much more damaging to learning than days in the 70s.

Though air conditioning is not the only way to improve school temperatures, it can be an effective solution. Little reliable data on school air conditioning exists at the state or national levels,  so we surveyed students to learn whether their classrooms had effective temperature controls. Using their responses as a proxy for school air conditioning, we show that annual fluctuations in heat have little impact on the learning of students in schools that appear to be fully air-conditioned.

One troubling outcome of hot classrooms is that the impact is substantially larger for Black and Hispanic students than for White students, for at least two reasons. First, Black and Hispanic students tend to live in parts of the country that experience more days of extreme heat. Second, Black and Hispanic students are less likely to attend schools (or live at home) with adequate air conditioning. We show that the combination of these two factors implies that hot school days explain about 5% of the racial gap in PSAT scores. Improving school infrastructure could reduce that substantially.

Heat also interferes with the learning US students in younger grades and of students around the world.  In “Learning is Inhibited by Heat Exposure, Both Internationally and Within the United States,” Jisung Park, Patrick Behrer and I show that U.S. students in 3rd to 8th grade have lower test scores in years where their school district experiences unusually high temperatures. Again, this effect is driven largely by low-income, Black, and Hispanic students. This pattern appears again in international test score data (PISA). Compared to themselves in other years, countries have lower PISA scores when they experience hotter than usual school days.

In every data set we explored, hot school days interfere with the learning outcomes of students. All this suggests that temperature- and ventilation-focused school infrastructure investments will have large payoffs, particularly given that climate change implies school days will be increasingly hot. As a bonus, improved ventilation also reduces infection risks (from COVID-19, the flu, or the common cold) and indoor pollutant levels, which a growing body of research shows also interfere with cognition.

Most adults now work in buildings that we expect to be fully air-conditioned. Good temperature control and ventilation will help students and teachers be more productive in extreme heat, just as it helps the rest of us. We should provide no less to our students and teachers.

Joshua Goodman is an associate professor of education and economics at BU Wheelock. His research focuses on quasi-experimental estimation of the impacts of educational interventions.