Sweating Over Student Behavior in Hot Schools, with Kristen McCormack

Notable Quotes

• Increases in hot temperatures driven by climate change will require rapid action to protect students from the long-term effects of extreme heat: “It’s pretty remarkable that even the additional warming that’s happening over a relatively short period of time translates to substantial increases in hot temperatures. That’s troubling when we think about the speed at which we’re going to need to act to protect students from behavioral changes and some of the adverse effects that are linked with the school discipline system, like changes in college attendance, arrests, and incarcerations.” (19:14)
• Schools are an important and appropriate place to prioritize safety as extreme temperatures become commonplace: “Warming temperatures could cause more behavioral issues in schools, potentially exacerbating existing disparities in the school discipline system. But, my findings also suggest that school AC might be protective, especially for students who have less access to AC at home. I think that’s really important. We want our schools to provide a safe learning environment for all kids, and temperature is part of that.” (21:26)
• Research on extreme temperatures informs how schools allocate scarce resources: “School districts have limited resources … The decisions about allocating scarce resources to meet a school district’s many needs are both challenging and also really specific to a district … I’m hoping that my results will help schools, districts, and policymakers at all levels who are thinking about school conditions better understand the effect of heat on students, the importance of these effects, and what can be done to better protect kids from these hot temperatures.” (21:50)

Top of the Stack

• Education Under Extremes: Temperature, Student Absenteeism, and Disciplinary Infractions by Kristen McCormack
• “The School to Prison Pipeline: Long-Run Impacts of School Suspensions on Adult Crime” by Andrew Bacher-Hicks, Stephen B. Billings, and David J. Deming
• "How Much Hotter Is Your Hometown Than When You Were Born?” interactive infographic from the New York Times
• “Meadowlark Sings and I Greet Him in Return,” a poem from the collection Devotions by Mary Oliver

The Full Transcript

Kristin Hayes: Hello, and welcome to Resources Radio, a weekly podcast from Resources for the Future (RFF). I’m your host, Kristin Hayes. Today, I’m very pleased to introduce our listeners to Kristen McCormack, RFF’s newest fellow who began at RFF, at least this time around, in July of 2025.

Kristen worked at RFF back in the day before heading off to get her PhD in public policy at Harvard, then working for several years at the US Treasury Department. We’re really thrilled that she’s officially now back in the RFF fold.

So, today, we’re going to be talking about some of Kristen’s work on education under extremes: temperature, student absenteeism, and disciplinary infractions. I will say, this is only one of many interesting topics that Kristen both knows about and has interest in. And we opted to focus on this one today as, I’d say, a good entrée into one of the types of questions that Kristen is likely to consider in years to come. So, consider this your introduction to someone who’s got a wide range of interests and skills. I’m really excited that you get to e-meet her today, so stay with us.

Hi, Kristen. Thank you so much for joining me on Resources Radio.

Kristen McCormack: Thanks for having me. I’m really excited to be here.

Kristin Hayes: Oh, great. Well, we always like to welcome our new fellows on the podcast. And I have to say, I specifically requested to be able to have this conversation with you because I thought it would be really fun to have a Kristin-Kristen chat. So, that’s what we’re doing today. I’m pretty psyched about that.

I said a little bit about your research trajectory, but it was a very little bit. So, let me invite you to share more about your academic and your professional journey that landed you back at RFF.

Kristen McCormack: Sure. So, I’ve been interested in environmental policy for a long time. When I was younger, I was mainly interested in animals and wildlife, and I gradually moved more toward climate and environmental issues more broadly. That combines a lot of things that I both care about and find exciting to work on.

One of my formative experiences was in late high school and early college, when I had a really cool opportunity to intern at the National Center for Atmospheric Research. I spent a few summers there working mainly with people in the Research Applications Lab on projects with really interdisciplinary teams that were policy relevant. That experience gave me an opportunity to see how research can inform policy, and I’ve been really interested in that intersection ever since.

Kristin Hayes: Was that out in Colorado?

Kristen McCormack: Yeah, where I’m from.

Kristin Hayes: Oh, you’re from Colorado?

Kristen McCormack: Yeah.

Kristin Hayes: I didn’t actually know that. Well, great. Now everyone knows that, including both the hosts and our listeners, but that’s awesome.

Kristen McCormack: So, you mentioned that I was at RFF before. When I came back to RFF last month, I realized there was a 10-year gap almost to the day between my first-first day at RFF and my first day coming back after my PhD and time at Treasury. So, that’s kind of fun.

One of the things that I have appreciated more and more during my time away from RFF is the work that people do here to translate research to different audiences. When I was at Treasury, I was mainly working on tax policy implementation. So, that put me in a position to both be experiencing policymaking and also to primarily be a consumer of research for the first time. That experience really centered how important policy-relevant research and the communication that accompanies it is to informing policy. But yeah, I really loved working here before. It’s such a special place. Excited to be back.

Kristin Hayes: Oh, that’s awesome. Yeah, and I appreciate the shout-out for the communications part of that, too. We obviously have wonderful communications colleagues literally in the room with us right now as we record this podcast.

Kristen McCormack: Right. Of course.

Kristin Hayes: Yeah, it’s a really good reminder that it really is a whole life cycle of research production and then communication that makes it work. Great. Looking forward to talking to you in the future about many other topics.

But today, as I mentioned, we’re going to be talking about the impact of heat on schools and students. This is my second conversation on the podcast on related issues, but I will say it feels just as relevant today as it did the first time. We’re right back at the same time of year; students are going back to the classroom. And it’s just as relevant as it has been—and perhaps more relevant than it was—the last time that I spoke with a researcher about this.

So, there’s some interest here, but why did you in particular choose to look at this issue overall?

Kristen McCormack: Yeah, for the past few years—around this time of the year when kids are going back to school—we see a familiar wave of articles popping up, talking about how hot conditions are, how uncomfortable it is in these classrooms. These uncomfortably hot temperatures are being experienced by students right as they’re coming back to school. So, this is happening when they’re trying to get off on the right foot, when they’re trying to make good relationships with their classmates, with their teachers, and trying to create a solid foundation for learning for the rest of the year.

This is, I think, a real challenge facing our school systems. Schools have been experiencing a record number of high-heat days, and we can expect these uncomfortably hot temperatures to be increasingly just part of the reality as we look to the years and decades ahead. And at the same time, many of these schools are struggling with outdated or deteriorating infrastructure. That means it’s really challenging for them to protect their students from these conditions.

Something that I also find really troubling is how unequally these temperatures are being experienced. In the United States, Black, Hispanic, and lower-income kids live in hotter areas, and they have less access to air-conditioning (AC) both at home and at school. So, as we think about these hot temperatures increasing in the future, I think there’s a real concern that this is going to exacerbate existing inequality in educational outcomes.

But I did want to say one thing that is more on a positive note, which is that research in these areas might help policymakers understand the effects of hot temperatures and how they can better protect these students from these conditions. That’s actually another reason that I decided to work on this topic. I find working on these kinds of questions, where there’s a real gap in knowledge and also a clear use for that knowledge and policymaking, to be really rewarding.

Kristin Hayes: That’s excellent. I didn’t really think about the temporal nature of some of these challenges: that the hottest conditions are going to be right when students are coming back into the classroom, establishing that foundation, as you noted, and then right at the end of the year, potentially when students are really trying to sort of solidify knowledge. Everyone’s a little bit tuned out anyway by the end of the year. They’re hard times for these hot temperatures to really be coming into play. So, that’s a good flag. I’m glad you mentioned that.

Okay. Well, clearly this is an important issue. So, you in your work looked at the impacts of warmer temperatures on two specific aspects of student behavior. One was absences, and one was disciplinary referrals. I think there are a number of metrics that people could conceivably look at. Why are these particularly interesting metrics to consider?

Kristen McCormack: Both of the aspects of behavior that I study in this paper cause learning to be disrupted, and there’s also evidence that they both lead to worse academic outcomes and also worse later-life well-being.

Another thing that I also want to mention before I talk a little bit more about these specific later-life effects is that across the country, there have been large disparities in income, race, and ethnicity for these outcomes, as well. I talked about the disparities in terms of exposure, but this is also something that we see in terms of existing disparities. I think that understanding these outcomes and how they’re going to be affected by temperature is really important, partly because of that.

I want to focus on the effects of disciplinary referrals for a moment to explain some of the reasons we might care about that outcome in particular. What we know from the literature is that receiving a referral, especially one that results in a serious response like a suspension or an expulsion, can have really detrimental effects. There’s a paper I cite in my paper by Andrew Bacher-Hicks and others that shows that kids who are assigned to stricter middle schools are more likely to be suspended, are less likely to attend four-year colleges, and are more likely to be arrested and incarcerated in early adulthood.

There’s also a really sizable literature pointing to a physiological effect of hot temperatures that you might expect to also affect students and teachers. Hot temperatures have been shown to affect physical and mental health, and many studies also support the hypothesis that on hot days, people have worse impulse control, are more irritable, and also just cognitively don’t perform at the same level. So, we might expect all of these effects to translate into behavioral changes.

Kristin Hayes: I used the term disciplinary referral without actually defining that very well myself. So, maybe I can ask you to sort of say just a little bit more, because you put some color around that, which I think is important. What does a disciplinary referral actually mean?

Kristen McCormack: It’s capturing a wide range of behavior. And actually, this is another reason that I think this topic is interesting. Most of the studies of these kinds of adverse behavioral outcomes are studying adult behavior. But in the school setting, you can capture a lot more. The majority of the disciplinary referrals in this setting are for disruptive or defiant behavior. The data that I have captures even minor behavioral issues. This range of behavior represents real disruptions in learning, focus, and in interpersonal relationships. But even though those disruptions probably exist in other settings, we aren’t able to observe them as easily. At least they’re not showing up in our data sets.

I also wanted to mention another reason that studying behavior in this setting is beneficial. This is getting a little bit more into methodology, but schools are very scheduled spaces. Most of the time, school calendars are the same regardless of what’s happening with the weather, with snow days being the obvious exception.

And student attendance is also observed, as well. This means that I’m able to avoid a really common challenge in observational studies of the effects of temperature on behavior, where temperature might affect not only the types of behavior occurring, but also the number of interactions people are having and how observable those interactions are.

Kristin Hayes: Right, because you can’t just not go to school if it’s … I mean, I suppose you could not go to school if it’s hot, and I think we’re actually going to talk about that.

Kristen McCormack: Right.

Kristin Hayes: But presumably, the idea is that this is a setting in which people are going to be doing the same things at the same time of day, regardless of what the temperature is. So, that gives you a baseline to work with. Yeah.

Kristen McCormack: Right. You can see when kids are absent. And then when kids are at school, they’re going to the same classes. It’s very different from observing adult behavior where people’s schedules might change a lot.

Kristin Hayes: Interesting. Okay. So, just to put a little bit more color on this story, I’d like to ask you to just paint a picture of the school district with whom you worked on this study. I know we’re not able to specifically identify them, but any sort of mental image you could build for us of the type of school district that we’re talking about?

Kristen McCormack: Sure. My setting is a large urban school district in the United States. It’s one of the 50 largest districts in the country and the largest in its state, so it serves a lot of students. In terms of the environmental conditions, it’s a place that experiences all four seasons including really hot and really cold temperatures. During my study period, many of the schools in the district also had incomplete access to air-conditioning.

One of the reasons I was excited to partner with this district is that they’ve been working to increase school AC access. I’ve learned a lot from working with them, and it’s been a really valuable partnership.

Kristin Hayes: It’s great to understand, again, the type of environment in which you were doing this analysis. One other stage-setting question about the analysis I wanted to ask was: I think a key challenge for many studies is trying to identify not just correlation between things, but causation. We hear that a lot: “correlation, not causation.” It’s almost a nerdy refrain sometimes, but it matters, right, to know not just that these things are happening in parallel, but that one is actually causing the other.

So, I wanted to ask if you can give us kind of a layperson’s version of how you were actually able to estimate causal impact in this study. I promise we’ll get to the findings, but it feels like the causality is important here. So, anything that you could share about your data that would shed light on that would be great.

Kristen McCormack: Yeah, of course. Maybe I’ll start by describing the data, because I think it’s going to be helpful to understanding what I’m doing.

The data from the school district are really detailed. I use tens of millions of observations where I can see for every student and on every day where students were enrolled in school, whether they were present or absent, and whether they were involved in any kind of behavioral incidents. I also have information on student grades and demographics. And then I link these data with other sources of information that help me paint a picture of the environmental conditions students are experiencing throughout their day.

So, first, I bring in daily environmental data, like outdoor temperature, snow, rain, pollution, et cetera. Then, I construct a data set to characterize access to AC at school and at home. At the school level, I have building AC information. And to get a picture of the temperatures students are experiencing at home, I link student census blocks with county assessors data to come up with a census-block-level average residential AC measure.

I was really excited to be able to look at AC in this paper, because data on AC are notoriously difficult to come by, and I think the data were able to really give me a helpful sense of the conditions that students are really experiencing.

And then the final piece of data—and I promise I’m getting to your question—was school calendar information.

So, this brings me to this question of how you estimate the causal impact of temperature on students. It’s one of the most common questions that I get asked from economists, from non-economists, from friends. I think everyone has a sense that temperatures change throughout the seasons along with many, many other things that could affect student behavior. So, I wanted to be able to make sure I was isolating the effect of temperature from all of these other seasonal changes.

I actually have controls for every single day of the school year. Those controls—along with controls for school, year, demographic characteristics, et cetera—mean that when I ask, “What is the effect of temperature and behavior?”, what I’m essentially doing is asking, “If it’s hotter on the first day of school this year than it was on the first day of school last year, what does that do to behavior?” Then I’m asking that question again and again for every single day of the school year. And then I use that information that I constructed about student demographics and access to AC at home and at school to try to get a sense of who is most affected by temperature and why.

Kristin Hayes: Fascinating. And it’s such a reminder, too, that this wasn’t a single data set. It obviously was a massive data set that you had access to, but you as a researcher also brought in complementary data sets that you were able to put together to answer these complicated questions. So, it wasn’t just one.

That was my main question: Were there multiple sources that you were looking at piecing together to be able to answer the questions you wanted?

Kristen McCormack: Right. Multiple sources. It ended up being a lot of data, and I like these big data projects.

Kristin Hayes: It sounds like big, big data, but great. Well, I always like to ask that question, too, because I think it is valuable just to really, again, set the stage of how many different and clever ways, quite frankly, people are thinking about teasing out real impacts and not just saying, “Yeah, it’s hotter. We imagine this is going to happen,” but really getting into the details and being able to answer that with some conviction. So, that’s great.

Okay. I’ve stalled long enough, or I’ve asked enough background questions. I really do want to make sure to turn to the findings here and give those time. So, you identify three main findings in the work, and I wanted to just invite you to talk us through those in turn.

Kristen McCormack: Sure. First, I find that both hot and cold temperatures lead to more student absences, and I find the largest effect among Black, Hispanic, and lower-income students. And then, consistent with previous literature, I also find that absences increase with snow.

Second, I find that hot temperatures—like temperatures above 80 degrees—lead to an increase in disciplinary referrals. But I find that this increase in referrals is only happening in schools that aren’t air-conditioned.

This is a working paper at the moment, but to give you a sense of magnitude, my current numbers suggest that in these schools, referrals increase by 7 percent and 21 percent on days with temperatures between 80 and 90 degrees and days above 90 degrees, respectively, relative to days with temperatures between 60 and 70 degrees.

I also see that the largest effect I’m observing here is among students who are less likely to have AC at home, who in this district are more likely to be Hispanic students and more likely to be lower income.

Kristin Hayes: Interesting. So, when there’s no respite from the heat, it’s particularly challenging, because students who don’t have access to AC at home or at school are sort of doubly exposed to the heat over time. Is that a fair assessment of that finding?

Kristen McCormack: Yeah, I think that my results suggest that hot temperatures are going to be hardest for those students.

Kristin Hayes: Yeah. Okay. Well, I will say when I first looked at this paper, the increases that you mentioned, they do seem very intuitive to me. And I’ll say this as a person who gets, honestly, super grumpy on hot days. This is very anecdotal and not very scientific, unlike your study, but it’s certainly something that I think is familiar to a bunch of us, that there’s a different level of anger and frustration, and, I don’t know, grumpiness.

I think in the paper … There’s a pretty significant body of literature on how higher temperatures are correlated—at least correlated—with a range of disruptive behaviors. So, again, no surprise that it’s showing up here, but I’m guessing there were also some surprises, perhaps, in your findings. I don’t know if those were about magnitudes or if they were about populations, or anything, but I’m curious if you just want to share anything that really surprised you. It doesn’t have to be a radical surprise, but findings that were perhaps a little bit unexpected for you.

Kristen McCormack: I think I’ll mention two things. I’m actually going to start with something that surprised me in the descriptive statistics. I think I mentioned before that, during the years I focus on in my study, about half of schools have AC. One thing that’s noteworthy in this district is that access to AC at schools doesn’t tend to follow all the same patterns that we see in national trends. For example, in this district, students attending air-conditioned schools are less likely to be white. And my measure of household income is pretty similar between students attending air-conditioned and non-air-conditioned schools.

I also mentioned this because, when we compare outcomes between two groups of students, we always want to think about what differs between these two groups. In this district, there’s very little evidence that students who may be more vulnerable to heat are disproportionately attending non-air-conditioned schools.

Something else that isn’t so much of a surprise, but definitely increased in salience in my mind, was the speed at which hot temperatures are becoming more common. I do this exercise at the end of my paper where I basically look at a 50-year window of time between 2000 and 2050 and ask how the expected temperature changes over this period would affect student behavior and later-life outcomes if additional adaptation—like more AC—doesn’t happen. It’s pretty remarkable that even the additional warming that’s happening over a relatively short period of time translates to substantial increases in hot temperatures. That’s troubling when we think about the speed at which we’re going to need to act to protect students from behavioral changes and some of the adverse effects that are linked with the school discipline system, like changes in college attendance, arrests, and incarcerations.

Something else that made the timing of these warming temperatures more salient to me was the experience of presenting these results to different places in the United States. When I was giving presentations to different audiences, I usually started my presentation with this really nice New York Times infographic where you can enter your hometown and then see what’s going to happen to hot temperatures over your lifetime. This exercise of pulling up this plot for different places and seeing people’s reactions really emphasized to me how ubiquitous this problem is going to be—and is already—and the importance of trying to get ahead of the challenges we can expect to result from these changing temperatures.

Kristin Hayes: Interesting. Okay. Well, on that note—and I think you make a good point about how this is only going to grow in salience as time goes on, given the data that we have now—I think it’s really important to talk about the point you made right at the beginning, which is how this is an area where the information gap that you are aiming to fill can actually feed back into the decisionmaking context and really help decisionmakers moving forward.

So, you noted, too, that school districts, almost inevitably … I think it is somewhat inevitable that they have scarce resources to allocate for things like building upgrades, putting in more air-conditioning, really helping to sort of find these technological solutions that might help make this better. But, clearly, there’s still a resource constraint here. So, how can this work and the findings that you have been uncovering in this research help inform some of those decisions?

Kristen McCormack: First, I think that my findings can help districts, educators, even families, plan better for these really hot days. Recognizing the physiological toll that hot days have, who is most vulnerable, and how that could translate into changes in attendance and behavioral issues can hopefully help people prepare for what they’re going to see in the future.

The results of my paper suggest something really troubling. In the absence of additional adaptation, warming temperatures could cause more behavioral issues in schools, potentially exacerbating existing disparities in the school discipline system. But my findings also suggest that school AC might be protective, especially for students who have less access to AC at home. I think that’s really important. We want our schools to provide a safe learning environment for all kids, and temperature is part of that.

You did mention something really critical, which is that school districts have limited resources. In my paper, I present some results that might help districts think about how installing AC in different schools might affect their students. But the decisions about allocating scarce resources to meet a school district’s many needs is both challenging and also really specific to a district.

Schools that aren’t able to air-condition are attempting to cope in other ways, like shifting the school year or canceling school more frequently. I’m hoping that my results will help schools, districts, and policymakers at all levels who are thinking about school conditions better understand the effect of heat on students, the importance of these effects, and what can be done to better protect kids from these hot temperatures.

Kristin Hayes: Yeah, that’s interesting to think about, the various solutions aside from just air-conditioning, which to me is the go-to of like, “Oh, that’s clearly the solution here.” But obviously that’s an energy-intensive solution in its own right. Are there other things that might be less cost intensive and potentially even less energy intensive that could help? Any other examples you have of how people are shifting and trying to address these warmer conditions?

Kristen McCormack: I think that there is literature on different things that schools have been trying. I also think it’s really important to point out—because it does come up in some conversations—the fact that AC is energy intensive. And I think one thing that—and I’ll just say this from my personal view—I think that this is a really important place to get temperature control right. There are a lot of different places that you could save on energy, and this is an area where I feel like it makes sense to prioritize these kids’ environments in their schools.

Kristin Hayes: Yeah, that’s a good point. On the cost-benefit ledger here, the benefits of using AC in this particular context seem really high.

Great. Well, I am sure that some parent somewhere is going to be listening to this as they think about sending their children back to school. We’re recording in early August of 2025, so we’re pretty close to the beginning of this school year. It’s great to have a reminder of this really important issue right as people are heading back that direction.

Thank you so much for talking us through this; again, one of your many interests. One of your many both past and future studies. Hopefully we’ll have other opportunities in the future to discuss.

Kristen McCormack: Yeah. Thanks so much.

Kristin Hayes: Let me close with our regular feature, Top of the Stack. As you may be aware from previous episodes, I would invite you to recommend good content of any variety that you might want to suggest, that might be of interest to our listeners. So, Kristen, what’s on the top of your stack?

Kristen McCormack: Well, I’ve always really loved reading poetry, especially in the morning. I find it to be a really good way for me to start my day. And I’ve been coming back recently to Mary Oliver, who’s one of my favorites. I find her deep attention to the natural world to be really grounding. And I’ve been trying to be a bit more deliberate about reading poetry or a bit of a book in the morning, because my instinct is to jump right into the news.

The poem that actually is on top of my stack is a poem that I read yesterday called “Meadowlark Sings and I Greet Him In Return.” I really like meadowlarks. They’re very common in Colorado, and so the poem feels a bit more special because of that.

Kristin Hayes: Oh, that’s so great. I don’t think we’ve had too much poetry or poems on Top of the Stack before. So, I learned something else new about you. That’s a great new genre for our top of the stacks to head in.

Kristen McCormack: Great.

Kristin Hayes: Really appreciate it. Highly recommend. Great. Well, thank you again. It’s been a pleasure to catch up.

Kristen McCormack: Thanks so much for having me.

Kristin Hayes: You’ve been listening to Resources Radio, a podcast from Resources for the Future, or RFF. If you have a minute, we’d really appreciate you leaving us a rating or a comment on your podcast platform of choice. Also, feel free to send us your suggestions for future episodes.

This podcast is made possible with the generous financial support of our listeners. You can help us continue producing these kinds of discussions on the topics that you care about by making a donation to Resources for the Future online at rff.org/donate.

RFF is an independent nonprofit research institution in Washington, DC. Our mission is to improve environmental, energy, and natural resource decisions through impartial economic research and policy engagement. The views expressed on this podcast are solely those of the podcast guests and may differ from those of RFF experts, its officers, or its directors. RFF does not take positions on specific legislative proposals.

Resources Radio is produced by Elizabeth Wason, with music by RFF’s own Daniel Raimi. Join us next week for another episode.