Paper finds a surprising link between warm temperatures and math test scores | Hacker News Reader

Paper finds a surprising link between warm temperatures and math test scores

washingtonpost.com

40 pointsjfuhrman34 comments

Those who have been around long enough may recall the late Daniel Moynihan's discovery that math proficiency scores had a strong inverse relationship to the distance of the state capital from the Canadian border. As I recall, he thought it an excellent way for northern-tier states to raise revenue by renting out enclaves where other states might place their capitals.

For the unenlightened the point here is you need to control for race, which I doubt the study in the referenced article did.

Is it race or education policy? Hmmm...

Probably race in some way, because when you control for race all ethnic groups in America outperform the respective groups overseas. Swedish Americans score higher than Swedes, black Americans score higher than west Africans, etc.

Do you have a source for that?

Race is extremely difficult to control for opposed to income, employment, education, etc. It's difficult to understand from a european racial perspective, but it's like stuffing everyone from the mediterranean to scandinavia into the same race. People who are otherwise genetically disparate share skin color. This is extremely evident in Africa proper but is also evident in e.g. the american "African American" ethnicity.

TL;DR unless you're trying to correlate with a specific gene, you're gonna be called racist for trying to correlate skin color to anything. Race in general is a word that is difficult to use correctly in an argument unless you're shooting down someone else's argument. It is virtually meaningless outside of how people extrapolate phenotypes to ethnic identification.

> but it's like stuffing everyone from the mediterranean to scandinavia into the same race

That's a perfectly reasonable thing to do almost any way you try to look at it.

> That's a perfectly reasonable thing to do almost any way you try to look at it.

...why? Europe is a culturally, politically, and genetically diverse place. It makes no sense to group them by "race".

> genetically diverse place

On a relative basis, no it isn't.

I suffer from allergies compounded by an injury. This has led to a varying degree of obstructed breathing and discomfort.

Warmer weather tends to bring greater allergic and perhaps other factors influencing this condition. I notice a significant cognitive decline under these conditions.

When I work in air conditioned environments, I perceive my cognitive ability to increase as the cooler, drier, and filtered air gradually mitigates the symptoms of congestion and discomfort.

In my experience, there is a lot that medicine and formal study do not recognize nor study. For lack of a better word, a "hollistic" perspective that looks beyond easily reducible and isolated factors is significantly absent and is so particularly with respect to acknowledging and addressing current conditions as opposed to promoting one particular theory or product.

So, it's anecdotal, but for me, I have little difficulty believing such a correlation is possible. One might think that populations would adapt to warmer environments to not be cognitively challenged by them. But we've been moving around a lot, and changing our environment a lot, in recent decades. As just one possible reason why not.

Something I have always looked at was that, from a historical point of view, the fastest development in the industrial revolution occurred in more temperate areas, up to a pint where severe winters curtailed the effect. Thus Manchester developed more than Moscow.

A long time ago I came up with the half baked theory is that people in these climate zones had to work harder for food and shelter, and that the hard work for that had a motivating effect, or perhaps increased the reproductive success for those willing to put in the work. Whereas someone in a warmer climate could be half-assed with their shelter and food production and still get along fine.

It's possible that warmer temperatures just slow down brain function as well. People from warmer areas tend to have slower, drawn out speech patterns. Maybe there is something in it.

This would explain why the introduction of central heating slowed down speech pa... wait

You might be interested in Guns, Germs, and Steel. It has a similar N-S vs. E-W thing going on.

And with all geographical determinism there's gonna be too many edge cases for the theory to hold any water. See: diamond's academic reputation.

I also recommend reading "Guns, Germs and Steel" ... and some other literature. Some of the points you make have obvious counter-examples.

>> People from warmer areas tend to have slower, drawn out speech patterns

Where do you get that from? How many languages do you know? Have you spoken to an anverage Andalousian or somebody from Sud Tirol? To an average Dane, Belgian? To a Chilean or an Argentinian? I only speak about 4 and a half languages, but as soon as you consider dialects, local variations, etc ... speed and temperature don't seem to have any effect (and my own datapoints show the inverse relationship, but it's too little data).

Regarding working harder in cold areas leading to civilization ... didn't western civilization develop first in the middle east (fertile crescent)? ... and then the torch of the "leading western civilization" was passed around many warm nations (Greeks, Romans, Al-andalus & North Africa, etc) before the northerners got anywhere?

Regarding the language thing, I was talking about English speakers in general. I'm from a warmer climate and people tell me I have a slow drawl. Maybe it's a city thing that makes people talk faster and cities are all over the latitudes....I'm making a personal observation here and not seriously suggesting a scientific theory.

On the Fertile Crescent - yes agrarian civilisation spread in those areas but I was specifically referring to the industrial revolution - the point being that once people started using technology the more temperate regions flourished - as opposite to the more agrarian societies. My personal thought is that maybe this is because the rewards of the application of technology benefit colder climates.

I'll check out the recommended book, I'm always interested in these sorts of topics.

> “I’ve got to tell you, we puzzled over that for at least a year, to try to really wrap our heads around whether we’re doing it right and what might explain such a pattern of results,” said Graff Zivin.

I don't understand the reason for puzzlement. The result is straightforwardly intuitive. Discomfort is distracting, not to mention it takes a lot of energy to maintain homeostasis when it's either too hot or too cold. If anything, the researchers should be trying harder to falsify the obvious conclusion rather than dig up more reasons to believe it.

Googling "temperature" and "productivity" leads to multiple articles explaining how different office temperatures influence workers' productivity. The sweet spot seems to be somewhere between 70F and 75F, and productivity decreases as the temperature moves away from this range. Why would children respond any differently to discomfort? Does anyone seriously believe that the temperature has absolutely no impact on productivity?

So discomfort itself can be a confounding factor. I would very much like to know whether discomfort explains all the variance, or whether there's something else about high temperatures that affects the human brain in a different way. (Are our brains like CPUs that need to throttle down when they get too hot?) Likewise, I'd like to know whether acclimatization can explain the lack of long-term effects, because if so, the "global warming is making us stupider" angle is just plain stupid.

> But poorer math performance...on particularly hot days

> Overlay test scores with the average temperature in the county where they lived

Wait, what? Is the study about hot days or warm climates?

You cut off the quote too early:

> "...on the day of testing."

The "average" means they're not taking the high or low for the day, but it's definitely warm days, not warm climates.

Much more likely to have warm days in warm climates.

This is a HUGE claim. As many have said, the larger (stranger) the claim, the more skeptical we should be and the more evidence should be required (http://en.wikipedia.org/wiki/Marcello_Truzzi#.22Extraordinar...).

I like how you provided a citation for that, in case anyone disagrees.

The history of attribution is pretty interesting. :)

Hang on...

So in the second part of the study, these guys are trying to determine if the heat effects are cumulative, and find that they are not (obviously, since susceptibility to permanent damage due to > 70 degree weather would have crippled our species intellectually from the get go).

They then go on to explain that parents are somehow "adapting" to this by, for example, sending kids to math camp, thus negating the cumulative effect. Huh?

They explain the possible A/C confound like this:

“We don’t know what the kid is doing 2 hours before the test implementer arrives at the house, it’s quite possible the kids are outside,” he says. “It’s quite possible these kids are exposed to some heat even if they can then come in and cool off.”

I actually had a different and precisely contradictory explanation: the test-takers on hot days were more likely to be in heavily air-conditioned environments and thus the air in the test-taking environment is too cold (and dry, and probably mouldy).

It's certainly an interesting effect but they need to do a lot more work to establish the cause.

It also could be the effect that it's a nice day outside and you are inside taking a test (regardless of a/c)

Interesting; I was recently binge-reading some old blog posts and came across this[1] not dissimilar claim from 2012. Maybe it wasn't so crazy after all.

[1] http://squid314.livejournal.com/320770.html?thread=2386178

Beyond being just a crappy study... People have lower performance when they are uncomfortable? Who knew.

Or when they'd rather be playing outside.

Lower in math, not reading, so it goes.

if this isn't an exercise in correlation not being equal to causation then i don't know what is.

The part about the researchers being surprised that the effect isn't cumulative seemed like something from The Onion.

All the trust those researchers have on their finding, and all that certainty that the correlation isn't cumulative because children in hot weathers must study more.

No, the Onion wouldn't be able to get away with an article like that. People would claim up front that it's too contrived and unrealistic for comedy.

Care to elaborate? The study design seemed pretty good from the abstract--longitudinal looking at changes in scores of specific students.

Reply on news.ycombinator.com