# Odd Gender Pattern At Maths Olympiad

Cousin E pointed out something odd about the International Mathematical Olympiad. It’s an annual competition for high school students. And girls do super poorly in it. We ran some stats on the data for 2015 and 2016, and found that a national team with more than one female member gets less than half the median points per capita of an all-male team. With one female member, it’s 59-78%.

The question I want to address is not whether women are in empirical fact worse at maths than men. Nor do I, if this is the case, want to discuss whether it’s because of nature or nurture. I want to understand how the IMO works. Look at this.

1. Every country finds its six best maths students to make a team. In some countries, some of the best students are girls and beat out large numbers of boys to get on the team.
2. When these teams compete against each other internationally, suddenly teams with a female member do way worse than all-male teams.
3. Why are these girls super strong on the national level, but super weak on the international level? Is it because they are on male-dominated teams instead of working solo?

1. #1 rork
October 4, 2016

Maybe “suddenly” is not accurate. Hypothesis: 1) some countries have weaker teams. 2) they are more likely to contain a female member. 1 may partly cause 2, rather than the reverse.
“less than half the median points per capita” – seeing data would be nice since that seems astonishing. “per capita” seems not needed if all teams have 6 members, so I worry about what’s being said.

2. #2 Martin R
October 4, 2016

Some countries certainly do have weaker teams. We know that because teams do not get identical scores in the IMO. Hello?

Are you suggesting that the process which selects team members is prone to sub-optimal results in some countries?

Indeed, not all teams have 6 members.

3. #3 Eric Lund
October 4, 2016

One theory that I have heard (and I don’t know whether there is anything to it) is that the average boy and the average girl have about the same skill, but the variance for boys is much larger than for girls.

By its nature, this competition draws from the right-hand tail of the distribution. So if your country can find six boys who are far enough out on that tail, they are likely to do well in international competition. The few girls who are that far out on the tail aren’t far enough out to outperform those boys, according to this theory.

Of course, there could also be some gender dynamics issues. Some boys don’t play well with girls. And some countries have cultural issues with girls doing well at math (e.g., the United States) or anything in general (e.g., Saudi Arabia).

4. #4 John Massey
Hong Kong
October 4, 2016

The problem arises because your statements 2 and 3 are not true. Hong Kong, Canada and Hungary, which all included a female member, did not do that badly, and came in well above very many teams that were all male.

5. #5 Martin R
October 4, 2016

True, there is variation inside the two categories all-male and male-plus-a-few-girls. But the median per capita scores are extremely different.

6. #6 John Massey
Hong Kong
October 4, 2016

The reason is what Eric said. The far right of the distribution of IQ in any population is heavily dominated by males, and this obviously applies also to ability in Maths. (The far left of the distribution of IQ is also heavily dominated by males – there are very few female idiots anywhere.) The highest scoring female was Qi Qi from Canada, and she only managed a silver medal. But she still beat a hell of a lot of males at international level who were the best that their countries could come up with.

Plus, if there were no differences between the Gaussian distributions of male and female intelligence, the teams would not be so obviously heavily male-dominated, unless something is discouraging females from competing (which I would not discount, incidentally.)

7. #7 Lowell Gilbert
Massachusetts, United States
October 4, 2016

Based on my experiences with upper levels of the Science Olympiad in my country, I think it’s a reasonable assumption that team selection is sub-optimal as a rule. And sometimes extremely so. Competitors are preparing in their spare time, and if they do poorly in a regional competition, they may not get a chance to improve their performance with what they learned.

There is also a level of strategy built into the Olympiad format. The six top competitors from a country may all have very similar strengths, in which case someone who is weaker as an individual may improve the team score more.

In any case, it’s hard to even guess at causative relationships without knowing how the girls compared to their teammates.

8. #8 Martin R
October 4, 2016

Drawing on his experience from Chinese middle school, Cousin E suggests that the discrepancy may be because the national test and the international competition measure different things. He believes that you can win on the Chinese national level simply by being neat and organised, while the international competition also demands great mathematical skill.

9. #9 Eric Lund
October 4, 2016

unless something is discouraging females from competing (which I would not discount, incidentally.)

This could well be the case in certain countries, but will be very culture-dependent. For instance, many people in the US have an attitude that “girls can’t do math or science”. I know from personal experience that this is absolutely not true–indeed, my current boss is a US-born woman–but the attitude is still out there. The reason I know for sure it’s cultural: Whenever I visit a mom-and-pop restaurant run by people of East Asian ethnicity, one of the women (usually the mother, sometimes a daughter) operates the cash register.

I’m also aware that in Japan particularly, and other East Asian countries generally, women are expected to take charge of the household finances, as opposed to their American counterparts, who are assumed to want to marry Mr. Right or Prince Charming. Which leaves many American widows (women are on average younger than their husbands, plus they have a longer life expectancy) and divorcees scrambling for help when their husband dies/walks out, leaving them in charge of household finances for possibly the first time in their lives. This may be less of an issue for my generation, but it definitely applies to people my mother’s age and older.

10. #10 Eric Lund
October 4, 2016

He believes that you can win on the Chinese national level simply by being neat and organised

There are a couple of possible factors in play here. One is that the Chinese may place a cultural emphasis on “neat and organized” which in some cases carry comparable or higher weight than exceptional ability (there have been times in China’s past when this appears to have been the case). The other is if the national exam is not sufficiently rigorous to separate the top levels of mathematical ability, they may have to resort to “neat and organized” as a tiebreaker. These factors are not mutually exclusive.

I know that the SAT, the most common college examination in the US, is not sufficiently rigorous to distinguish people with levels of math ability beyond high school algebra from each other: among this year’s first-year class at my undergraduate alma mater, the median SAT math score is the highest score possible. I’m sure the selection examination for the Math Olympiad team is more rigorous than the SAT, but it might not be rigorous enough, especially in a country the size of China, to select six and only six students.

11. #11 Helen Simonsson
Sweden
October 5, 2016

Your number 3 is interesting, Martin. A lot of researchers are discussing what you are actually measuring when it comes to competitions – are you measuring the students ability to do maths or are you measuring their ability to compete or to do maths under pressure or in a particular situation? Perhaps the competion shows your ability to perform in groups, or how the students are performing in groups of mixed sexes? Research suggests that girls perform better than boys in maths and science in schools, but still both boys and girls believe boys are better. Girls have less confidence in their ability. Dweck noticed in a study in the 80s that the higher the girls’ IQ, the more likely they were to give up. Girls prematurely concluded that they don’t have what it takes to succeed. Bright boys, on the other hand, saw the difficult material as a challenge. They were more likely to redouble their efforts, rather than give up. Less confidence is a great problem in a competition situation. (http://www.voanews.com/a/girls-get-better-grades-than-boys-math-science/1904864.html And here is the study discussing the gender gap in competitions in maths http://web.stanford.edu/~niederle/NV.JEP.pdf )

12. #12 Kaleberg
October 5, 2016

Doing well in a math competition require two things: a lot of mathematical talent and a particular problem solving skill set. My high school had a serious math team, and I attended a few of their sessions. I’ve also known international class mathematicians and a lot of people who use a lot of mathematics in the course of their work.

Math competitions require a broad set of heuristics for solving problems that are intentionally hard to solve. Mathematicians tend to work on problems that are hard to solve by their intrinsic nature, not because they were designed that way. A basket of heuristics might be helpful, but it isn’t going to get you to the new level of abstraction that you need. If you use mathematics in your work, heuristics can be helpful, but you usually want a set that are informed by your field, not for general mathematical puzzle solving.

I’m guessing that you are seeing a cultural thing. I’ve known girls who enjoy solving mathematical puzzles, but I’m not sure how many girls get into the puzzle thing as a competition as opposed to for honing their skills. It’s one thing to do crossword puzzles to build up your working vocabulary, but it’s another thing to make it a major part of one’s identity. My experience is that girls, and women, tend to be more practical.

—-

P.S. My high school math team actually had a touch football team that played against teams from other teams. I remember them beating the debate team football team by a huge margin. Yes, it was a weird school.

13. #13 John Massey
Hong Kong
October 5, 2016

“He believes that you can win on the Chinese national level simply by being neat and organised.”

I sincerely doubt that. I know something about the Chinese education system, and it places great emphasis on Maths ability, because these days you need Maths for everything. And East Asians generally (and Ashkenazim) excel at Maths. I wouldn’t mind betting a lot of the team members from non-East Asian countries are Jewish.

If everything else is equal, then ‘neat and organised’ might give you the edge over another student of otherwise equal ability. But otherwise…no, I don’t believe it. Not in China. Not in Hong Kong either. Of course, presentation matters, it always matters, but first you need to show exceptional Maths ability, which is a function of IQ (noting that Chinese have a mean IQ of 106 and Ashenazi Jews have a mean IQ of 112, and the tails of the distributions are fat, i.e. there are a shitload of very smart Chinese and Jewish kids.)

I work a lot with young Chinese engineers/trainee engineers in Hong Kong, and most of them are outstandingly bright – in both intelligence and application they run rings around their Anglo Australian counterparts.

14. #14 John Massey
October 5, 2016

The acid test of what Cousin E has claimed is to notice that all of the members of the top American team are in fact Mainland Chinese, and the reason America is top is that the wealthy elite Chinese intelligentsia target America and its ivy league universities as the most desirable place to get to. So America got to field the cream of the Mainland Chinese Maths students.

And they didn’t come top just by being neat. Organised, yeah, obviously.

15. #15 John Massey
October 5, 2016

Maybe not all – the one called Alan Liu sounds like he might be a HK Chinese. HK Chinese also have a very high mean IQ.

Ashwin Bhattacharya is obviously South Asian. Michael Kural? Dunno, could be anything. Possibly Ashkenazim by ancestry.

But I’d bet money that the other 4 are Mainland Chinese, or first generation Americans of Mainland Chinese parents.

16. #16 Martin R
October 5, 2016

Nature / nurture.

17. #17 John Massey
October 5, 2016

Other 3, I mean. Having just failed basic arithmetic, I obviously would not have done well in the competition.

18. #18 John Massey
October 5, 2016

#16 – It’s an iterative interaction of genes and environment.

19. #19 John Massey
October 5, 2016

My daughter came out very bright, but a day dreamer at school. She cruised through her A Levels without paying overly much attention. She learned the Maths because she needed to, but it didn’t really excite her much. She was much more into Biology and Chemistry.

We were never ‘tiger parents’, never pushed her, just let her go at her own pace, so she just remained on cruise control all the way through.

But now, doing post-grad, she is needing to revise all the Maths she learned, and she has discovered a love for Maths that she didn’t have before. So she’s absolutely creaming it, because now she’s excited/motivated to really understand it and dig into it further and further.

Just brains alone doesn’t do it – it needs brains plus application, which means motivation, and a lot of motivation is an environmental thing.

20. #20 Eric Lund
October 5, 2016

Michael Kural? Dunno, could be anything.

Definitely not West European, and definitely not East Asian, but beyond that, not a name I can easily place. If pushed I would guess South Asian, but that’s a very low confidence WAG. Ashkenazi ancestry is possible if he is Central/East European or Middle Eastern, but not likely if he’s Turkic or South Asian.

21. #21 John Massey
Hong Kong
October 6, 2016

A quick google of ‘Kural’ suggests it is a Tamil name. Ashwin Sah is also South Asian – I found him straight off by googling.

India has a national mean IQ of 85, but the Indian population is so genetically diverse, and there is so much genetic substructure among the population that national averages don’t mean much for India, and any country with such a huge population is bound to produce the occasional mathematics genius. They also excel in English spelling competitions – you can safely bet the farm that any national or international spelling competition will be won by a South Asian, usually a female (which should also not surprise anyone).

So, I make out the all conquering USA team to comprise 3 South Asians, 2 Mainland Chinese and 1 Hong Kong Chinese.

It’s a safe bet that all of the South and North Koreans are Koreans, Singaporeans are all Chinese, and Taiwanese are all Chinese. Then you get to Russia. Then the UK, whose team includes 2 HK Chinese and 1 South Asian. Then Hong Kong (who included one girl – yay for HK!!!), then Japan, then Vietnam, and then Canada whose team was all Chinese, including Qi Qi, who was the highest scoring girl in the competition and is obviously Mainland Chinese.

It’s not looking great for WASPs, is it?

22. #22 Martin R
October 6, 2016

Hey, I’m doing my bit for eugenics! One kid is 1/8 Ashkenaz and the other is 1/2 Han Chinese.

23. #23 John Massey
Hong Kong
October 6, 2016

That’s not eugenics, Martin, that’s just sensible avoidance of inbreeding.

Plus think of all the good we’re doing, preventing the main population groups of the world from drifting further apart genetically and speciating.

24. #24 Martin R
October 6, 2016

If those foreigners want to speciate, then they shouldn’t look so damn tasty.

25. #25 John Massey
Hong Kong
October 6, 2016

It’s pretty safe – with two genetically distinct populations, you only need one intermarriage (euphemism) per generation to avoid them drifting further apart genetically.

I don’t know how long it would take for a human population to speciate, but the Australian Aborigines didn’t, and they were pretty well genetically isolated for about 50,000 years.

In the age of jet travel, I don’t see any human population being able to remain genetically isolated for hundreds of thousands of years. Not with adventurous people like you and me around, anyway.

26. #26 Lassi Hippeläinen
October 6, 2016

John Massey #6: “The far left of the distribution of IQ is also heavily dominated by males – there are very few female idiots anywhere.”
That explains why there are so few female politicians.

27. #27 Eric Lund
October 6, 2016

Lassi@26: “Suppose you were an idiot, and suppose you were a member of Congress; but I repeat myself.” –Mark Twain

28. #28 John Massey
Hong Kong
October 7, 2016

Prompted by Lassi’s observation, I was just checking out Sarah Palin’s Wikipedia page, and I discovered that Moscow is a city in northern Idaho.

Who knew?

29. #29 John Massey
Hong Kong
October 7, 2016

There is also a borough in Pennsylvania called Moscow.

Altogether, there are 24 places in the USA called Moscow, plus one in Canada, one in India and one in Scotland.

Seriously, the stuff you find that you don’t want to know.

30. #30 Eric Lund
October 7, 2016

Joah@28: The Moscow in Idaho is a university town, so it’s probably one of the more liberal (Communist, if you believe certain right-wing types) areas of the state.

The university town in which I live has the same name as a much more famous university town in North Carolina. Both are named after a university town in northern England. On one occasion I was supposed to interview with a group south of London who initially assumed I was coming from the last. On one other occasion a vendor saw the word “University” and the name of the town in our address and shipped our order to North Carolina.

Then there was the time I was flying from Los Angeles to Manchester, and the lady at the check-in desk (this was in the days before online check-in was available) asked to see my passport. I was going to the Manchester in New Hampshire, not the one in England, and since this was a domestic trip, I didn’t have my passport with me (in the US one normally uses a driver license for such things). She should have known from the airport codes (MHT and MAN, respectively) that I was a domestic passenger.

31. #31 Phillip Helbig
Tyskland
October 7, 2016

Someone once tried to sue a travel agent because he wanted to travel to Florida but ended up in Russia or vice versa: St. Petersburg.

32. #32 Eric Lund
October 7, 2016

Phillip@31: I’ve heard a different version of the story: that someone from the USSR wanted to travel to the St. Petersburg in Florida but ended up in Petersburg, Alaska. Allegedly, the godless Commies dropped the “St.” part of the name. A more likely explanation, for either this version or yours, is that somebody wasn’t aware that the St. Petersburg in Florida doesn’t have its own airport (if you are traveling there by air you fly to Tampa, just across the bay) and sent the passengers to a Petersburg/St. Petersburg that did.

An actual case I am aware of was the couple that found cheap air tickets online from London to Sydney via Halifax. It was only when they boarded the connecting flight in Halifax (a puddle jumper of some kind) that they realized there is a Sydney in the Canadian province of Nova Scotia.

The largest city in Maine and the largest city in Oregon are both called Portland. The latter is named after the former. According to Oregon lore, a guy from Maine and a guy from Massachusetts played a game of chance for the right to name the new city, and the guy from Maine won. Supposedly, if the contest had gone the other way, that city would have been called Boston. Check the airport code carefully if you are flying to either city: PWM is about 4000 km east of PDX.

33. #33 Candice H. Brown Elliott
https://candicebrownelliott.wordpress.com
October 7, 2016

In this post, and most especially, in the coments, I note a number of both highly sexist and racist themes that need to be examined and unpacked.

First, as to the gender differences in ability: much has been made in recent years of the Greater Male Variability Hypothesis, to wit, that men as a population exhibit greater variablity, and thus dominate (yes, I chose that word on purpose) at the very high tail end of the distribution. I have no doubt that as measured, this was correct. But that alone can NOT be used to assert that this is an essentially biological determined phenomena. Social factors can easily explain this difference as well as any putative biological one. For example, if two equally matched people marry… in todays (and especially in the past) gendered expectations, one quickly finds that one of these people is expected to shoulder a far greater part of the household chores… leaving fewer hours with which to hone ones skills… and indeed with lower “energy” left over to with which to do so in those fewer hours.

Second, if something is biologically determined, we would expect that the phenomena would be stable over both time and space. But even the AVERAGE IQ is unstable over time with the well noted Flynn Effect showing that as a society becomes wealthier and better educated, IQ (perhaps even g itself?) increases. So, do we see a stable difference between men and women at the high end over time? No. In the early ’80s, the ratio of boys to girls at age 13 who scored 700 on the SAT was 13:1 but in 2005, it had dropped to 2.8:1 So, how about stability over space? In the US, the variability ratio for mathematics tests was 1.19 while it was 1.00 and 0.99 in the Netherlands and Denmark respectively.

If I can share some personal anecdotes? In 1983 I was admitted to the Materials Science Dept. at Stanford… into the first class that had more than a token woman at 30%. The very first day of the very first class the professor showed up to class wearing a shirt with cartoons printed on it (it didn’t seem very “professorial”)… upon examining those cartoon images, the motive was of physical violence of a man beating a women. The women in the class turned to each other with an expression that universally said “Are you seeing what I’m seeing?” The men in the class were totally oblivious. So… how many women do you think would want that professor to be their advisor? Personally, I dropped out to pursue a VERY lucrative and successful career in Silicon Valley start-ups (of which I have founded several).

I had a close friend, another woman graduate student in the Physics dept. who had similarly bad experiences… and eventually transfered to the Geophysics dept. at another university where women were much better treated. My friend is now a manager at the Lawrence Livermore Labs.

A final anecdote… as a child, I HATED mathematics, though I was very good at it. I especially hated my freshman algebra teacher who was a misogynist pig (and only taught the class so that he could be the football coach in the afternoon). As an adult, I make my living developing digital signal processing algorithms for images based partly on psychophysic of human vision.
How many girls get turned off by mathmatics and science because of exposure to men like the above?

A final word on teams and gender. It is common for boys to have a style of communication that excludes girls by requiring one to be excessively (by girl’s gender role expectations) aggressive. Thus, mixed teams will fail to take advantage of their female teammates skills. The team would have to make a practiced effort to be inclusive and in some competitions, don’t have the chance to develop this strategy before hand.

34. #34 John Massey
Hong Kong
October 8, 2016

Assuming that you have unpacked the highly sexist themes in the comments (not that I can discern, but whatever), you don’t seem to have unpacked the highly racist themes that you have noted.

It is not at all clear to me how the teams in the competition operate, but individual team members get individual scores. For example, looking at Canada, two males got a gold medal, two members including one male and one female got silver medals, and one male got a bronze medal, and one male got an Honourable Mention (i.e. no medal).

But there is also a Team Leader and Deputy Leader for each team, and it is totally unclear to me what those people actually do in the competition.

So it looks to me like they are working on the problems individually and getting individual scores, and then those scores are aggregated to get the team score.

In any case, Canada alone shoots down Martin’s contentions 2 and 3: Canada came 12th out of (I count) 109 countries – that’s hardly doing much worse than all-male teams, and Qi Qi, the Canadian girl, in scoring a silver medal, is hardly weak at the international level – she’s way up there, not right at the top, but close to the top level.

What is of interest to me is not how the girls performed relative to the boys, because I’m not interested in the far right tail of the distribution – in Civil Engineering, which is maths-heavy but at a level at which you need to be good, but not actually a mathematics genius, in Hong Kong and China, the girls are holding their own against the boys, and often out-competing them, and on average around 50% of Civil Engineering graduates every year are female, often including the top graduates. (This is not true in Australia, where female enrolment in all engineering courses combined is only at 12%, and I doubt it’s true in America.) What interests me is why all of the teams are so heavily male dominated. I know plenty of girls in Hong Kong who are really into mathematics, and very aggressively competitive against their male counterparts. Why are more girls not entering this competition?

And please don’t give me a lot of stuff from America about this, because America is not the world, despite what many Americans obviously think, and other parts of the world are very different from America, and from the Netherlands and Denmark (about the most homogeneously white country you could have chosen) too. I want to know what is happening at a global scale.

BTW, I have looked at IQ scores in China, and the Greater Male Variability Hypothesis does not seem to be borne out there – I could discern no differences in the distributions of male and female IQ.

35. #35 John Massey
October 8, 2016

“if something is biologically determined, we would expect that the phenomena would be stable over both time and space.” No, actually dead wrong. There is such a thing as fœtal alcohol syndrome and fœtal drug syndrome. Intelligence in adulthood can be heavily affected by disease, parasites and malnutrition during childhood. In fact, malnutrition has an epigenetic effect – it takes three generations of adequate nutrition before its effects are eliminated. Mercury poisoning, particularly methyl mercury poisoning, crosses the placental barrier, so pregnant women who ingest methyl mercury (most commonly from contaminated seafood) at sub-toxic levels can have children who suffer from neural diseases because methyl mercury attacks the central nervous system.

So there is no such thing as ‘purely biologically determined’ except in a perfectly controlled environment in which all environmental factors can be eliminated, including those going back three generations. And that perfectly controlled environment can never be created in practical terms.

That is why geneticists say that intelligence is about 60 to 70% heritable. If there were no gene/environment interaction, it could possibly be 100%, but there is no perfect world in which to test that. And the heritability of intelligence at 60-70% is as measured in developed countries where there is very little childhood malnutrition, not much childhood disease (check out how much of a depressing effect on intelligence catching a dose of the measles can have – it’s pretty scary) and few parasites.

In Africa and India, where childhood malnutrition, disease and parasites are all at much higher levels, the heritability of intelligence is a lot lower; in other words, the environmental factors are having a bigger influence on intelligence.

The Flynn effect is flattening out now in modern developed countries, because (hypothesis) the adverse environmental influences on intelligence have gradually been eliminated or reduced to low levels. Plus in modern developed countries, there are now many more ways for children’s brains to be stimulated, particularly in the very young – it looks like mental stimulus during the first three years of life might make a difference.

So it’s just dumb to talk about intelligence being “biologically determined” – it is largely, but even if it is 70% heritable, that is far from “purely biologically determined”. Plus you have to factor in epigenetic effects, like if your grandmother suffered malnutrition as a child, which would show up in inherited intelligence, but is actually an environmental factor.

If you are really interested in this subject, it is worth looking up what happened to the crime statistics in America after the discontinuation of use of leaded fuel in automobiles. That alone probably gave the Flynn Effect a bit of a kick-on.

36. #36 John Massey
Hong Kong
October 8, 2016

#31 – LOL! You’d be pissed, wouldn’t you? You deplane, expecting to see palm trees and beaches, and there’s some guy demanding to see your passport in Russian.

37. #37 John Massey
Hong Kong
October 8, 2016

OK, I’ve read the rules now. The Team Leaders and Deputy Leaders form a jury to select the problems, and to adjudicate on marking the scores. They are quarantined from the participants during the competition, and play no part in solving the problems.

Team members work individually on solving the problems, and then their scores are aggregated to get the team score. So, in effect, each team member is working solo.

Things that might count against females in the competition? Well, they don’t allow calculus, which surprised me, but they do allow problems from branches of mathematics which are not conventionally taught in schools. So, it sounds like you need to be a real nerd to do well. Maybe females are going to be more focussed on doing well in the mathematics they need to cover in school, in order to get a good university place, and less likely to indulge themselves in things that are off the course, i.e. maybe they tend to be less nerd-like. I don’t know.

Not including calculus, but including things that are outside of the normal mathematics curriculum would certainly put me off from competing. I was dead focused on getting the marks I needed to get into engineering at university, and learning the mathematics that I would need to do engineering. I didn’t have time for any off-curriculum nerdishness. At my school I also had to engage in mandatory military training and team sports, so I didn’t have any spare time to engage in nerdish pursuits – I was too busy running around in the Australian bush lugging a Lee Enfield .303 rifle and a field radio to ‘make a man of me’, or trying to mangle poor little bastards on the opposing team on the rugby field (at which I excelled, incidentally – not at playing rugby generally, just the mangling part).

There are so few girls in this competition that I suspect the trends that Martin and Cousin E have detected could just be artefacts of the data. The smartest girls will have their eyes on what really matters, which is getting into the course they want to do at the university they want to do it at; i.e. fewer female nerds who like to hang out with other nerds. That would fit with my observations of male and female engineering students, graduates and young engineers – the females are very focused on doing precisely those things they need to do to advance themselves professionally; the guys tend more to be distracted and mess around. But that’s just a subjective impression.

In open cut mines in Australia, they need people to drive those huge off-road trucks, hauling ore around, and the people they most prefer to choose to drive them are 18 year old girls. The reasons they give are that girls at that age are more responsible, concentrate on the job better, day-dream less, muck around less, and take fewer stupid, unnecessary risks than guys do. I’m not making that up; a lot of young women have worked in mining doing just that, and some of them have made so much money doing it that they have their house paid off by the time they are 24. I’m guessing that they are less inclined so spend what they earn on drinking as well.

38. #38 Lassi Hippeläinen
October 8, 2016

There is a Sea of Moscow in the Moon. However, it is not named after the city but after a state of mind.
http://www.nature.com/nature/journal/v488/n7412/full/488429b.html

39. #39 John Massey
Hong Kong
October 9, 2016

I assume to reach this state of mind, you have to consume a sufficient quantity of vodka.

But I’m just guessing.

40. #40 Omega Centauri
October 9, 2016

There is are issues of breath versus depth of ones knowledge. Will you do better in this particular competition if your knowledge is well-rounded, versus being exceptional in a few areas? Traditional math is very proof heavy and plodding in pace, engineering would be very fluency heavy, and physics requires a combination of creativity and audaciousness (try an expansion or whatever, and don’t be put off because you don’t have rigorous proof that it works). Different tests could select for one or another of these somewhat specialized skills.

A cooperative team competition might be more interesting -and more like the real world. Diversity of the skill set would then be more important, than broad but shallow knowledge for each member. Modern research exploits a diversity of people with deep, but not necessarily broad skills. Testing for individual scores of a team would not select for this sort of thing.

41. #41 John Massey
Hong Kong
October 10, 2016

Not really. The idea that IQ testing is flawed because people have different kinds of intelligence is not borne out by testing. Intelligence even correlates pretty well with musical ability.

I think you are sort of on the point – the IMO looks to me like something that appeals to maths nerds, not just highly intelligent students who are good at mathematics. The fact that it excludes calculus is a clue and, in my view, a major flaw.

“engineering would be very fluency heavy” – hmmm.

Fluency = “The quality of consistently applying skill correctly in the manner of one well-practiced at it…” Well, yes, sort of; “..requiring little deliberate thought to perform without mistakes” – hell no. Engineers work to standards and codes of practice, and really need to. But every new problem is a new problem; it’s not like a process of mindless repetition; not at all. Show me the engineer who is working like that, and I will show you a very bad engineer. And bad engineers are very bad news. A doctor who makes an error could kill someone. An engineer who makes an error could kill hundreds. There’s a reason we have independent checking by senior, experienced people.

42. #42 Phillip Helbig
Tyskland
October 10, 2016

“The idea that IQ testing is flawed because people have different kinds of intelligence is not borne out by testing.”

As Isaac Asimov said, the higher one’s IQ, the less one thinks that it can be meaningfully measured.

43. #43 Sean M
October 10, 2016

Philip: I can sure say that whenever people start talking about how important and real IQ is, they don’t sound as intelligent as they usually do 🙂 Whatever its uses for psychologists, I don’t think that it helps people without training in statistics have helpful conversations.

44. #44 Martin R
October 10, 2016

A friend of mine once quipped, “If you’re really smart, then surely you wouldn’t want to be a member of anything resembling Mensa?”

45. #45 Phillip Helbig
Tyskland
October 10, 2016

” Whatever its uses for psychologists, I don’t think that it helps people without training in statistics have helpful conversations.”

There is the famous story where President Eisenhower was shocked when an aide mentioned that half of the US population is below average in intelligence!

46. #46 Phillip Helbig
Tyskland
October 10, 2016

“If you’re really smart, then surely you wouldn’t want to be a member of anything resembling Mensa?”

I think their criterion is the top 1%, which would mean tens of millions of people would qualify.

47. #47 Eric Lund
October 10, 2016

“If you’re really smart, then surely you wouldn’t want to be a member of anything resembling Mensa?”

I sent the club a wire stating, “PLEASE ACCEPT MY RESIGNATION. I DON’T WANT TO BELONG TO ANY CLUB THAT WILL ACCEPT PEOPLE LIKE ME AS A MEMBER”.

48. #48 Robert
October 10, 2016

@Phillip Helbig

“There is the famous story where President Eisenhower was shocked when an aide mentioned that half of the US population is below average in intelligence!”

I’ve heard the same story, but with Reagan…

49. #49 BDoyle
October 11, 2016

Remember that these are teenagers. It’s possible that if you put them on a mixed-gender team, they will not only be thinking about math. I’m not just making a joke comment, I’ve been there, and it’s a distraction.

50. #50 John Massey
Hong Kong
October 11, 2016

#46 – It’s actually the top 2%. Plus you’re out by an order of magnitude, unless you are referring to a specific subset of countries. Actual membership is estimated at only 121,000, which means that even in the countries with the highest mean IQ, the very large majority of people with IQ>130 don’t bother.

#42 – But what else would you expect Asimov to say? You can just about tick them off on your fingers: Gould, Lewontin,…He did have something of a point though – the higher people’s intelligence, the more difficult it is to measure differences; like when you get up into the 160s – Terence Tao territory.

51. #51 Phillip Helbig
Tyskland
October 11, 2016

“Plus you’re out by an order of magnitude”

Note the “would”. Our points are the same: only a small fraction of those who could, do.

52. #52 Phillip Helbig
Tyskland
October 11, 2016

“He did have something of a point though – the higher people’s intelligence, the more difficult it is to measure differences”

Perhaps, but what he meant was that people who are really intelligent realize that quantifying intelligence in one number is to coarse. (Of course, part of the joke is self-referential, since this real intelligence is measured via IQ, but that’s part of the joke.) People who aren’t really intelligent, in turn, are impressed by a high IQ. Like poor kids who are more impressed by conspicuous consumption than real rich people are.