Average IQ in Europe by country (map) | Hacker News Reader

Average IQ in Europe by country (map)

jakubmarian.com

30 pointsgivan61 comments

One of the authors of one of the reference materials (Tatu Vanhanen) of Lynn and Vanhanen is fairly suspect in his methodology of applying and analysing of IQ scores.

Fact #1: It's not actually obvious what IQ scores measure, what amount of cultural bias the scores have and does it even make sense to compare IQ scores between different populations. Well, it's a number so yes, you can numerically compare it, but - as we engineers would put it - it's not obvious that the scalar value of 'IQ' has the same dimension in disjoint populations.

There are various attempts to map human beings mental capabilites on various basis functions (IQ, Briggs-Meyers etc.). From the point of view of armchair natural scientist none of the the approaches are very convincing as hard core analysis tools. As facilitation tools they can provide value, but facilitation always includes discussion and holistic analysis of what actually the number means in each situtation.

Fact #2: Tatu Vanhanen has quite a lot of publicized work where he expresses views that are quite openly racist - veiled in a thin cloack of pseudoscientific analysis. It appears that he starts from the theory that some populations have higher IQ:s and then he massages the numbers until he can "prove" his original hypothesis of bias between some populations. I will not trust any number that man quotes because figuring out who he computed it would be a waste of time. IMHO, YMMV and so on but this has been his way or working for at least three decades. So, yes, crackpot radars ahoy.

Although it would be cool to pat oneself in the back and say yes, we finns are really smart... seeing Vanhanens work quoted here makes me actually a bit shamed.

If you look at the home page of the referred book "Intelligence: A Unifying Construct for the Social Sciences" http://www.ulsterinstitute.org/intelligence.html the first referred reviewer is http://en.wikipedia.org/wiki/J._Philippe_Rushton who wikipedia links also to racially biased research.

I don't have qualms about trying to understand human condition through measurement - but starting from some very limited, simple hypothesis and applying it to a complex emergent system like the human civilization is scientifically fairly suspect. You will always get correlation between large set of unconnected numbers and the way these guys work is that they collect IQ 'data' and then muddle through the numbers until they reach conclusions for which any 1920's western racial theorist would be proud.

This "research" would be totally in the "Pirates and global warming" category of the Spaghetti Monster creed http://en.wikipedia.org/wiki/Flying_Spaghetti_Monster#Pirate... - except it's not funny.

It turns out that calling things "facts" does not, unfortunately, make them facts.

To go beyond the armchair, try looking at, for example, http://books.google.ca/books?id=jj4tKn9gb1IC&pg=PT172

There's something wrong here... most (and most populous) countries have IQ below 100, but IQ 100 is defined as being the average of human populations. Unless other continents have significantly more intelligent people than in Europe (seems unlikely), then this chart must be significantly uncalibrated.

Maybe the overall IQ has lowered since we set the standard of 100? :)

As the movie "Idiocracy" put it, evolution does not necesserily reward intelligence.

On the contrary, raw IQ rises about 3 points per decade. The Flynn Effect.

http://en.wikipedia.org/wiki/Flynn_effect

OTOH, the 2012 map shows generally lower numbers than the 2007 map, which seems to suggest that Europe is running contrary to the general trend.

The whole article is about how the 2007 map was biased and how the 2012 map attempts to correct this bias. So you can't compare them the way you do it.

The only thing in the article about the relation to the older map is "Since the maps above [the 2012 maps] are based on more recent data, the map shown here [the 2007 map] should be considered outdated."

There's not one word in the linked piece about bias in the older map, so saying that's what the whole article is about is, well, odd.

Is a single point difference significant?

> Unless other continents have significantly more intelligent people than in Europe (seems unlikely)

Why does this seem unlikely?

It's unlikely that 1) there are significant differences in intelligence between different continents 2) that IQ, which is significantly influenced by education, would be lower in Europe where education is rather decent (probably at least in the better half, if you were to sort the world by quality of education).

> It's unlikely that 1) there are significant differences in intelligence between different continents

Given the known individual drivers of IQ, which are things not equal by continent, that seems dubious.

> that IQ, which is significantly influenced by education, would be lower in Europe where education is rather decent (probably at least in the better half, if you were to sort the world by quality of education).

Yeah, but the 100 IQ average wasn't set globally in the first place, its the average of the norming sample of the test, and most (all) of the test were developed in the first world with norming samples drawn from there.

Western Europe being a hair below first-world average (and Eastern europe farther below it) in the first world isn't at all surprising.

I've always thought that IQ-tests are biased towards education, too, and not just "raw intelligence". This seems to confirm it. The richer countries tend to have higher average IQ than the poorer ones.

No, not necessarily the richer - I can spot a direct correlation between the quality of the education system and the IQ reported here. Finland, the country with maybe the best schools(not Universities, schools) in Europe is leading.

Yet, the quality of schools could be correlated to teachers and politicians with higher IQ. Chicken and egg...

Or fewer low IQ people. Whites and Asians in the US have a higher average IQ than the US as a whole for instance.

If you were to compare a similar map from 1970 with the current one the notion of which came first would be much more clear. Before WWII IQ was pretty evenly distributed across the regions of Germany, but during the partition the IQs of children born in East Germany fell way behind those born in West Germany. And now with reunification and economic convergence IQ scores have converged too. Genetics may play a fairly large role in explaining outcomes for individuals, but if it affects measurable differences in outcomes between large groups the effect is small enough to be drowned out by other factors.

Its a long-known, well-demonstrated effect on an individual level. A more likely causal mechanism than the kind of broad social effect in the reverse direction that you suggest is that access to quality education also affects experience with, patience with, and attitudes toward the kind of assessments used to measure IQ, thus influencing results.

(There's other wealth-related -- and, on a country-level, social-support-network related effects, too, as early childhood nutrition and other social-environment factors have an effect on IQ that, again, has been demonstrated on an individual level.)

> No, not necessarily the richer - I can spot a direct correlation between the quality of the education system and the IQ reported here.

The first sentence of the linked post, well before the actual maps, is "Please note: Quality of education greatly influences IQ scores, i.e. a lower average IQ is more indicative of lower access to wide-scale quality education rather than innate intelligence (as demonstrated, for example, here). Also, testing conditions influence results; wealthier countries are more likely to be able to afford better testing conditions for participants."

However, whether we like to admit it or not, intelligence is partially affected by training (i.e. education).

It's impossible to measure some pure form of "genetical intelligence". After all, this is the best strength of our species: we're born idiots, and we learn our way up. We adapt.

More than intelligence, performance on tests is affected by training.

and lower in the ones with more recent wars too.

Why do you think the richer countries are richer?

Geography, accidents of history and positive feedback loops?

A higher IQ population will be better able to adapt and take advantage of other factors.

In addition to all the standard arguments against IQ (tailor at specific culture circles, can be taught and improved upon, measures subset of intelligence) it is important to remember that this is a sample of people who took the test, which by and far may not be a representative sample of a population. Most probably it's upwardly biased, but one has no way of knowing if the bias is constant across countries, which in turn makes any inference impossible.

How can you get an accurate average IQ? I've never taken a formal IQ test in my life. I asked serval people around if they have and no one else has in their adult life.

I have taken a couple taken cold as I was diagnosed with dyslexia as a child and later on my employer got me some extra tutoring for adult dyslexics.

I think my best was in the mid 130's or as I put it high enough :-)

I have, in high school we were all asked to take a test.

Did you go to a public school? Most tests I remember taking were the ITBS, PSAT, SAT and ACT.

http://en.wikipedia.org/wiki/List_of_standardized_tests_in_t...

Yes however this was in Europe.

Romania to be specific.

It wasn't mandatory or anything just encouraged.

What? There are tons of IQ tests online. I took one and found I had an IQ of 185. And for only $25 I could have gotten a certificate stating this fact. I'm sure that these tests are highly accurate. After all, who wouldn't want to think they are a genius?

When I was young, as part of my move into the public school system (from a private school in my first and second grades), I had to take one. Later, I was identified by a gifted program and had to take an entire battery of tests for most of the rest of my schooling, including numerous IQ tests. In those circles, I didn't know anybody who had never taken an IQ test.

Interesting perhaps I took one but didnt realize it. How accurate can an IQ test for elementary schoolers be for an adult population?

Maybe the test I took were not called IQ tests but were very similar. Most test I took were similar to what the SAT was, but I never got back an "IQ" score. Usually they gave us back a percentile score against the whole pool, but never an IQ number or range.

Yeah, I took a bunch of those as well, like the ITBS. They're more aptitude tests than IQ tests, but I imagine there some correlation. Usually they're looking for specific subjects or learning styles

There's a bunch of different IQ tests as well (WISC, SB5, DAS, WJ) and some designed for adults, but they should give you a score. There's some special tests for people with high percentile IQs as well.

I don't know if I'd call any IQ test accurate. Mine have varied wildly over time and depending on the test.

> How can you get an accurate average IQ?

The linked source has both maps based on only IQ data and maps based on IQ data adjusted with additional data from other assessments that have a demonstrated correlation to IQ results.

I dont find it very accurate to adjust an hard score based on other assessments, presumably those test arent IQ scores. Did every country take the same test?

Maybe it's just me, but I would find those numbers much more interesting if they were accompanied by a measure of spread of the IQ distribution.

The blog post author did a terrible job of research on IQ by encountering the writings of Richard Lynn (funded by the Pioneer Fund, and later an administrator of the Pioneer Fund) and not finding the writings of James R. Flynn,[1] the acknowledged expert on IQ trends over time, after whom the Flynn effect[2] is named. Flynn's TED talk[3] is the place to start for understanding what is really going on in IQ score trend and in national average IQ differences to date.

By contrast, Lynn is decried as a sloppy scholar,[4] and is not taken seriously by the mainstream of human intelligence researchers.

[1] http://www.amazon.com/What-Is-Intelligence-Flynn-ebook/dp/B0...

http://www.amazon.com/Are-Getting-Smarter-James-Flynn-ebook/...

http://www.amazon.com/Intelligence-Human-Progress-Story-Hidd...

[2]

http://www.nytimes.com/2012/11/04/education/edlife/james-r-f...

http://www.theguardian.com/technology/2012/sep/23/james-flyn...

http://www.timeshighereducation.co.uk/books/what-is-intellig...

[3] http://www.ted.com/talks/james_flynn_why_our_iq_levels_are_h...

[4] http://wicherts.socsci.uva.nl/wichertsRavenAfr2010rej.pdf

http://www.nybooks.com/articles/archives/1994/dec/01/the-tai...

Excellent critique, Sir.

I thought Unz's criticisms of Lynn were an excellent blend of accessible and thorough.

http://www.theamericanconservative.com/articles/race-iq-and-...

Is the difference between say 97 and 101 IQ really significant?

I'd attribute the scores that are lower than 90 (or maybe 95, or 98, I don't know) directly to lackings regarding national education, but does fluctuation among the others really mean something?

With a large enough sample size small differences can be significant. It's bad practice to display a more precise IQ number than your confidence intervals, so I'll assume the differences are real.

its in theory a normal distribution Two standard deviations include 95% of the population

I wonder why people focus so much on IQ instead of better metrics.

In this article, "The play deficit":

http://aeon.co/magazine/being-human/children-today-are-suffe...

there are these very interesting sentences:

"Other research, by the psychologist Mark Runco and colleagues at the Torrance Creativity Center at the University of Georgia, shows that scores on the TTCT are the best childhood predictors we have of future real-world achievements. They are better predictors than IQ, high-school grades, or peer judgments of who will achieve the most."

TTCT is "Torrance Tests of Creative Thinking".

People should really be much more interested in metrics like TTCT, rather than IQ tests, SAT tests or other not very relevant metrics.

Probably because general intelligence is "the best-established, most predictive, most heritable mental trait ever discovered in psychology." (Geoffrey Miller, Spent)

I googled it and found the following:

http://www.barrelstrength.com/2013/12/31/ramachandran-iq/

where there is:

"Intelligence (says Prof. Miller at page 189 of Spent) is positively correlated to:

    brain size
    speed of performing basic sensory motor tasks (“reaction time is a factor”, as the cop said in Blade Runner)
    height
    symmetry of face and body
    semen quality (!)
    health, physical and mental
    longevity
    sexual attractiveness for long term relationships"

So it is most predictive of many things, but maybe not of future real-world achievements...

Interesting how you'll never meet a person whose IQ is less than 100 (or even 120) on the Internet. It's a rather flawed metric.

Have you tried twitter?

I was going to suggest youtube comments ..

The point is, nobody would ever admit to having low (even above average) IQ, not that they are actually smart.

Isn't the average IQ by definition 100?

The overall average, sure. But just as that doesn't mean every single person's IQ is 100, it doesn't mean every country's average is the same.

IQ as a measure is like comparing computers on speed of the CPU. It is only one measure that defines us and with that many aspects make us what we are.

This is a map of how well someone can use English, IQ tests are biased towards that.

IQ tests are usually done using the person's first language, not English. What you are saying is only true if English tests are used for non-English-speaking people.

>Tip: Did you know that “iron” is pronounced as “I earn”, not as “I Ron”?

Who pronounces those the same? And who ever said it was the "correct" way?

Relevant: http://www.paulgraham.com/say.html

Also relevant: http://48laws-of-power.blogspot.co.uk/2011/05/law-38-think-a...

The study might be true, but there's little benefit to us discussing it right now.

Why? Are you worried it might bring forward discussions about race superiority and that kind of crap?

In my opinion, this study can start a reasonable discourse on the difference between education systems.

Not worried, it's just a waste of time.

Reply on news.ycombinator.com