They all disproved your theory that IQ is even an accurate way to measure intelligence. I also saw no evidence of intelligence thresholds based on race.
I will provide an overview of precisely what IQ measures because this is something that most social science professors, much less NB posters, do not understand (they do not understand because they tend to be mathematically illiterate by the standards of university professors).
In the early 20th century an Englishman by the name of Charles Spearman, an extraordinarily mathematically gifted man who was originally an engineer but later made remarkable contributions to both statistics and psychology, made a common sense observation that everyone in high-school notices: that school grades tend to be correlated.
Spearman noticed that British school pupils who got good grades on Latin very often also attained good grades on mathematics and history. Likewise pupils who did poorly on one subject often did poorly on most of them. And he wondered, why? Why was it that performance on subjects as unrelated as mathematics and Latin were so correlated?
Spearman being the natural mathematician he was wanted to study this issue precisely, i.e. quantitively and mathematically, so he invented a statistical technique known as factor analysis to do so. This statistical technique essentially allows you to ‘extract’ a few ‘factors’ that explain the correlation among many variables such as many school subjects.
He then found that the correlations between all the subjects studied by school pupils could be compressed to just one factor which he named ‘g’ for the ‘general factor.’
An analogy that might help clarify this for you would be to see that, given the same amount of practice, people who are good at a few sports tend to be good at most sports while people who are bad at a few sports also tend to be bad at most sports. If you apply factor analysis to performance on sports you would extract a factor that could quantitatively explain the correlation between performance on all sports. This factor would be something like a general ‘athleticism’ and people who score high on ‘athleticism’ are good at most sports and the converse, that people who score low on ‘athleticism’ are bad at most sports, is also true.
Now you can precisely quantify a correlation between these general factors and individual subject scores. For those unfamiliar with statistical correlation – a correlation of 0 means two things aren’t related at all while a maximum correlation of 1.0 means that they are perfectly proportional (a straight-line on a graph). If you found that, for example, basketball correlates with ‘athleticism’ at 0.8 while cricket only correlates with it at 0.4 you could say that basketball is a better test of ‘athleticism’ than cricket and people who are good at basketball are more likely to be good at most sports than people who are merely good at cricket.
Now do you see the amazing utility of factor analysis? Factor analysis essentially allows you to quantify just how ‘good’ a test is at measuring the general factor. When it comes to school subjects, a subject like physics tends to have a high correlation with ‘g’ (we say that physics is highly g loaded) compared to something like art. And that’s not surprising at all, I’m sure you remember from your high-school days that the kids who were good at physics tended to be ‘smart’, i.e., they tended to do well academically in general whereas the same couldn’t be said as strongly for the artistic kids.
Well, it turned out that ‘g’ isn’t limited to school subjects. In fact it turns out that literally any battery of tests that require some form of thinking, for example tests of short term memory and how fast you can do arithmetic calculations (a test of mental processing speed) and tests of the ability to mentally rotate shapes in your head, will all show the same pattern of universal correlation between the individual tests just like school subjects. And if you apply factor analysis to them you will extract the single general factor.
And guess what?
The general factor extracted from school grades is very highly correlated with the general factor extracted from a battery of basic mental tests (tests of memory, mental speed etc).
Now we can finally define mathematically what a perfect or ideal IQ test is: it is an infinite number of mental tests (these tests can be anything that requires thinking whether a basic memory test or a physics exam). This is because the correlation between the general factor and the individual tests tends to 1 in the mathematical limit of N tests as N tends to infinity.
This is why good real world IQ tests that psychologists use are all batteries of individual tests – things like a test of vocabulary, abstract pattern recognition, mathematical reasoning etc. A real IQ test isn’t those fun jokes of IQ tests on the internet. A much better example of an IQ test than ‘internet IQ tests’ would be the American SAT or GRE or LSAT or MCAT.
Here is a fun fact for you: did you know that physics majors typically come out as the top LSAT scorers on average? Even though physics is all mathematics and the LSAT is a law-school aptitude test consisting of a battery of verbal thinking tests, physics majors do better than English literature majors. Why is that? Well just as in high-school – the kids who are good at physics tend to be ‘smarter’ than the kids who merely do well in English literature. And those physics majors go on to perform better in law-school and then become better lawyers, on average. Likewise physics majors outperform biology majors on medical school aptitude tests like the MCAT and GAMSAT.
And here is another fun fact for you, whether it is the SAT or GRE or LSAT or MCAT, East Asians outperform Whites who outperform Blacks.
Charles Spearman thought that ‘g’ was a measure of something like ‘mental horsepower’ and I think that is a good way of thinking about it. IQ tests measure your ability for reasoning in general whether that reasoning is about engineering or law or medicine. Intelligence is a much more vague term but if you think that engineering, law or medicine has something to do with it then I venture to say that IQ has something to do with intelligence.
