One advantage of the vocabulary tests is that because they are crystallized and usually have no time limit they have a high reliability (they measure the underlying "vocabulary" factor better), I believe that if you correct for reliability the arithmetic tests should have a higher g loading than the vocabulary tests.
"The systematic and realiable differences in choice of distractors indicate that most subjects have been exposed to the word in some context, but have inferred the wrong meaning."
This is right some of the time. But the other reason why people choose the same consistent distractors is the presence of fake morphemes.
A morpheme is the smallest unit of language that carries meaning. For example, consider the word "unhappiness". The "un-" at the beginning of the word negates what comes after. Importantly, this morpheme consistently carries that meaning in whatever word it appears (which is what makes it a morpheme).
But sometimes, components of words get misinterpreted/mistaken for morphemes due to an etymological coincidence. In the examples that Jensen gave, I suspect that the "fat" in "fatuous" is causing people to think "fatuous" has something to do with size. And the "fec" in "fecund" is causing people think about "fecund" has something to do with fecal matter. But if you look it up, despite the superficial similarity in spelling, the pairs of words are etymologically unrelated.
What's important is that for this confusion to occur, it doesn't require for the person to have seen "fatuous" or "fecund" before, undercutting his argument that consistency of distraction implies that someone has seen the word before. (Though I overall agree with Jensen's explanation for why vocabulary is so g-loaded.)
There is now a fad about taking anti-aging suplements, driven by folks like Dr Sinclair ("Lifespan" author). They mention resveratrol, NMN and other suplements that may reverse aging, they do point out to papers like this one https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4854911/
Could you have a look from the statistical and sceptic pov to see if any of these might actually make sense?
Seems like a win for people who think IQ tests aren't valid, as I can imagine parents who make their kids read will lead to a boost in scores.
On the other hand, Jensen's reasoning seems sound. I see this all the time in London, not with language, but with geography. Some people who live in London, without explicitly trying to, just seem to know where different places in London are. You tell them a location, and they intuitively remember what part of London it's in, and what tube line goes there. It's strange, but seems to track well with Jensen's ideas around vocabulary.
But vocabulary shows contamination with age, Openness to Experience, and (modestly) the Flynn Effect. Don't get me wrong, I love vocabulary as an informal measure of intelligence, but shouldn't psychometricians rather be focusing on things like choice reaction time in IQ test construction to avoid confounding?
Maybe in the past vocabulary tests were the highest g-loaded, but nowadays people are exposed to less vocabulary than ever before. Nobody reads. You might as well just test kids on what internet lingo means, not obscure words.
Terence Tao had a math SAT score of 760 at the age of 8 (99% percentile) and a verbal score 290 (9% percentile). Verbal tests clearly do not work for children. There are no child prodigies that can write novels.
In an internet era with absolute dominance of the English language in every field, I would not be surprised if verbal scores in non-English speaking countries drop. Maybe you can tell us your personal experience. English education starts early in Scandinavian countries. If a lot of what you watch/read/write is in English, how can you increase your vocabulary in Danish or Swedish? And testing you in English vocabulary for IQ is obviously not reasonable as you're not a native speaker. If your daily interactions with other people are in Danish but your work and study is in a mix of Danish and English, I doubt that many people can be good at both.
Jensen really was one of the greats of the 20th century, I hope we'll realize that some day, collectively. Wouldn't hold my breath though.
One advantage of the vocabulary tests is that because they are crystallized and usually have no time limit they have a high reliability (they measure the underlying "vocabulary" factor better), I believe that if you correct for reliability the arithmetic tests should have a higher g loading than the vocabulary tests.
What are "pedigrees?"
"The systematic and realiable differences in choice of distractors indicate that most subjects have been exposed to the word in some context, but have inferred the wrong meaning."
This is right some of the time. But the other reason why people choose the same consistent distractors is the presence of fake morphemes.
A morpheme is the smallest unit of language that carries meaning. For example, consider the word "unhappiness". The "un-" at the beginning of the word negates what comes after. Importantly, this morpheme consistently carries that meaning in whatever word it appears (which is what makes it a morpheme).
But sometimes, components of words get misinterpreted/mistaken for morphemes due to an etymological coincidence. In the examples that Jensen gave, I suspect that the "fat" in "fatuous" is causing people to think "fatuous" has something to do with size. And the "fec" in "fecund" is causing people think about "fecund" has something to do with fecal matter. But if you look it up, despite the superficial similarity in spelling, the pairs of words are etymologically unrelated.
What's important is that for this confusion to occur, it doesn't require for the person to have seen "fatuous" or "fecund" before, undercutting his argument that consistency of distraction implies that someone has seen the word before. (Though I overall agree with Jensen's explanation for why vocabulary is so g-loaded.)
Spatial ability has lowest g loading. Is this consistent with male advantage in intelligence? There is even small female advantage in word knowledge.
https://www.semanticscholar.org/paper/Brother%E2%80%93sister-differences-in-the-g-factor-in-of-Deary-Irwing/7e8bdb5a9e5835d00c2cc51355f69d8919d8748d
This you?
https://rationalwiki.org/wiki/Scandza_Forum
Emil, a request.
There is now a fad about taking anti-aging suplements, driven by folks like Dr Sinclair ("Lifespan" author). They mention resveratrol, NMN and other suplements that may reverse aging, they do point out to papers like this one https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4854911/
Could you have a look from the statistical and sceptic pov to see if any of these might actually make sense?
Seems like a win for people who think IQ tests aren't valid, as I can imagine parents who make their kids read will lead to a boost in scores.
On the other hand, Jensen's reasoning seems sound. I see this all the time in London, not with language, but with geography. Some people who live in London, without explicitly trying to, just seem to know where different places in London are. You tell them a location, and they intuitively remember what part of London it's in, and what tube line goes there. It's strange, but seems to track well with Jensen's ideas around vocabulary.
But vocabulary shows contamination with age, Openness to Experience, and (modestly) the Flynn Effect. Don't get me wrong, I love vocabulary as an informal measure of intelligence, but shouldn't psychometricians rather be focusing on things like choice reaction time in IQ test construction to avoid confounding?
Maybe in the past vocabulary tests were the highest g-loaded, but nowadays people are exposed to less vocabulary than ever before. Nobody reads. You might as well just test kids on what internet lingo means, not obscure words.
Terence Tao had a math SAT score of 760 at the age of 8 (99% percentile) and a verbal score 290 (9% percentile). Verbal tests clearly do not work for children. There are no child prodigies that can write novels.
https://www.lesswrong.com/posts/5Xrv5aXfb95uHcseN/innate-mathematical-ability
In an internet era with absolute dominance of the English language in every field, I would not be surprised if verbal scores in non-English speaking countries drop. Maybe you can tell us your personal experience. English education starts early in Scandinavian countries. If a lot of what you watch/read/write is in English, how can you increase your vocabulary in Danish or Swedish? And testing you in English vocabulary for IQ is obviously not reasonable as you're not a native speaker. If your daily interactions with other people are in Danish but your work and study is in a mix of Danish and English, I doubt that many people can be good at both.