Nassim Taleb has been criticizing IQ tests lately. In this nice short blog, @JamesPsychol shows that everything Taleb claims is either incoherent, or factually wrong, or pointlessly ad hominem, or already addressed by psychometricians decades ago.http://www.unz.com/jthompson/swanning-about-fooled-by-algebra/ …
-
-
This Tweet is unavailable.
-
What you said makes no more sense than his argument did. Try to rephrase?
4 replies 0 retweets 15 likes -
Replying to @primalpoly @bonafidecustom and
Simply: IQ has no property of a measure. And you consider it as a measure. However, no matter how many times someone repeats this it will be of no avail to you and your peers. It is another domain of thinking.
4 replies 1 retweet 53 likes -
Replying to @PredragBrajovic @bonafidecustom and
It's a great measure, it's the best one we have in the behavioral sciences, and it's proven useful in dozens of domains. Sorry you've drunk his kool-aid. If you want to wake up from his cult, read the
@StuartJRitchie book.9 replies 0 retweets 42 likes -
Replying to @primalpoly @PredragBrajovic and
The mathematical argument is that the correlation is only for low IQ, and vanishes for high IQ. But the "average" correlation is like .5. So having high IQ has 0 correlation with, say, high SAT. Therefore, it only works for low IQ. Where correlation is actually 0.8
3 replies 2 retweets 13 likes -
Replying to @wall_sd @primalpoly and
This is stupid. The blog post at the top shows plenty of examples where higher IQ matters after 100 IQ. There's even one that shows increased chances of various outcomes at the top 1% of ability. I can't believe people ate up Taleb's made-up graphs.
1 reply 0 retweets 5 likes -
Replying to @amiguello1 @wall_sd and
Just because having higher IQ 'increases your chances' of certain outcomes does not mean that you can actually predict outcomes from IQ. If I tell you that a particular person has tested at 160 IQ, but no other info, can you tell me how successful they are?
2 replies 1 retweet 3 likes -
Replying to @thepiclord @amiguello1 and
If there are a lot of high IQ high performers, but also a lot of High IQ low performers, and a LOT of high IQ mediocre performers, then IQ is NOT actually telling you much about an individual person's performance. As in it isn't actually *explaining* their success in full.
3 replies 1 retweet 5 likes
You don't understand how correlation works or why it's useful.
-
-
Replying to @primalpoly @amiguello1 and
Nope. Understand both. Also have extreme respect for its LIMITS. Which is really where the debate lies.
1 reply 0 retweets 3 likes -
Replying to @thepiclord @primalpoly and
The limits of the predictive power of a correlation are shown by the coefficient itself. It is the improvement over guessing which gives it its utility. Your examples show that you have not digested that point.
1 reply 0 retweets 1 like - 2 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.