You can login if you already have an account or register by clicking the button below.
Registering is free and all you need is a username and password. We never ask you for your e-mail.
Never heard of absolute deviation until just now. How would absolute deviation work better? I can't even find the mean absolute deviation of IQ, so all I can do is approximate that it would be 15÷[√(π÷2)]≈12. That would just prove my point even further...
[–]rektumsempra[S]0 points
1 point
1 point
(+1|-0)
ago
(edited ago)
the problem here was that I thought IQ was normally distributed. I'm changing my stance now, after seeing how easily gigantism and dwarfism fuck up the distribution of human height, which might otherwise be normally distributed but is obviously not, since the tallest man in the world was 9.275 standard deviations out. Obviously we're not gonna get a 1 in 10 quintillion guy. I expect IQ to have confounding variables in the same way, making it not normally distributed. -.-
Wow. I just realized now that IQ isn't normally distributed... because I looked at the SD of male height, which is 4, but the tallest man ever was 107.1in. Without the phenomenon of gigantism, a person like that is 1 in 10 quintillion, so height isn't normally distributed either. Wtf is with these bogus standard deviations... Also "outlier", then, is synonymous with "non-normal distribution". There couldn't be outliers in a normal distribution with enough data points, and human height has more than enough.
view the rest of the comments →
[–] rektumsempra [S] ago
Never heard of absolute deviation until just now. How would absolute deviation work better? I can't even find the mean absolute deviation of IQ, so all I can do is approximate that it would be 15÷[√(π÷2)]≈12. That would just prove my point even further...
[–] dan_k ago
Maths only explains reality, not define it.
[–] rektumsempra [S] 0 points 1 point 1 point (+1|-0) ago (edited ago)
the problem here was that I thought IQ was normally distributed. I'm changing my stance now, after seeing how easily gigantism and dwarfism fuck up the distribution of human height, which might otherwise be normally distributed but is obviously not, since the tallest man in the world was 9.275 standard deviations out. Obviously we're not gonna get a 1 in 10 quintillion guy. I expect IQ to have confounding variables in the same way, making it not normally distributed. -.-
[–] urx ago
outlier.jpg
[–] rektumsempra [S] ago
Wow. I just realized now that IQ isn't normally distributed... because I looked at the SD of male height, which is 4, but the tallest man ever was 107.1in. Without the phenomenon of gigantism, a person like that is 1 in 10 quintillion, so height isn't normally distributed either. Wtf is with these bogus standard deviations... Also "outlier", then, is synonymous with "non-normal distribution". There couldn't be outliers in a normal distribution with enough data points, and human height has more than enough.