I agree with (1), but would claim that it also selectively attracts introverts (and I’m unsure whether or not it will bias J-P to the P side)
(2) For each of these, I tried not to look at the data after finding the poll. I made predictions first. Just for fun / to correct for hindsight bias, anyone reading might want to do the same. To play, don’t click on the link or read my prediction until you make yours. Also, here is some data which claims to represent the general population—http://mbtitruths.blogspot.com/2011/02/real-statistics.html for comparison. I’ve already seen similar data on another site, so I won’t state my predictions on this one.
A website posts stats for people who have taken the test. Unlike the above simple random sample, this selects for internet users.
Prediction: I’d consider this “shallow internet”, so very weak biases to (I). The general population is (S), I’d expect a weak bias to (N) but not enough to overcome the general population’s S centering.
Result: apparently I suck at predictions. In hindsight all the top three would be predicted score high “Fi” on a Jungian cognitive function test, and Fi in theory would be more interested in taking personality tests. But that’s hindsight, and I’m not sure if connection between MBTI and Jung hasn’t been verified empirically.
Here is a “deep internet” forum that I wouldn’t ever visit… Christian singles chat forum! This should not suffer from the sample bias you mentioned earlier (He stated that websites I visit are likely to have users with similar personalities to me [ENTP])
Prediction: I tried my best not to look at the data despite the high visual salience as soon as you open that link. Here is my prediction: I’d predict strong biases towards Introversion (because internet), slight biases towards iNtuition (because religion is idea-based), moderate bias to Feeling (I think religious people are illogical) and … let’s say a slight tilt towards Judging. Call it a hunch, life experience says that Si (judging + sensing) is particularly predisposed to religion.
Result: OK, looks like my trends were right but my magnitude was way off. My “hunch” was correct but I didn’t listen to it closely enough and vastly underestimated the Judging bias, while my personal prejudice overestimated the Feeling bias. My predictions about intution and introversion were essentially correct though.
Prediction- Strong N, slight T bias. I don’t think T actually means “intelligent” as I define it, but I do think it would help on some portions of the IQ test.
Prediction—strong N, moderate T. I’m not sure about J-P. I think people who choose science tracks and go into academia will be P (creative types), whereas kids who get good grades but ultimately do not choose science will be J. I’m not sure which group they are looking at (I didn’t permit myself to read it yet, so I’m a bit vague on what exactly they did). I don’t think E—I will matter at all.
Result -
NT take high level science a lot more, Introverts take them slightly more. J-P is irrelevant.
Intuition really helps in school at all levels.
Feeling relates to high GPA in the easy courses but not the hard course (that’s pretty unexpected)
Introversion relates to high GPA in hard course but not in easy courses.
Percievers start out with a pretty big edge both in IQ and GPA in the lower level courses, Judging takes a slight lead in both those metrics in the advanced course. Not sure if this is noise.
Side finding—they also did IQ measurements. Again, only N related to IQ (in fact, F won out over T)...but it did not relate as much in the advanced courses. I think the advanced course chopped off the lower end of the IQ bell curve, leaving only smart Sensors. By the way, Extroverts have an IQ edge, despite getting lower grades and not taking advanced courses as often.
Thoughts? I think in general my ideas about introversion not mattering for intelligence, but mattering a lot for internet use, bear out. Apparently Thinking doesn’t really matter either...which I sort of felt was true, but I didn’t actually expect the IQ test scores to agree with me on that. It might have to do with self reported vs actual use of logic.
Of course, we are looking at the center of the bell curve, whereas on LW we are (presumably) looking at the far right edge.
They say that they found IQ correlates with I, N, T, and P. However, they claim that were surprised about the “I” correlation, because a large number of other studies have found that E is positively correlated. They go on to talk about how different testing conditions might favor E vs I. Some interesting further reading in there...it seems that N only correlates on the verbal reasoning section,
I agree with (1), but would claim that it also selectively attracts introverts (and I’m unsure whether or not it will bias J-P to the P side)
(2) For each of these, I tried not to look at the data after finding the poll. I made predictions first. Just for fun / to correct for hindsight bias, anyone reading might want to do the same. To play, don’t click on the link or read my prediction until you make yours. Also, here is some data which claims to represent the general population—http://mbtitruths.blogspot.com/2011/02/real-statistics.html for comparison. I’ve already seen similar data on another site, so I won’t state my predictions on this one.
A website posts stats for people who have taken the test. Unlike the above simple random sample, this selects for internet users.
http://www.personalitypage.com/html/demographics.html
Prediction: I’d consider this “shallow internet”, so very weak biases to (I). The general population is (S), I’d expect a weak bias to (N) but not enough to overcome the general population’s S centering.
Result: apparently I suck at predictions. In hindsight all the top three would be predicted score high “Fi” on a Jungian cognitive function test, and Fi in theory would be more interested in taking personality tests. But that’s hindsight, and I’m not sure if connection between MBTI and Jung hasn’t been verified empirically.
Here is a “deep internet” forum that I wouldn’t ever visit… Christian singles chat forum! This should not suffer from the sample bias you mentioned earlier (He stated that websites I visit are likely to have users with similar personalities to me [ENTP])
http://christianchat.com/christian-singles-forum/34516-meyers-briggs-type-indicator-mbti-poll.html
Prediction: I tried my best not to look at the data despite the high visual salience as soon as you open that link. Here is my prediction: I’d predict strong biases towards Introversion (because internet), slight biases towards iNtuition (because religion is idea-based), moderate bias to Feeling (I think religious people are illogical) and … let’s say a slight tilt towards Judging. Call it a hunch, life experience says that Si (judging + sensing) is particularly predisposed to religion.
Result: OK, looks like my trends were right but my magnitude was way off. My “hunch” was correct but I didn’t listen to it closely enough and vastly underestimated the Judging bias, while my personal prejudice overestimated the Feeling bias. My predictions about intution and introversion were essentially correct though.
http://personalitycafe.com/myers-briggs-forum/28171-mbti-demographics.html
Click the ppt, it has data by education.
Prediction: NT’s pursue higher education, SF’s do not. Other two dichotomies don’t matter as much, but J helps slightly.
Result: seems about right. Eyeballing, J seems not to matter much until college, at which point it prevents dropping out.
For IQ—http://asm.sagepub.com/content/3/3/225.short
Prediction- Strong N, slight T bias. I don’t think T actually means “intelligent” as I define it, but I do think it would help on some portions of the IQ test.
Result: N bias only. interesting.
Finally, Scientific aptitude: http://www.amsciepub.com/doi/pdf/10.2466/pr0.1970.26.3.711
Prediction—strong N, moderate T. I’m not sure about J-P. I think people who choose science tracks and go into academia will be P (creative types), whereas kids who get good grades but ultimately do not choose science will be J. I’m not sure which group they are looking at (I didn’t permit myself to read it yet, so I’m a bit vague on what exactly they did). I don’t think E—I will matter at all.
Result -
NT take high level science a lot more, Introverts take them slightly more. J-P is irrelevant. Intuition really helps in school at all levels. Feeling relates to high GPA in the easy courses but not the hard course (that’s pretty unexpected) Introversion relates to high GPA in hard course but not in easy courses. Percievers start out with a pretty big edge both in IQ and GPA in the lower level courses, Judging takes a slight lead in both those metrics in the advanced course. Not sure if this is noise.
Side finding—they also did IQ measurements. Again, only N related to IQ (in fact, F won out over T)...but it did not relate as much in the advanced courses. I think the advanced course chopped off the lower end of the IQ bell curve, leaving only smart Sensors. By the way, Extroverts have an IQ edge, despite getting lower grades and not taking advanced courses as often.
Thoughts? I think in general my ideas about introversion not mattering for intelligence, but mattering a lot for internet use, bear out. Apparently Thinking doesn’t really matter either...which I sort of felt was true, but I didn’t actually expect the IQ test scores to agree with me on that. It might have to do with self reported vs actual use of logic.
Of course, we are looking at the center of the bell curve, whereas on LW we are (presumably) looking at the far right edge.
EDIT: here is another IQ one with bigger sample size. http://www.psytech.com/Research/Intelligence-2009-08-11.pdf
They say that they found IQ correlates with I, N, T, and P. However, they claim that were surprised about the “I” correlation, because a large number of other studies have found that E is positively correlated. They go on to talk about how different testing conditions might favor E vs I. Some interesting further reading in there...it seems that N only correlates on the verbal reasoning section,