CronoDAS comments on Open Thread, November 1-15, 2012 - Less Wrong

4 Post author: OpenThreadGuy 02 November 2012 02:11AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (373)

You are viewing a single comment's thread.

Comment author: CronoDAS 03 November 2012 11:33:57PM 2 points [-]

If my IQ was measured as "really high" (140-ish) back when I was 8, but I'm now 30, should I expect the same kind of "really high" score if I took an IQ test today, as an adult?

Comment author: gwern 03 November 2012 11:36:28PM 6 points [-]

No. You should expect regression to the mean compounded with the higher variance of childhood scores and the decline of Gf beginning in the 20s. I don't know how much lower those would make one predict IQ at age 30, though.

Comment author: Epiphany 07 November 2012 05:29:16AM *  -1 points [-]

If your IQ was over the highest possible score on the test because they based that on an estimation calculated using your age (for a hypothetical example: you scored IQ 100 as a 2 year old, the equivalent of an average adult. You're much smarter than an average adult then, doing that at 2, so they'd give you a ridiculously high score.)

Unless this was the case, and it doesn't seem like it (it's not likely the test they gave you had a ceiling below 140) then you should expect a similar score. IQ is supposed to stay the same throughout one's life. Whether it actually does, I'm not sure. I haven't done research to confirm that IQ does what it is supposed to. But I can say that it is supposed to stay the same.

One problem you may run into is that if you take a different test it may give you a different score. This is less likely if your IQ is 100, and increasingly likely the higher your IQ is. That's because it's difficult to get the tests to behave properly for rare people due to not being able to find enough of them to test the test on.