• lime!@feddit.nu
    link
    fedilink
    arrow-up
    45
    ·
    4 days ago

    i mean, iq is a normal distribution. it caps at 200. 160 represents the 99.996th percentile, and above that the error bars are so large that the result is uneless.

    • Pommes_für_dein_Balg@feddit.org
      link
      fedilink
      arrow-up
      52
      ·
      edit-2
      4 days ago

      It’s not capped, really. But the claimed 250 IQ would be 10 standard deviations from the mean, so he’d be the most intelligent person in a population of ~1024 people.
      ~1011 humans have ever lived on earth.

    • Telemachus93@slrpnk.net
      link
      fedilink
      arrow-up
      12
      ·
      4 days ago

      Strictly speaking, a normal distribution doesn’t cap, neither at 0 nor at 200. Maybe the scores achievable by standardized tests do, of course.

      • lime!@feddit.nu
        link
        fedilink
        arrow-up
        7
        ·
        4 days ago

        they usually cap at 150. but yeah it’s not a hard cap, it’s an asymptotic curve. statistically the chance of getting 201 or higher is the same as getting -1 or lower.

    • taiyang@lemmy.world
      link
      fedilink
      arrow-up
      2
      ·
      4 days ago

      I was going to mention the same thing. Even the “smartest man alive” would be in those useless upper bounds.

      To explain that upper bounds issue to others, imagine being the top score on a leaderboard. Some of that’s going to be random chance and other factors, even if most of the time you score in the top 1% of scores consistently.