• Got_Bent@lemmy.world
    link
    fedilink
    arrow-up
    23
    ·
    1 year ago

    It’s been quite some time since I took stats, but wouldn’t six standard deviations put it in the true outlier category? If I’m even twenty percent correct in what I’m trying to communicate, that’s frightening.

    • coffeeguy@lemmy.world
      link
      fedilink
      arrow-up
      27
      arrow-down
      1
      ·
      edit-2
      1 year ago

      Person who does lots of stats checking in. That’s a good question. We usually refer to Sigma (1-sigma, 2-sigma… 6-sigma) as the probability that an observation could occur by random chance.

      The probability of 6-sigma occurring by random chance is about 1:1-billion.

      So you’re definitely right to characterize it as an outlier. In terms of sea-ice this means that based on our observations of ice extent recorded going back to 1989 (based on the image) it is extremely unlikely we would expect to observe a sea ice extent so far below the norm suggesting something else (climate change) explains the deviation.

    • Dagnet@lemmy.world
      link
      fedilink
      English
      arrow-up
      22
      arrow-down
      1
      ·
      1 year ago

      Been a while since I studied standard deviations but I remember 2 being already an outlier, 6 is a lot of deviation

      • ivanafterall@kbin.social
        link
        fedilink
        arrow-up
        7
        arrow-down
        3
        ·
        1 year ago

        I never took statistics, but 6 is at least triple the deviations of 2, and probably even more for math reasons I neither will nor can get into.

    • Seraph@kbin.social
      link
      fedilink
      arrow-up
      12
      arrow-down
      2
      ·
      1 year ago

      Correct, outlier until it’s repeated next year and the following. Then we can statistically say we’re fucked.