• perestroika@lemm.ee
    29·
    6 days ago

    Hypothesis A:

    • his name is J. Random Error

    Hypothesis B:

    • he’s real, has a brain anomaly, and is taking androgen antagonists (e.g. 5-alpha-reductase inhibitors) for breakfast, lunch and dinner. It’s not testosterone which has most biological effect, but dihydrotestosterone. T is just the guidance signal from the brain. DHT does the heavy lifting in tissue. If you don’t want to operate on the brain, you can block the result and leave the guidance signal going wrong.
  • Blackmist@feddit.ukEnglish
    24·
    6 days ago

    Smart Hulk.

    There’s probably severe health complications with that level of testosterone though.

  • Aeri@lemmy.world
    101·
    7 days ago

    So I’m not a data scientist but doesn’t this basically say that your testosterone and IQ are pretty independent of each other?

    • Gladaed@feddit.org
      61·
      6 days ago

      The depiction does not tell you anything since you cannot see structure in the black blob. It shows that outliers do exist which is expected since biology is messy.

      There is the slightest anti correlation for measured testosterone levels and IQ, but on the individual level this is meaningless. There are people with testosterone resistance to various levels which would require diligent treatment to destill a more useful graph, but that’s infeasible.

  • ZILtoid1991@lemmy.world
    2·
    5 days ago

    Could be someone with mild testosterone insensitivity, people like that often have elevated testosterone levels without the effects of elevated testosterone levels.

  • Bogus007@lemm.ee
    412·
    6 days ago

    So, because women have a much lower content of Testosterone, we can conclude that there are much more stupid than men or what does this silly correlation wants to tell us???

    • ILikeBoobies@lemmy.ca
      131·
      6 days ago

      It’s about men but your takeaway would be that women are smarter because as IQ goes up, Testosterone goes down

      • drascus@sh.itjust.works
        7·
        6 days ago

        this graph very clearly states it’s only related to men and has nothing to do with women at all. the only thing it shows is that the majority of men have average intelligence and average testosterone levels. R2=0.19 value means only 19% of the variation in testosterone levels can be explained by IQ, leaving 81% unexplained. This is a very weak relationship, bordering on statistical noise.

          • UnderpantsWeevil@lemmy.worldEnglish
            2·
            6 days ago

            The thing about very superficial analysis like this is you can just go fishing for correlations and come back with something meaningless but publishable.

            • Feathercrown@lemmy.worldEnglish
              1·
              6 days ago

              That exploitation of the scientific system is not what’s happening here. This is naive maybe, but not malicious, and it seems valid and without unwanted motives.

      • GoodEye8@lemm.eeEnglish
        21·
        6 days ago

        His takeaway is that women are as smart as men (as in the IQ is the same) but because men have higher testosterone holding them back men are actually smarter than women. What I think he’s actually trying to say is that he’s a high testosterone misogynist.

        • drascus@sh.itjust.works
          2·
          6 days ago

          If that is someones takeaway they are wrong. This chart essentially shows that these things are not actually correlated.

          • GoodEye8@lemm.eeEnglish
            1·
            6 days ago

            Look. I don’t make the rules of misogyny. You want things to make sense, you need to look elsewhere.

        • Bogus007@lemm.ee
          2·
          6 days ago

          Your conclusion is based on deduction which in many occasions - like this one - leads to false conclusions. Try again.

    • UnderpantsWeevil@lemmy.worldEnglish
      6·
      6 days ago

      women have a much lower content of Testosterone, we can conclude that there are much more stupid than men

      This poster is High T. Ask me how I can tell.

      • Bogus007@lemm.ee
        13·
        6 days ago

        Understand the meaning of question marks, you do not? Much to learn, little Padawan, you still have.

    • drascus@sh.itjust.works
      2·
      6 days ago

      The only thing this graph shows is that average people are average. This graph essentially shows no -> very weak correlation between testosterone levels and IQ. in fact I would say this graph argues that correlating testosterone levels with IQ is essentially flawed. However because of the way the data is represented it makes it look as though there might be a correlation.

      • Bogus007@lemm.ee
        2·
        6 days ago

        AFAIR the R2 is (almost) equal to rho in the Pearson correlation. I just see two variables, a linear fit from - possibly - an OLS. The small R2 is likely due to the outlier (though a single outlier by this mass of points raises my eyebrows as the MSE (or take the RSME) won’t be affected as such by a single point when there are 15’000 points centered closely around an estimate, but CCV would tell) and R2 says nothing about the p-value, which is determined by the amount of information in a system/about variables, and hence likely way below 0.05.

        This relationship aka in this case correlation says pretty much nothing about real world, because IQ is (possibly) not only determined by IQ, but way many other factors. The picture is utterly simplified. It is similar to the relationship between the number of babies and the number of storks.