• Vampire [any]
    hexagon
    ·
    edit-2
    9 months ago

    Chance would dictate that the participant would guess correctly 25% of the time. The results showed a 26.6% hit rate. This is already easily explainable by pure chance.

    "In a total of 6,000 trials, there were 1,559 hits (26.7%), significantly above the chance expectation of 25%" – the p value in that case is 0.041

    I won't be investigating your other sources, because there is no evidence

    "I will not look at the evidence because there is no evidence"..... thank you for conceding have a nice day.

    • The_Jewish_Cuban [he/him]
      ·
      edit-2
      9 months ago

      It's just literally impossible lol

      Magic isn't real. Your expectation for random people online to look through X amount of random studies reads the same as anti-vax garbage. Go talk to people in the field and publish and argue with them about this information if you really think it matters. Otherwise you're trying to convince a bunch of low information amateurs who don't really have the background for the conversation. Which frankly leads me to heavily doubt all that you have to say.

        • GarbageShoot [he/him]
          ·
          9 months ago

          You've had people be very rude to you in this thread, which I think is undeserved despite the fact that I also think your position is unequivocally false, but Jewish_Cuban here was being extremely reasonable in his critique of your approach.

          • Vampire [any]
            hexagon
            ·
            9 months ago

            your position is unequivocally false

            It's a question of evidence. We'll know that from the evidence

    • Abracadaniel [he/him]
      ·
      9 months ago

      26.7% really doesn't strike me as significant. Certainly not enough to be convincing of such an amazing claim on its own. It does mean the study is worth scrutinizing to rule out any issues.

      https://www.graphpad.com/guides/prism/latest/statistics/common_misinterpretation_of_a_p_value.htm

      I just ran a script to generate 6000 random numbers between 0 and 1, then count what proportion were below 0.25. One result was as low as 23.9%

      • Vampire [any]
        hexagon
        ·
        edit-2
        9 months ago

        I agree p=0.041 is not very conclusive.

        I'll also note that saying the evidence should be approached as an "amazing claim" is explicitly saying you bring a biased approach to the evidence.

        No one study is conclusive really, that's why we need meta-analyses

      • TRexBear
        ·
        9 months ago

        deleted by creator