• skittle07crusher@sh.itjust.works
    link
    fedilink
    arrow-up
    1
    ·
    7 months ago

    These kinds of well-needed comments full of doubts and questions of what all was controlled for in the research and whether confounding variables remained, the kinds that always come up on reddit and now (hooray!!) lemmy, make me wonder whether research could in some cases be dramatically improved by letting the internet loose on the research hypothesis ahead of time instead of once the paper is published.

    Scientists: “In our study, we will evaluate whether a is correlated with an increase in b. We will control for w, y, x, and z.

    The internet/reddit/lemmy: “You absolute imbeciles. Did it not occur to you to control for α, β, and gamma through omega?!

    Scientists: well, we will certainly consider all those and do our best to do so now!

    • AmidFuror@fedia.io
      link
      fedilink
      arrow-up
      2
      ·
      7 months ago

      Like everything on the internet, that feedback would quickly devolve into political bickering.

      There’s a lot of low quality science that gets done out there. Scientists are pretty good at spotting it, but they can’t do much to stop it. The real problem is science reporting misleading the public about what the current state of the science is.