• whotookkarl@lemmy.world
      link
      fedilink
      English
      arrow-up
      1
      ·
      2 months ago

      Selection bias, Self selection bias, reward bias, bias in the questions and how they are presented, etc. You have to really know some stuff to not have a fucked survey result and even then some of those biases will still be present. If they don’t even bother with error % boundaries it’s probably going to have some other problems too.