> If the research can't demonstrate the DK effect, then the research is bad, like poorly designed.
It’s the opposite problem. The research is bad because it “demonstrates the DK effect” even for cases where it cannot possibly be true. Like “people who tend to roll lower dice rolls overestimate their own ability to roll high dice values and people who tend to roll higher values tend to underestimate their own ability”
Which is just obviously incorrect and the true result should have been “there’s no correlation between people’s estimate of their ability to roll dice and their actual outcomes”
If you want to demonstrate an actual effect like DK, you have to set up an experiment such the if it truly doesn’t exist then the experiment tells you so. A great way of checking this is to throw random data at it in which case it should come out saying “I saw no correlation” but that’s no the case with their setup.
It’s the opposite problem. The research is bad because it “demonstrates the DK effect” even for cases where it cannot possibly be true. Like “people who tend to roll lower dice rolls overestimate their own ability to roll high dice values and people who tend to roll higher values tend to underestimate their own ability”
Which is just obviously incorrect and the true result should have been “there’s no correlation between people’s estimate of their ability to roll dice and their actual outcomes”
If you want to demonstrate an actual effect like DK, you have to set up an experiment such the if it truly doesn’t exist then the experiment tells you so. A great way of checking this is to throw random data at it in which case it should come out saying “I saw no correlation” but that’s no the case with their setup.