2 ms·

Unfortunately, there's no good way to say that a p > 0.05 is a failure to reject the null hypothesis (which does not imply the null hypothesis is correct) witho

Unfortunately, there's no good way to say that a p > 0.05 is a failure to reject the null hypothesis (which does not imply the null hypothesis is correct) without making nonstatistican readers bored. Statistical writing is hard.

wrs 3y ago
“Bloomberg didn’t show that ChatGPT is racist” is probably the best you can do for the headline. (They didn’t do that either.)
- dang 3y ago
  Ok, we've put that in the title above. Thanks! If someone has a better (i.e. more accurate and neutral) title to suggest, we can change it again.