What prompted this wishlist request?
I've seen the misuse of "each pair, student 's t -test" when doing analysis of variance in JMP by people who don't know much about multiple comparisons. (When selecting a test for means) I've heard of the term t -test, but I've never heard of other tests in statistics classes, so I don't use them...
What improvements would you like to see?
In future updates, we will use "each pair, student 's t -test" when conducting analysis of variance.
I would like it to be removed from the default settings or made a hidden option.
Why is this idea important?
This makes it less likely that results from analysis of variance will be misinterpreted.
Thank you for posting, BicycleRider. As you stated, the t-test is the "classic" statistical test for comparing means and is often misused in situations for multiple comparisons of more than 3 groups. We will discuss internally potential options to better illustrate that a multiple comparisons method using family-wise error rate correction should be employed for comparing more than 3 groups.
