p-value for rank correlation?

It has been asked in the past(here and here), but I’d like to reiterate the request for a p-value statistic in linear correlation and rank correlation tests.

To be frank, all other statistics are meaningless if you don’t provide a p-value or confidence interval.

Hi @rosstripi - welcome to the forum and thanks for the feedback. I will pass this along to the developers and we’ll see what we can do about making our statistics nodes produce more robust outputs.

(I actually saw your tweet about this the other day and was wondering if you would post here :slight_smile: )

EDIT: for future reference, this is AP-10030 internally


Thank you!

Our team is in the process of pivoting from python/jupyter notebooks to Knime and this was one of the holdups I had about making the transition, so it will put me and I bet others at ease having those stats :smile:


This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.