Signifiance level of a correlation

Hi every one,
today i'm trying to find a way to compare 2 correlation to know if the difference between this 2 correlation is significant.
BUT, I can't find any Rscript or any function to make it on the web.
You may can help me to find a solution to this probleme?

Hello @Solpero,

Just to get more clarity. You want to compare two variables, derive a something like a Pearson's correlation coefficient and see if it is significant? Or do you actually want to compare two correlations? The latter wouldn't make any sense.

It was the second option... I don't understand why it's a non sense?

Surely, you'd rather want to see if there is a significant difference for measure A in the pre and post and then the same for measure B in the pre post and not the combination effect of the relationship between those? This will clearly highlight what has changed as it might be that B has changed and not necessarily A. You're going to have to provide data otherwise to make your case clear.

Hello,

I dived a bit into this. It seems your best bet would be to do a Fisher r to z conversion. Essentially, taking the correlation converting it into a z score and then you'll be able to test for significant differences.

See here: http://personality-project.org/r/psych/help/fisherz.html
And also here: https://www.rdocumentation.org/packages/DescTools/versions/0.99.38/topics/FisherZ

Ok thank you i'm gonna look how it can work!
I'll be back to say how it's going :slight_smile:

1 Like

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.