- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hello,
Could anyone help clarify how the p-value is calculated for Pearson correlation coefficient using PROC CORR procedure? I know the null hypothesis is r=0, but not sure if it's calculated using 2-sided t-test or 1-sided F-test, or anything else.
I have looked through all materials and discussion threads posted online but haven't found the answer to my question.
Thanks a lot,
Erin
Accepted Solutions
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
The documentation for PROC CORR contains a details section on the Pearson correlation. The definition of the t statistic that is used to test for r=0 is specified at the bottom of the page:
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
two sided t although t**2 = F so it does not matter.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Thank you very much.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
The documentation for PROC CORR contains a details section on the Pearson correlation. The definition of the t statistic that is used to test for r=0 is specified at the bottom of the page:
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hello,
and thanks for the valuable information.
Does anyone know how to get the t-test values for testing the population correlation coefficient from PROC CORR, or other PROC for that matter? That is, the actual value of
I have saved the correlation coefficients (r) generated by PROC CORR and then coded the formula myself. But isn't there an easier way to do it?
Thanks.