Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
Reporting Point-Biserial Correlation
Message posted by Diane (via 209.149.54.126) on May 24, 2001 at 2:12 PM (ET)
If I am using a point-biserial correlation can I treat it as I would a regular Pearson's R and use the R-sq to explain variance?
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: Reporting Point-Biserial Correlation
Message posted by Bill (via 192.231.71.108) on May 29, 2001 at 4:26 PM (ET)
The point-biserial is mathematicly the same as the Pearson. That is, if you perform a Pearson correlation using one variable (eg sex) coded 0,1 you get a point-biserial. I GUESS that you can square the resulting "r" and interpret it as usual.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.