Help with point biserial correlation!

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Help with point biserial correlation!
Message posted by sue (via 64.44.78.50) on April 30, 2001 at 3:51 PM (ET)

I'm confused about this equation. To calculate the variance in my statistics class we have been summing all the squares of the scores, divided by N, then subtracting the mean squared. However, we learned that this number needs to be "adjusted" by multipling by N/N-1. So I do not know which standard deviation score to use for this point biserial equation. Is it the square root of the original variance, or the square root of the corrected variance? Thank you for any clarification!

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Help with point biserial correlation!
Message posted by Bill (via 192.231.71.108) on May 1, 2001 at 6:06 PM (ET)

I can't help with which quanity to use. But, the point-biserial correlation is simply a Pearson correlation when the dichotomous variable is coded as 0,1.

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.