Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: Point Biserial Correlation Calculation
From
"Baur, Joshua" <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: Point Biserial Correlation Calculation
Date
Fri, 25 Feb 2011 15:31:14 -0800
Any suggestions on the following question will be appreciated:
In doing a point biserial correlation calculation following t-tests, the calculation procedure is to include dfs in the denominator. But if one does a ttest with unequal variances (after running the sdtest command to check), the Satterthwaites dfs are much lower than the dfs for a ttest
assuming equal variances. In calculating the point biserial correlation for a ttest in which equal variance is not assumed, is it appropriate to use Satterthwaites dfs? Or would you use the dfs as if equal variance WAS assumed. That is, the Satt. dfs are much smaller than the dfs for a ttest with equal variance, so the point biserial comes out much larger using Satt. dfs. In fact, the point biserial is nearly twice as large.
Any suggestions on what is the "correct" calculation of point biserial correlation for groups with unequal variance?
thanks.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/