On Fri, 2 Feb 2001, Harry Avis wrote:
> Remeber that a correlation of .50 means that 25% of the variance is a
> population is caused by the variable selected ( Variance = correlation
> squared)
Curiously enough, this is not true for twin study correlations.
In this case a correlation of .50 means 50% of the variance
explained. I've tried on a number of occasions to understand why
this is so, but generally failed to follow the explanation.
For anyone who wants to try, though, Plomin (1990) discusses this
point, in a box titled "Interpreting correlations" on p. 43 of
his book _Nature and Nurture_. Bouchard et al, in a letter to
_Science_ 250, 1990, p. 1498) titled "when kin correlations are
not squared" has a go at the issue as well.
So I just believe it.
-Stephen
------------------------------------------------------------------------
Stephen Black, Ph.D. tel: (819) 822-9600 ext 2470
Department of Psychology fax: (819) 822-9661
Bishop's University e-mail: [EMAIL PROTECTED]
Lennoxville, QC
J1M 1Z7
Canada Department web page at http://www.ubishops.ca/ccc/div/soc/psy
Check out TIPS listserv for teachers of psychology at:
http://www.frostburg.edu/dept/psyc/southerly/tips/
------------------------------------------------------------------------
