On Wed, 22 Feb 2012 05:52:37 -0800, Marc Carter wrote:
>Hi, all --
>I fear this to be a really stupid question, but...
>
>In ANOVA, why are the squared deviations in the SS between groups scaled
>(multiplied) by _n_?  I was once told it was to weight them, but that somehow
>doesn't seem right.
>
>Can someone help?

Just to add to what others have said, take a look at the attached PDF
that is Table 15.3 from Glass & Hopkins (3rd ed; 1996).  It lays out
the components for a one-way independent groups ANOVA graphically
and numerically.   The numerator of the F ratio estimates the population
variance by using the variance error of the mean (i.e., the variance of
the sampling distribution of means; the square of the standard error
of the mean) and multiplying it by N -- remember that the variance
error of the mean is equal to the population variance divided by N.

-MIke Palij
New York University
[email protected]

---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=16212
or send a blank email to 
leave-16212-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Attachment: G&H-Fig15-03-ANOVA Layout.pdf
Description: Adobe PDF document

Reply via email to