Of course, you could continue to add additional parameters that affect
power, such as the relative efficiency of the estimator that you employ, the
correlations between samples in nonindependent samples designs, the exact
shape of the distributions from which the samples were drawn, the
reliability of your measurements, and so on.
----- Original Message -----
From: "Christopher D. Green" <[EMAIL PROTECTED]>
To: "Teaching in the Psychological Sciences" <[EMAIL PROTECTED]>
Sent: Tuesday, November 12, 2002 4:58 PM
Subject: [Fwd: SPSS & Power]


Earlier today I wrote:

> Power is a funtion of two independent components: effects size and sample
size
> (suitably adjusted depending on the design).

This was, of course, not quite correct. Power is a function of *three*
independent
components: effect size, sample size, and *alpha* (the probability of making
a type I
error). Effect size and sample size usually turn up explicitly in the
formula. Alpha
usually turns up only in the table of non-central F that is used. Sorry for
the
inexactitude.

Regards,
--
Christopher D. Green
Department of Psychology
York University
Toronto, Ontario, Canada
M3J 1P3



---
You are currently subscribed to tips as: [EMAIL PROTECTED]
To unsubscribe send a blank email to [EMAIL PROTECTED]

Reply via email to