From: Martin Sherman <MSherman@loyola.edu>	
Reply-To: Martin Sherman <MSherman@loyola.edu>
To: SPSSX-L@listserv.uga.edu
Date: Jul 16, 2006 12:42 PM
Subject: Re: Levene-Test for related samples?
Reply | Reply to all | Forward | Print | Add sender to Contacts list | Delete this message | Report phishing | Show original | Message text garbled?
From David Howell's Stat Text.

 t = [(F-1)*Sqrt(n-2)]/[(2*Sqrt(F(1-r squared)))].

HTH,

df for this t is n-2.
F = largest variance divided by the smallest variance
r = correlation between the matched or pair scors.
Check t for sign. If it is not significant than you can safely conclude
that you have met the equal variances assumption for the dependent t
-test.