Date:         Thu, 10 Jul 1997 23:00:32 -0500
Reply-To:     heft@swbell.net
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         Laura and Bill Heft <heft@SWBELL.NET>
Organization: Southwestern Bell Internet Services
Subject:      Help: What does SPSS do for an ANOVA with unequal sample sizes
Content-Type: text/plain; charset=us-ascii

I am trying to test for difference between 2 independent groups on
several dependent variables.  The groups have substantial differences in
sample size. I know that I could use t-tests, ANOVA, or GLM to look for
mean differences, but I don't know what SPSS does to adjust for unequal
n. I think I need to use a weighted or pooled average variance estimate.
My question is, does SPSS provide this automatically with unequal n, or
is there a way to indicate that I want a weighted variance estimate?  My
variances appear to be homogeneous (non-significant results from
Levene's test for equality of variance).  Any ideas???  Thanks, Laura
Heft