| Date: | Tue, 5 Nov 1996 13:09:22 -0500 |
|---|
| Reply-To: | "William B. Ware" <wbware@EMAIL.UNC.EDU> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU> |
|---|
| From: | "William B. Ware" <wbware@EMAIL.UNC.EDU> |
|---|
| Subject: | Re: Adjustment factors when using two-factor experiments with
repeated measures. |
|---|
|
| In-Reply-To: | <96Nov5.123136-0500_est.7220-59108+78@email.unc.edu> |
|---|
Are you referring to adjusted F-statistics, or adjusted F-values. The use
of the F-distribution (F-values) in repeated measures designs is
predicated on some rather restrictive assumptions regarding the nature of
the variance-covariance matrix of the repeated measures. If that
assumption is violated, then one may adjust the df using the results of
Geisser and Greenhouse. This adjustment assumes a "maximum" violation.
However, one can estimate the degree to which the assumption is violated
and calculate a "constant" that is used to adjust the df for that
particular situation. This "exact" adjustment is available in the SPSS
Manova procedure using either of the two "epsilons".
______________________________________________________________________________
William B. Ware, Professor and Chair Educational Psychology
CB# 3500 EMAIL: wbware@unc.edu
University of North Carolina PHONE: (919)-966-5266
Chapel Hill, NC 27599-3500 FAX: (919)-962-1533
URL:http://www.unc.edu/~wbware/
______________________________________________________________________________
On Tue, 5 Nov 1996, Mike Broyles wrote:
> I note that adjusted F-values are sometimes used in repeated measures
> when certain conditions are not met. How does one determine as to when
> adjusted values should be used?
>
> Mike Broyles
>
| 
