| Date: | Mon, 12 Aug 2002 13:02:04 +0100 |
|---|
| Reply-To: | Simone Leber <leber@stud.uni-frankfurt.de> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
|---|
| From: | Simone Leber <leber@stud.uni-frankfurt.de> |
|---|
| Subject: | Welch-James and Brown-Forsythe |
|---|
| Content-Type: | text/plain; charset="iso-8859-1" |
|---|
Hi Marta,
thanks so much for your help, it works fine and seems to be the right thing
to do, even if it's not the Welch-James Test. I was really just looking for
something to substitute the 2way ANOVA in the case of unequal variances, and
Brown-Forsythe seems to do the trick as well. So thanks for that!
The reference I've got for the Welch-James procedure is Algina, J. &
Olejnik, F.(1984). Implementing the Welch-Lames procedure with factorial
designs. Educational and Psychological Measurement, 44, 39-48.
Good luck
Simone
Simone Leber
2F2, 8 Thistle Place
Edinburgh
EH11 1JH
Scotland, U.K.
++44/ 131/ 228 6471
email: leber@stud.uni-frankfurt.de
|
