Check out Tabachnick & Fidell (Using Multivariate Statistics): In short, it is necessary to have more cases than DVs in every cell. In addition, power is reduced unless there are more cases than DVs in every cell because of reduced df for error.

Scott Millis ~~~~~~~~~~~ "Kunst ist schön, macht aber viel Arbeit."

Scott R Millis, PhD, ABPP (CN,CL,RP), CStat, CSci Professor & Director of Research Dept of Physical Medicine & Rehabilitation Dept of Emergency Medicine Wayne State University School of Medicine 261 Mack Blvd Detroit, MI 48201 Email: smillis@med.wayne.edu Tel: 313-993-8085 Fax: 313-966-7682

--- On Thu, 10/8/09, Salbod, Mr. Stephen <ssalbod@pace.edu> wrote:

> From: Salbod, Mr. Stephen <ssalbod@pace.edu> > Subject: MANOVA with small sample size and unequal ns > To: SPSSX-L@LISTSERV.UGA.EDU > Date: Thursday, October 8, 2009, 7:39 AM > > > > > > > > > > > > > > > > Dear Friends, > >                > I’ve > been asked to do  one-way (3 levels) MANOVAs where the > group sizes are n=25, > n=7, and n=6; and, the  number of dependent variables > can range from 6 to > 12.  I’ve read somewhere that it is > inappropriate to do MANOVA in > this situation (small ns + unequal ns + large DV set > relative to the ns). Does > any know a reference addressing this problem? > >                > Any > help will be greatly appreciated. > >  > > Stephen Salbod, Pace University, > NYC > >  > >                > > >                >    > > > > > > >

===================== To manage your subscription to SPSSX-L, send a message to LISTSERV@LISTSERV.UGA.EDU (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD