www.statisticsdoc.com
Stephen Brand
Marco,
All other things being equal, decreasing the number of dependent variables
will increase the power of the MANOVA, because each dependent variable uses
one degree of freedom. The problem is that "all other things being equal"
covers a lot of assumptions. Power can be affected by the intercorrelations
between the dependent variables that are remaining, and their effect sizes
(see Cole et al (1994) in Psychological Bulletin, Vol 115, 465-474 ). Also,
check to see whether you satisfy the assumptions of MANOVA with the full set
of dependent variables as well as the subset that you might retain.
HTH,
Stephen Brand
For personalized and professional consultation in statistics and research
design, visit
www.statisticsdoc.com
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]]On Behalf Of
Marco van de Ven
Sent: Monday, November 27, 2006 2:07 PM
To:
[hidden email]
Subject: MANOVA question
I am wondering whether there is a decrease of statistical Power if one
carries out a MANOVA with many dependent variables compared to just a few.
Or are no automatic Bonferroni adjustments carried out for MANOVA? The
reason for asking this is the following...I have devised a new testing
method for one of my experiments, which decreases the number of dependent
variables from 6-8 to maximally 3. Would this increase the Power of my
testing?
MV
Send instant messages to your online friends
http://uk.messenger.yahoo.com
Locked
