http://spssx-discussion.165.s1.nabble.com/multi-vs-univariate-tests-in-GLM-RM-anova-tp5037385p5039918.html
From below, 1 Dec,
"But in RM Anova, the univariate
tests are not showing us effects separately for each measure. "
... "each measure"?
- In RM Anova, the construction of the problem and hypothesis is such that
you have only *one* measure, with multiple reports. Simple tests are done
on the sum of the reports, or perhaps on polynomial contrasts if reports are
ordered (like "time") and request them. Individual contrasts are unusual, and
not very interpretable if there is a trend-line.
From below, Nov 30,
"The multivariate tests show all within-subject effects as significant
but the
univariate tests show two of those effects as non-significant."
- I'm not sure which you are pointing to as "all within-subject effects", but you
ought to be looking at the tests for the between-subject factors. The simple case
of MANOVA is discriminant function; you do not ordinarily see or want to see a
"within-subject" test as to whether the predictor variables in a DF are have equal
means (which implies a matrix of paired tests, as followup) or have means equal
to zero.
A multivariate test is a test on the *pattern* among the variables. The pattern
can differ as a consequence on one or more components, or (merely) the
pattern among them, even with no variable being nominally significant. That is
one reason why the general MANOVA is a good choice for an overall test
performed with total agnosticism, but is a poor choice when you have prior
knowledge about the expected results, or if you want to draw careful
conclusions about any of the components.
--
Rich Ulrich
Date: Thu, 1 Dec 2011 04:27:40 +0000
From:
[hidden email]Subject: FW: multi vs univariate tests in GLM RM anova
To:
[hidden email]
hello, i received a couple answers to my post which made me realize i may not have explained my issue well. I understand the difference between the multivariate and univariate tests in the MANOVA (multiple dependent variables) but I
am not sure i understand it in the RM Anova (the same DV measured multiple times). In MANOVA, the multivariate tests tell us if there are effects across all DVs. Then the univariate tests show us the effects separately for each DV. But in RM Anova, the univariate
tests are not showing us effects separately for each measure. We still have exactly the same set of effects that include the within-subject factor as we have in the multivariate tests. So, how are they different?
bozena
On Wed, Nov 30, 2011 at 9:03 PM, Zdaniuk, Bozena
<[hidden email]> wrote:
Hello, i am running one within, two between subject RM ANOVA using GLM. The multivariate tests show all within-subject effects as significant but the univariate tests show two of those effects as non-significant. My intuition is to use the univariate tests
but I don't know exactly why (and whether it is what I should do). Why would there be such a difference between the two types of tests? My N is very large (70, 000)
thanks in advance for help.
[snip, other response]