Date: Mon, 3 Apr 2006 14:04:09 0600
ReplyTo: Matthew Pirritano <mpirrita@unm.edu>
Sender: "SPSSX(r) Discussion" <SPSSXL@LISTSERV.UGA.EDU>
From: Matthew Pirritano <mpirrita@unm.edu>
Subject: Re: Difference between MANOVA and RepeatedMeasures MANOVA
InReplyTo: <6250203B042D8349A920AA610383093601106726@UTHEVS3.mail.uthouston.edu>
ContentType: text/plain; charset=ISO88591; format=flowed
The variables are all in the same scale. Does that mean to Repeated
Measures would work?
I believe that in calculating the betweensubjects effect repeated
measures just compares the betweensubjects factor across the average of
the repeated measures. Is that correct? Of course, in addition to this
the repeated measures manova also tells you if your between subjects
effect varies as a function of your multivariate DV. When it says that
the effect varies as a function of the DV how is this determined. Is it
just a comparison of the univariate effects across the different levels
of the DV?
Since the main variable of interest here is the betweensubjects effect
it's still unclear to me which makes more sense, straight manova or
repeated measures. A comparison across the average dv (repeated
measures) seems more interpretable than comparing across the root that
best fits the dvs.
The data involves the comparing of the amount of care given to a child
by mom with the amount of care by dad. The four variables are measuring
different aspects of the construct of care. Along this same line, the
researcher had wanted to look at what I've been calling the
betweensubjects factor "parent" as a withinsubjects factor, because
they relate to the same kid. I didn't think this made sense. They are
after all two different people. If we could consider parent a
withinsubjects factor than we could look at the withinsubjects effect
of parent. Any thoughts. Any references. This is a bit beyond my
understanding.
Thanks
Matt
Matthew Pirritano, Ph.D.
National Science Foundation PostDoctoral Fellow
College of Education
Department of Individual, Family & Community Education
MSC05 3040
1 University of New Mexico
Albuquerque, NM 871310001
Telephone (505)2777115
FAX (505)2778361
Swank, Paul R wrote:
> If the variables are measurng the same construct but are not in the same
> scale then MANOVA is the best approach. Different scales will make the
> interpretation of the differences between scales problematic, unless you
> are interested in a profile analysis. I that case we would only be
> interested in parallelness.
>
>
> Paul R. Swank, Ph.D.
> Professor, Developmental Pediatrics
> Director of Research, Center for Improving the Readiness of Children for
> Learning and Education (C.I.R.C.L.E.)
> Medical School
> UT Health Science Center at Houston
>
> Original Message
> From: SPSSX(r) Discussion [mailto:SPSSXL@LISTSERV.UGA.EDU] On Behalf Of
> Zdaniuk, Bozena
> Sent: Monday, April 03, 2006 11:27 AM
> To: SPSSXL@LISTSERV.UGA.EDU
> Subject: FW: Re: Difference between MANOVA and RepeatedMeasures MANOVA
>
> If I may have an opinion here, I think you have a regular MANOVA
> (multiple variables from the same subjects) but not Repeated Measures
> MANOVA. Am I wrong?
> b.
>
> Bozena Zdaniuk, Ph.D.
>
> University of Pittsburgh
>
> UCSUR, 6th Fl.
>
> 121 University Place
>
> Pittsburgh, PA 15260
>
> Ph.: 4126245736
>
> Fax: 4126244810
>
> email: bozena@pitt.edu
>
>
> Original Message
> From: SPSSX(r) Discussion [mailto:SPSSXL@LISTSERV.UGA.EDU] On Behalf Of
> Matthew Pirritano
> Sent: Monday, April 03, 2006 12:07 PM
> To: SPSSXL@LISTSERV.UGA.EDU
> Subject: Re: Difference between MANOVA and RepeatedMeasures MANOVA
>
> The four repeated measures in this case are different measures of
> parental care. So, while they are not the same measure repeated at
> different points in time I thought it would make more sense to use
> repeated measures MANOVA as opposed to a MANOVA.
>
> Thanks
> Matt
>
> Matthew Pirritano, Ph.D.
> National Science Foundation PostDoctoral Fellow College of Education
> Department of Individual, Family & Community Education
> MSC05 3040
> 1 University of New Mexico
> Albuquerque, NM 871310001
> Telephone (505)2777115
> FAX (505)2778361
>
>
>
>
> Fredric E. Rose, Ph.D. wrote:
>
>> It would depend on what those 4 measures are. If they are the same
>>
> thing
>
>> (say, aggressive acts at different times or under different
>>
> conditions),
>
>> then a 2 (mother vs father) x 4 (condition/time) mixed ANOVA with
>>
> condition
>
>> as the repeated measure would be the way to go. If, however, the
>>
> measures
>
>> are different, such as parental attitude, # of negative comments made,
>>
>
>
>> aggression frequency, and empathy score, then a MANOVA would be the
>>
> way to
>
>> go. The main issue is whether the DV is the same "thing" or
>>
> different.
>
>> F
>>
>>
>> On 3/31/06 2:33 PM, "Matthew Pirritano" <mpirrita@unm.edu> wrote:
>>
>>
>>
>>> Say I have four measures on a mother, and the same four measures on
>>>
> the
>
>>> father. Being that I have four measures on a person I think that I
>>>
> have
>
>>> repeatedmeasures. I have sample of 100 mothers, and 100 fathers. I
>>>
>
>
>>> want to compare mom and dad on these four measures. Does it make
>>>
> more
>
>>> sense to do a MANOVA or a RepeatedMeasures MANOVA, and why?
>>>
>>> Thanks
>>> Matt
>>>
>>> Matthew Pirritano, Ph.D.
>>> National Science Foundation PostDoctoral Fellow College of Education
>>>
>
>
>>> Department of Individual, Family & Community Education
>>> MSC05 3040
>>> 1 University of New Mexico
>>> Albuquerque, NM 871310001
>>> Telephone (505)2777115
>>> FAX (505)2778361
>>>
>>>
>>
