jmp-l
[Top] [All Lists]

Re: repeated measurements

To: jmp-l@lists.biostat.wustl.edu
Subject: Re: repeated measurements
From: roncross <roncross@cox.net>
Date: Thu, 31 Mar 2005 21:07:50 -0800
In-reply-to: <693A217160D3324788FBA783ED4CF39C0361629D@apaupkvmbx01.cslg1.cslg.net>
References: <693A217160D3324788FBA783ED4CF39C0361629D@apaupkvmbx01.cslg1.cslg.net>
Thanks Gunter, I tried both approaches using the MANOVA as you suggest but I don't see anything about the run to run or within run variation calculated. I do see the pull down for Between Subjects and Within Subjects but there is nothing in there that I recognize as the mean square for calculating the variation. Where is this information located?

thx
RLC
On Mar 31, 2005, at 6:53 PM, <Gunter.Hartel@csl.com.au> wrote:

First you need to transpose the data so you have
Wideget S1      S2      S3
1               xx      xx      xx
2               xx      xx      xx
...
6               xx      xx      xx
Then you can use the MANOVA (Fit Models Platform) to do a repeated measures analysis on those three measures. I assume that for each widget S1 is a measurement in the same location, ie all theS1's are comparable, etc.

You could also use the random effects univariate approach by stacking the data:
Widget  Location        Measurement
1               1               xx
1               2               xx
1               3               xx
2               1               xx
2               2               xx

etc
then use fit models with measurement as Y, and Widget and location and possibly their interaction in the model. If widget is a random sample then make widget a random effect - ie select widget in the model and choose 'random effect' from the Attributes menu below.
Location may also be a random effect.
The analysis will then tell you what proportion of the variation is due to widget and what proportion due to location.

Of locations S1 to S3 are just random locations on the widget and don't correspond, ie the labels could be permuted without changing the meaning, then you can run location as a nested effect within widget.

Cheers
Gunter


-----Original Message-----
From: jmp-l-owner@lists.biostat.wustl.edu
[mailto:jmp-l-owner@lists.biostat.wustl.edu]On Behalf Of roncross
Sent: Friday, 1 April 2005 11:47 AM
To: jmp-l@lists.biostat.wustl.edu
Subject: [jmp-l] repeated measurements


Ok, I have a question about how to look at repeated measurements in jmp.


How do I perform an analysis for repeated measurements:

For example, I have the following data:

                w1              w2              w3              w4              
w5              w6
                s1              s1              s1              s1              
s1              s1
                s2              s2              s2              s2              
s2              s2
                s3              s3              s3              s3              
s3              s3

w1 through w6 are independent runs of wigets.  s1, s2, s3 are different
measurements location on a given widget, but each widget is measured in
the same location.

What do I need to do in jmp to get the variation from run to run and
the variation within each run? If I plot Y by X in a oneway anova, am I
making the assumption that  s1, s2, s3 are independent.

thx
RLC



*********************************************************************** ***************************************** This email and any attached files are intended solely for the named addressee, are confidential and may contain legally privileged information. The copying or distribution of them or any information they contain, by anyone other than the addressee, is prohibited. If you have received this email in error, please let us know by telephone or return the email to the sender
and destroy all copies. Thank you.

CSL Limited A.C.N. 051 588 348
45 Poplar Road Parkville Victoria 3052 Australia
Phone: +61 3 9389 1911  Fax: +61 3 9389 1434
*********************************************************************** ****************************************



<Prev in Thread] Current Thread [Next in Thread>