Output from GLM for RM ANOVA

[Bruce Weaver]

This is one of those questions for people close to the "inner sanctum", as David would say.  ;-)

CONTEXT.  In another thread, I posted some advice for someone on how to obtain a 2x2 mixed design ANOVA (i.e., one between-Ss factor and one within-Ss factor).  Unfortunately, if they follow that advice, they will be inundated with a lot of output, most of which is not needed for their simple design (or even for more complex designs when all repeated measures factors have only two levels).  

My question is this:  

Can GLM be tweaked so that it does not automatically spew the multivariate test results, the test of sphericity, and the 3 or 4 lines giving results that are adjusted for sphericity?  

As I allude to above, if all repeated measures factors have only 2 levels, there can be no problem with lack of sphericity; and in that case, *many* people won't want to see the multivariate results or tests of sphericity.  

It *seems* like the addition of a couple check-boxes to the dialog (Show or Don't Show multiviariate tests / sphericity tests) would be fairly straightforward.  For backward compatibility, the default setting could be to display all of that junk.  

Thanks for considering.

[Swank, Paul R]

Well, you could just do an ANCOVA instead. Or if all you care about is the interaction effect, you could do a t test on the difference score.

Paul

Paul R. Swank, Ph.D.
Professor, Department of Pediatrics
Medical School
Adjunct Professor, Health Promotions and Behavioral Sciences
School of Public Health
University of Texas Health Science Center at Houston


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bruce Weaver
Sent: Thursday, July 26, 2012 4:16 PM
To: [hidden email]
Subject: Output from GLM for RM ANOVA

This is one of those questions for people close to the "inner sanctum", as David would say.  ;-)

CONTEXT.  In another thread, I posted some advice for someone on how to obtain a 2x2 mixed design ANOVA (i.e., one between-Ss factor and one within-Ss factor).  Unfortunately, if they follow that advice, they will be inundated with a lot of output, most of which is not needed for their simple design (or even for more complex designs when all repeated measures factors have only two levels).

My question is this:

Can GLM be tweaked so that it does not automatically spew the multivariate test results, the test of sphericity, and the 3 or 4 lines giving results that are adjusted for sphericity?

As I allude to above, if all repeated measures factors have only 2 levels, there can be no problem with lack of sphericity; and in that case, *many* people won't want to see the multivariate results or tests of sphericity.

It *seems* like the addition of a couple check-boxes to the dialog (Show or Don't Show multiviariate tests / sphericity tests) would be fairly straightforward.  For backward compatibility, the default setting could be to display all of that junk.

Thanks for considering.





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Output-from-GLM-for-RM-ANOVA-tp5714488.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (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

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

[Jon K Peck]

You can always use OMS to suppress unwanted tables.  For example,

oms /if subtypes='Default Multivariate Tests' /destination viewer=no.
GLM ...
omsend.

Subtypes can list as many types as desired.

You could even turn this into a macro :-(

Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
new phone: [phone]




From:        Bruce Weaver <[hidden email]>
To:        [hidden email]
Date:        07/26/2012 03:25 PM
Subject:        [SPSSX-L] Output from GLM for RM ANOVA
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

This is one of those questions for people close to the "inner sanctum", as
David would say.  ;-)

CONTEXT.  In another thread, I posted some advice for someone on how to
obtain a 2x2 mixed design ANOVA (i.e., one between-Ss factor and one
within-Ss factor).  Unfortunately, if they follow that advice, they will be
inundated with a lot of output, most of which is not needed for their simple
design (or even for more complex designs when all repeated measures factors
have only two levels).

My question is this:

Can GLM be tweaked so that it does not automatically spew the multivariate
test results, the test of sphericity, and the 3 or 4 lines giving results
that are adjusted for sphericity?

As I allude to above, if all repeated measures factors have only 2 levels,
there can be no problem with lack of sphericity; and in that case, *many*
people won't want to see the multivariate results or tests of sphericity.

It *seems* like the addition of a couple check-boxes to the dialog (Show or
Don't Show multiviariate tests / sphericity tests) would be fairly
straightforward.  For backward compatibility, the default setting could be
to display all of that junk.

Thanks for considering.





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Output-from-GLM-for-RM-ANOVA-tp5714488.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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

[Bruce Weaver]

Sure, you could.  And depending on the design, ANCOVA would arguably be better than the mixed design ANOVA.  But in cases where one really does want a mixed design ANOVA (and it is of higher order than two-way), and where all of the repeated measures have only 2 levels, I don't want the tests of sphericity, the 3 or 4 results that adjust for sphericity, or the multivariate tests.  But there is no straightforward* way to avoid them when one uses GLM.  

* In included "straightforward" here in an effort to prevent people from jumping in with various kludges such as using OMS (or Python) to suppress the output I don't want.  Those options should be part of the GLM syntax, IMO.

Swank, Paul R wrote

Well, you could just do an ANCOVA instead. Or if all you care about is the interaction effect, you could do a t test on the difference score.

Paul

Paul R. Swank, Ph.D.
Professor, Department of Pediatrics
Medical School
Adjunct Professor, Health Promotions and Behavioral Sciences
School of Public Health
University of Texas Health Science Center at Houston


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bruce Weaver
Sent: Thursday, July 26, 2012 4:16 PM
To: [hidden email]
Subject: Output from GLM for RM ANOVA

This is one of those questions for people close to the "inner sanctum", as David would say.  ;-)

CONTEXT.  In another thread, I posted some advice for someone on how to obtain a 2x2 mixed design ANOVA (i.e., one between-Ss factor and one within-Ss factor).  Unfortunately, if they follow that advice, they will be inundated with a lot of output, most of which is not needed for their simple design (or even for more complex designs when all repeated measures factors have only two levels).

My question is this:

Can GLM be tweaked so that it does not automatically spew the multivariate test results, the test of sphericity, and the 3 or 4 lines giving results that are adjusted for sphericity?

As I allude to above, if all repeated measures factors have only 2 levels, there can be no problem with lack of sphericity; and in that case, *many* people won't want to see the multivariate results or tests of sphericity.

It *seems* like the addition of a couple check-boxes to the dialog (Show or Don't Show multiviariate tests / sphericity tests) would be fairly straightforward.  For backward compatibility, the default setting could be to display all of that junk.

Thanks for considering.





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Output-from-GLM-for-RM-ANOVA-tp5714488.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (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

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (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