test subcommand in mixed

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

test subcommand in mixed

Maguin, Eugene

I’m trying and failing to understand how to use the test subcommand in mixed. The examples and explanation in the documentation is far too stingy to be helpful. One thing that would be helpful are citations that go into detail on this topic. I welcome them.

 

I ginned up little 2x2 between factor dataset, x and m being the factors and y the dv.

The book shows an example that I adapted (page 1141, v26).

 

mixed y by x/fixed x/print solution cov/TEST = 'Contrasts of x' x 1/2 1/2; x 1 -1.

The result was this. Not sure why.

*  ERROR: L matrix not estimable.

 

However, this test command worked, although I should have said -1 1 to get the right sign.

TEST = 'Contrast of x' x 1 -1.

 

Moving on, I tried this.

mixed y by x m/fixed x m x*m/print solution cov/

   TEST = 'Contrast of x' x -1 1/

   TEST = 'Contrast of m' m -1 1/

   TEST = 'Contrast of x at m=1' x -1 1 m -1 1 m*x -1 1 -1 1.

 

Nothing was liked, same estimable matrix problem.

 

Some might wonder why I’m bothering with test when emmeans will do the same thing. Yes, it will and it is easily understandable, I assume. But for a growth curve model that has an x*time effect (x binary) and want to determine whether the interaction is significant within the range of the time variable

 

Thanks, Gene Maguin

 

 

 

 

 

 

 

===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: test subcommand in mixed

Ryan Black
Gene,

I‘ve posted a number of messages showing how to use the TEST sub-command. 

Here are a few messages you might find helpful:







HTH,

Ryan

Sent from my iPhone

On Sep 7, 2020, at 2:33 PM, Maguin, Eugene <[hidden email]> wrote:



I’m trying and failing to understand how to use the test subcommand in mixed. The examples and explanation in the documentation is far too stingy to be helpful. One thing that would be helpful are citations that go into detail on this topic. I welcome them.

 

I ginned up little 2x2 between factor dataset, x and m being the factors and y the dv.

The book shows an example that I adapted (page 1141, v26).

 

mixed y by x/fixed x/print solution cov/TEST = 'Contrasts of x' x 1/2 1/2; x 1 -1.

The result was this. Not sure why.

*  ERROR: L matrix not estimable.

 

However, this test command worked, although I should have said -1 1 to get the right sign.

TEST = 'Contrast of x' x 1 -1.

 

Moving on, I tried this.

mixed y by x m/fixed x m x*m/print solution cov/

   TEST = 'Contrast of x' x -1 1/

   TEST = 'Contrast of m' m -1 1/

   TEST = 'Contrast of x at m=1' x -1 1 m -1 1 m*x -1 1 -1 1.

 

Nothing was liked, same estimable matrix problem.

 

Some might wonder why I’m bothering with test when emmeans will do the same thing. Yes, it will and it is easily understandable, I assume. But for a growth curve model that has an x*time effect (x binary) and want to determine whether the interaction is significant within the range of the time variable

 

Thanks, Gene Maguin

 

 

 

 

 

 

 

===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: test subcommand in mixed

Bruce Weaver
Administrator
Those links take me to a page where I am asked to login using Ryan's gmail
address.  ;-)  

In my MIXED folder, I have a text file with excerpts from some of the TEST
examples Ryan has posted in the past.  I suspect some of them are ones he
tried to point to.  Here are some links to those posts in the Nabble
archive, which anyone should be able to see without logging in (as Ryan or
someone else).  

http://spssx-discussion.1045642.n5.nabble.com/Mixed-Model-Planned-Comparisons-tp5163929.html
http://spssx-discussion.1045642.n5.nabble.com/TEST-contrast-for-MIXED-td5715139.html

HTH.



Ryan Black wrote




-----
--
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.

--
Sent from: http://spssx-discussion.1045642.n5.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
bweaver@lakeheadu.ca
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.