TEST subcommand in MIXED

[prb]

Help needed using the /TEST subcommand in MIXED!

I am currently running a mixed model with 2 fixed factors (Nrecode 3 levels; Focal 20 levels) and TrialNum as a random factor. I want to make simple comparisons between the first and second level of Nrecode (Nrecode 1 -1 0) as well as between the first and third levels of Nrecode (Nrecode 1 0 -1). I am having trouble getting these contrasts using the /TEST subcommand.

I get the error message "This L matrix is not estimable, test results will not be produced".
The following is the syntax I used:

MIXED PropLength BY Nrecode TrialNum Focal
  /CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1) SINGULAR(0.000000000001) HCONVERGE(0,
    ABSOLUTE) LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE)
  /FIXED=Nrecode Focal Nrecode*Focal | SSTYPE(3)
  /METHOD=REML
  /RANDOM=TrialNum | COVTYPE(VC)
/TEST=Nrecode 1 -1 0; Nrecode 1 0 -1.

[Hart, Kimberly (hartkb)]

I will be out of office until Friday, May 24th, with limited access to email. I will respond to your email when I return.

Best, Kim

 

[Ryan]

Replace

 

/TEST=Nrecode 1 -1 0; Nrecode 1 0 -1.

 

with  

 

/EMMEANS=TABLES(Nrecode) COMPARE ADJ(LSD).

 

Ryan

On Tue, May 14, 2013 at 2:03 PM, prb <[hidden email]> wrote:

Help needed using the /TEST subcommand in MIXED!

I am currently running a mixed model with 2 fixed factors (Nrecode 3 levels;
Focal 20 levels) and TrialNum as a random factor. I want to make simple
comparisons between the first and second level of Nrecode (Nrecode 1 -1 0)
as well as between the first and third levels of Nrecode (Nrecode 1 0 -1). I
am having trouble getting these contrasts using the /TEST subcommand.

I get the error message "This L matrix is not estimable, test results will
not be produced".
The following is the syntax I used:

MIXED PropLength BY Nrecode TrialNum Focal
  /CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1)
SINGULAR(0.000000000001) HCONVERGE(0,
    ABSOLUTE) LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE)
  /FIXED=Nrecode Focal Nrecode*Focal | SSTYPE(3)
  /METHOD=REML
  /RANDOM=TrialNum | COVTYPE(VC)
/TEST=Nrecode 1 -1 0; Nrecode 1 0 -1.




--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/TEST-subcommand-in-MIXED-tp5720239.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