http://spssx-discussion.165.s1.nabble.com/GLM-repeated-measures-concern-question-tp5104022p5133121.html
I am trying to determine whether my data are missing at random using Little's MCAR test.
When I run the test via the missing values analysis, I also get a notification that "The EM algorithm failed to converge in 25 iterations".
When I increase the iterations (i.e., 50,75,100), the Chi-square value for Little's MCAR test also changes. In some cases from significant to non-significant.
I'm wondering if anyone could provide me with more information about the relationship between these and whether or not the failed convergence on the EM algorithm makes the Little's MCAR test invalid.
Thanks!
Amanda
Amanda Brouwer
Patient Advocacy & Research Lab
Pearse Hall, B53
University of Wisconsin - Milwaukee
P.O. Box 413
Milwaukee, WI 53201
[hidden email]
From: "eyeman03" <
[hidden email]>
To: [hidden email]Sent: Monday, January 9, 2012 8:14:08 AM
Subject: Re: GLM repeated measures concern/question
First, thank you* EVERYONE* for your input. It’s really appreciated.
Diana,
Let me expand a little on the design. We are studying the effects of two
types of simulated central blind spots on adaptation. The task is Visual
Search with 3 set sizes (1, 8, 32)
The research design is a mixed 2x2 ANOVA with adaptation as the within
effect, and blind spot type as the between effect. The statistical tool is
GLM Repeated measures.
As you know, our problem is the unequal spacing of the set sizes, and
whether or not SPSS intrinsically recognizes the spacing as equal. You
answered that question; yes, it does.
We’ve come to learn we can use the CONTRAST option for set size. However,
If we add set size as a covariate (as you suggested), then SPSS will not
recognize it as a factor (CONTRAST).
But if set size is added as a Between-Subject factor , we can now write
syntax as a CONTRAST: /CONTRAST(Set size)=Polynomial (1, 8, 32).
We ran two different syntaxes to compare their CONTRAST results. :
*/CONTRAST(Set size)=Polynomial (1, 8, 32)* vs /CONTRAST(Set
size)=Polynomial.
With the *former syntax*, only the linear contrast was significant. With
the latter syntax, both the linear and quadratic contrast were highly
significant.
So the output analysis with stating vs not stating the set sizes made a
significant difference.
We wanted to ask you opinion if this syntax solution: CONTRAST POLYOMIALS
(1, 8, 32) seems correct for our analysis
Dave
PS……I can send the Output data if anyone is interested.
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/GLM-repeated-measures-concern-question-tp5104022p5131418.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