Posted by
Maguin, Eugene on
Dec 28, 2011; 2:12pm
URL: http://spssx-discussion.165.s1.nabble.com/GLM-repeated-measures-concern-question-tp5104022p5105351.html
Dave,
I recall that the GLM documentation says that the default model for repeated measures is a polynomial. I think a first question I would have--especially since I know nothing about this area, is what shape of curve the data really do produce. So, are times for 8 stimuli 0.125 of those for 1 stimulus and those for 32 stimuli .03125 of those for 1 stimulus? Or, are they something else. Spss has a curve fit procedure, called that, I think, and that might be helpful here. Also, when you run GLM, you can shift the repeated measures contrast from polynomial to another alternative. You have three data points, 1, 8, 32, and those would define a cubic function. But you could have a power function of 2: 0, 3, 5.
By the way, you describe a 2 by 2 design. Just to be sure, 2 between factors and 2 within factors, right? I have assumed that stimuli number is a within factor, is that correct?
Gene Maguin
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of eyeman03
Sent: Tuesday, December 27, 2011 5:30 PM
To:
[hidden email]
Subject: GLM repeated measures concern/question
Hola,
We have a potential concern with analyzing our data:
*Quick Backgroud:* Research Design is 2x2 Mixed Model involving the effects
of central blind spots on Visual Search. The subjects are searching for a
target (O) embedded among distracters (Cs). The set sizes we have used are
1, 8, and 32 (1 stimuli on screen =1set size; 8 stimuli on screen= 8 set
size and so forth).
*Problem*: The potential concern is the set sizes (predictor variables) are
NON-LINEAR in spacing (unequal). Our plan is to use GLM Repeated Measures to
analyze the data, but were wondering if this GLM model can be used with the
set size non-linearityanother words does the GLM model implicitly assume
linear spacing with its analyses?
If yes, what are the potential solutions?
t/y
Dave
--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/GLM-repeated-measures-concern-question-tp5104022p5104022.htmlSent 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