help with repeated measures ANCOVA

[505stats]

I am new to writing syntax, and am trying to figure out how to set up the following analysis:

I have an experiment with learning at four points in time as the DV, and a between subjects factor of condition.  Additionally, I have a covariate (a pretest score) for each of the four times.  I want to run a repeated measures ANCOVA with a between subjects variable, and with the covariate specific to the time, in other words, I want pretest one to covary only with time one, pretest two with time two, and so forth.  When I try this in SPSS with the drop down for the repeated measures, I am not able to assign a specific covariate to the specific point in time, rather all the pretests are entered together, so when I run the analysis, the covariates interact with all four times.  Any suggestions on how to get the covariate to be specific to only the applicable time?
Thanks for your help!

[Julius Sim]

Not sure I can answer the syntax question, but in terms of the analysis
per se, could you run separate ANCOVAs for each time point so as to obtain
the parameter estimates, and then generate the covariate-adjusted scores
for each time point, so as to enter these into a repeated measures ANOVA
for the full set of timepoints?

Best wishes,

Julius

> I am new to writing syntax, and am trying to figure out how to set up the
> following analysis:
>
> I have an experiment with learning at four points in time as the DV, and a
> between subjects factor of condition.  Additionally, I have a covariate (a
> pretest score) for each of the four times.  I want to run a repeated
> measures ANCOVA with a between subjects variable, and with the covariate
> specific to the time, in other words, I want pretest one to covary only
> with
> time one, pretest two with time two, and so forth.  When I try this in
> SPSS
> with the drop down for the repeated measures, I am not able to assign a
> specific covariate to the specific point in time, rather all the pretests
> are entered together, so when I run the analysis, the covariates interact
> with all four times.  Any suggestions on how to get the covariate to be
> specific to only the applicable time?
> Thanks for your help!
> --
> View this message in context:
> http://www.nabble.com/help-with-repeated-measures-ANCOVA-tp14444281p14444281.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

[Matthew Pirritano]

I could be wrong, but I believe that Sara would be best off running a multilevel/ mixed model.

Matt

Matthew Pirritano, Ph.D.
Assistant Professor of Psychology
Smith Hall 116C
Chapman University
Department of Psychology
One University Drive
Orange, CA 92866
Telephone [phone]
FAX [phone]



----- Original Message ----
From: Julius Sim <[hidden email]>
To: [hidden email]
Sent: Thursday, December 20, 2007 3:00:05 PM
Subject: Re: help with repeated measures ANCOVA

Not sure I can answer the syntax question, but in terms of the analysis
per se, could you run separate ANCOVAs for each time point so as to obtain
the parameter estimates, and then generate the covariate-adjusted scores
for each time point, so as to enter these into a repeated measures ANOVA
for the full set of timepoints?

Best wishes,

Julius

> I am new to writing syntax, and am trying to figure out how to set up the
> following analysis:
>
> I have an experiment with learning at four points in time as the DV, and a
> between subjects factor of condition.  Additionally, I have a covariate (a
> pretest score) for each of the four times.  I want to run a repeated
> measures ANCOVA with a between subjects variable, and with the covariate
> specific to the time, in other words, I want pretest one to covary only
> with
> time one, pretest two with time two, and so forth.  When I try this in
> SPSS
> with the drop down for the repeated measures, I am not able to assign a
> specific covariate to the specific point in time, rather all the pretests
> are entered together, so when I run the analysis, the covariates interact
> with all four times.  Any suggestions on how to get the covariate to be
> specific to only the applicable time?
> Thanks for your help!
> --
> View this message in context:
> http://www.nabble.com/help-with-repeated-measures-ANCOVA-tp14444281p14444281.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

=====================
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