Repeated measures predictor variable

[Jill Stoltzfus]

Hello everyone. I need to analyze data where the outcome is absence or presence of gallstones and one of the predictor variables is amylase measurements taken at 3 points in time for each subject. The hope is that decreasing amylase levels will predict absence of gallstones. Can anyone tell me whether SPSS can handle such an analysis with through mixed models? 

Thanks very much.

Jill

[Bruce Weaver]

Jill Stoltzfus wrote

Hello everyone. I need to analyze data where the outcome is absence or presence of gallstones and one of the predictor variables is amylase measurements taken at 3 points in time for each subject. The hope is that decreasing amylase levels will predict absence of gallstones. Can anyone tell me whether SPSS can handle such an analysis with through mixed models? 
Thanks very much.
Jill

Jill, is the outcome variable (presence/absence of gallstones) also measured 3 times, or just once?  

--
Bruce Weaver
[e-mail]
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."

[Jill Stoltzfus]

The binary outcome is measured only once. 

Jill

--- On Thu, 7/2/09, Bruce Weaver <[hidden email]> wrote:

From: Bruce Weaver <[hidden email]>
Subject: Re: Repeated measures predictor variable
To: [hidden email]
Date: Thursday, July 2, 2009, 12:27 PM

Jill Stoltzfus wrote:
>
> Hello everyone. I need to analyze data where the outcome is absence or
> presence of gallstones and one of the predictor variables is amylase
> measurements taken at 3 points in time for each subject. The hope is that
> decreasing amylase levels will predict absence of gallstones. Can anyone
> tell me whether SPSS can handle such an analysis with through mixed
> models?Â
> Thanks very much.
> Jill
>

Jill, is the outcome variable (presence/absence of gallstones) also measured
3 times, or just once?

--
Bruce Weaver
bweaver@...
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."


-----
--
Bruce Weaver
bweaver@...
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."

--
View this message in context: http://www.nabble.com/Repeated-measures-predictor-variable-tp24307585p24309850.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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]

Jill Stoltzfus wrote

The binary outcome is measured only once. 
Jill

--- On Thu, 7/2/09, Bruce Weaver <[e-mail]> wrote:

From: Bruce Weaver <[e-mail]>
Subject: Re: Repeated measures predictor variable
To: [e-mail]
Date: Thursday, July 2, 2009, 12:27 PM

Jill Stoltzfus wrote:
>
> Hello everyone. I need to analyze data where the outcome is absence or
> presence of gallstones and one of the predictor variables is amylase
> measurements taken at 3 points in time for each subject. The hope is that
> decreasing amylase levels will predict absence of gallstones. Can anyone
> tell me whether SPSS can handle such an analysis with through mixed
> models?Â
> Thanks very much.
> Jill
>

Jill, is the outcome variable (presence/absence of gallstones) also measured
3 times, or just once?

If the outcome variable is measured just once, garden variety logistic regression should work.  But you may need to include some product terms, judging from what you said.  E.g., the effect of increasing the 3rd measurement by one unit may depend on what the values were for the 1st and 2nd measurements.  If you're not familiar with including product terms (interactions) in logistic regression models, you might find James Jaccard's Sage monograph on that topic (Interaction Effects in Logistic Regression) very helpful.  

--
Bruce Weaver
[e-mail]
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."