re: st: Repeated measures with no between-participants vars?

[David Airey <david.airey@Vanderbilt.Edu>]

> I am trying to assess the way in which participants' evaluations of 3
> different stimuli change over time, as indicated in a series of
> measurements. I am interested in checking whether the pattern of  
> changes
> in those evaluations is different for the different stimuli.
HOW DOES THIS WORK? DOES EACH SUBJECT SEE 9 PRESENTATIONS (3 TIMES  
FOR EACH OF 3 STIMULI) IN RANDOMIZED ORDER? SO SUBJECT 1 MIGHT HAVE  
SEEN (1,1,1) (3,3,3) (2,2,2) WHEREAS SUBJECT 2 SAW A DIFFERENT ORDER.  
JOSEPH IS THINKING YOU WERE CAREFUL ENOUGH TO BLOCK SUBJECTS INTO ONE  
OF THE 6 POSSIBLE ORDERS (1,2,3; 1,3,2; 2,1,3; 2,3,1; 3,1,2; 3,2,1).  
SO YOU PROBABLY HAD A MUTLIPLE OF 6 SUBJECTS. BECAUSE YOU HAVE MORE  
THAN ONE SUBJECT PER ORDER IF THIS IS THE CASE, ORDER IS A BETWEEN  
SUBJECTS FACTOR. YOU CAN MODEL IT OR NOT BUT IT SHOULD HAVE BEEN PART  
OF THE STUDY DESIGN.


> In other words, the two within-participants variables are (1) type of
> stimulus, and (2) time of measurement. There are no between- 
> participant
> variables.
>
> This seems like a simple task, but I haven't been able to get Stata  
> to do
> this for me. I tried a number of variations on the repeated ANOVA  
> command,
> but no luck. I would greatly appreciate any advice.
YOU CAN MODEL THIS WITH EITHER ANOVA, MANOVA, OR XTMIXED, IN ADDITION  
TO WHAT JOSEPH SAYS. FOR XTMIXED, SEE

<http://www.ats.ucla.edu/stat/Stata/faq/xtmixed.htm>

AND FOR ANOVA SEE

http://www.stata.com/support/faqs/stat/anova2.html

-DAVE
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/