st: Specifying a repeated measures analysis in anova and xtmixed

[David Pieczkiewicz <piecz001@umn.edu>]

Hello,

I am relatively new to Stata (and to the types of analyses I ask  
about below), and I hope my questions are not too elementary for the  
list. I've consulted the manuals and what online help I could find,  
but I was unable to resolve my issues. I'm hoping some of you can  
offer pointers.

Briefly, I am conducting research dealing with different ways of  
presenting medical data to clinicians. One aspect I am interested in  
is the time it takes for clinicians to reach decisions when using  
these displays, and whether different ways of presenting the same  
data take longer or shorter times than others.

My experimental design consists of a series of cases (i.e., data  
taken from individual patients), each shown in different display  
formats (i.e., modalities), to a variety of clinician readers, who  
then decide on what to do with that patient. *Every* case is shown in  
*every* modality to *every* reader. If I'm using the correct  
terminology, this is a complete, balanced design, with no nesting,  
blocking, etc. For each reader, the order of cases/modalities is  
randomized. Different data are collected from each combination of  
case, modality, and reader, including the time it takes for decisions  
to be rendered in each case/modality/reader combination.

My understanding is that this is a repeated measures design, with  
cases being repeated over both modality and reader. I have two  
questions, dealing with how to properly analyze this data with Stata  
(version 9):

1) Using the anova command, I believe this is an appropriate  
statement to examine the influence of modality on decision time:

anova lntime case modality reader modality*reader modality*case  
reader*case, repeated(modality reader)

where "lntime" is the natural log-transformed decision time(s), and  
case, modality and reader are the categorical variables as described  
above. While I am most interested in the factor "modality," I have  
some interest in the first-order interactions, hence their inclusion  
here.

Does this command look correct? Part of my uncertainty was that for a  
while, I was under the impression that the variables going into the  
repeated() clause were the experimental units being repeated (here,  
cases). I've come to understand that repeated() should instead  
contain the variables that the experimental units are being repeated  
over. I want to make sure I'm accurate in this before I get too far in.

2) If I wanted to analyze this as a mixed model, what is an  
appropriate way to use xtmixed? I consider case and reader to be  
random effects, as both are drawn from larger populations of possible  
cases and readers. Likewise, I consider modality to be a fixed  
factor, as there are only N modalities of interest in any particular  
phase of my study. As such, I came up with the following as a rough  
guess of how to express it:

xtmixed lntime modality || case: || reader:

Given that this is a repeated measures design, how can/should I  
express the fact that the cases are being repeated over all levels of  
modality and reader? I'm much less certain about this one, mainly  
because I have even less practical experience with mixed models than  
ANOVAs.

Any pointers, hints, clarifications, anyone can provide will be well- 
appreciated, not just for my understanding of how to examine this  
particular experiment, but repeated measures and mixed model analyses  
in general. Thank you!

Dave...

=====
David S. Pieczkiewicz, M.A.                    (612) 625-3241
Division of Health Informatics                    piecz001@umn.edu
University of Minnesota, Twin Cities Campus


*
*   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/