I'm looking at cross-sectional survey data where respondents were presented with a series of clinical scenarios. Each scenario varied by patient age (3-levels), condition severity (mild/severe), and risk (high/low). For each scenario respondents were asked to check all treatment options that apply (from a list of 12). Does this count as a repeated measures design, or not? My predecessor used a logistic model with GEE, but left no notes or programs, so I'm starting from scratch and cannot do that procedure in JMP (although I may have SPSS soon). Is there a better option, such as a mixed model? The outcomes are the treatment options endorsed by the respondents. Thanks!
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
respondents were presented with a series of clinical scenarios
That is, each subject saw more than one clinical scenario? Then yes, you have repeated measures.
answered Oct 26, 2016 at 17:03
-
$\begingroup$ Correct. Thanks. Would a mixed model best fit these data or the logistic with GEE? I don't think either can be done in JMP, but I may be able to take it to SPSS $\endgroup$– JWislarCommented Oct 26, 2016 at 17:47
-
$\begingroup$ @JWislar I think it really depends on the details like your research questions, how the different treatment options relate to each other, how many subjects there are, how many scenarios there are per subject, etc. But I'm not familiar with GEE, in any case. $\endgroup$ Commented Oct 26, 2016 at 17:51