Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.
up vote 2 down vote favorite
1

I am currently trying to find the best way to analyse my data.

Participants were tested longitudinally on up to five occasions. On each day participants were assessed for the presence or absence of a neurological marker (dichotomous), and were tested on a continuous measure (scoring 0-10). I simply want to compare how the positive or negative diagnosis on any given occasion predicted score on the continuous measure.

Unfortunately as the study took place in a critical-care hospital ward participants were all tested a different number of times, and were positive or negative for varying proportions of these observations. The distributions of scores for each group on the continuous variable were also non-normal.

So, I was wondering if there was a non-parametric repeated measures test that could handle unbalanced designs? I have been told Multi-level modelling might be an option, but I don't have much experience with it.

Thanks in advance for any help!

share|improve this question
There are non-linear mixed effect models. At least some of these are available in SAS (PROC NLMIXED and PROC GLIMMIX) and in R (nlme package, and probably other packages) and probably in other programs as well. What software are you using? – Peter Flom Sep 15 '12 at 12:45
@Peter does Stata have a non-linear mixed model? – pmgjones Sep 15 '12 at 15:14
The other main package in R is lme4. – gung Sep 15 '12 at 16:30
I do not know anything about Stata, sorry. – Peter Flom Sep 15 '12 at 16:45
I'm using R, never used SAS. Thanks for that, I'll have a look into non-linear mixed effect models. – Cam Sep 15 '12 at 20:59

Know someone who can answer? Share a link to this question via email, Google+, Twitter, or Facebook.

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.