Hello everyone,

I'm looking for some advice on how to analyze the results of an experiment.
It concerns a repeated measures clinical study with a 5-phase withdrawal design A-B-A-B-A.

I would appreciate any advice about how to evaluate the effect of a new treatment on test scores.

There are 250 subjects, scores are normally distributed and there could be some missing values.

At each A (non-treatment) or B (treatment) phase there are 2 different tests performed resulting in a pair of test scores per subject, i.e.:

A : subj1 score for test1, subj1 score for test2 ,.., subj250 score for test1, subj250 score for test2
B : as above
A : " "
B : " "
A : " "

I want to detect if there is a significant difference in test scores between phases A and B.

The test should be suitable for testing if any performance changes we observe are attributable to the treatment under study or occur just by chance.

Does anyone out there know of a technique particularly suitable for analyzing the results from this type of design?