0
$\begingroup$

The researchers are measuring a 5-point ordinal outcome on animals subject to treatment combinations (k = 4). This is being measured at three time points. Between each time point, some animals have been diverted at random to other uses and won't be measured again. What would be an appropriate model for this?

Without the diversions, a cumulative linked mixed model would be just fine, but do I need to do anything to account for the attrition?

$\endgroup$

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.