just a quick question I was a little unsure about. Any thoughts much appreciated!

Say I am running an 2 condition experimental trial where one of the outcomes is a z-score (e.g. cognitive test measured pre and post each condition).

To make the results more 'meaningful', would it be Kosher/correct to calculate the % difference in the z-score between the 2 conditions? To show there was x % difference in the z-scores between the 2 conditions.



Z-scores can easily be 0, so I don't think that's a safe thing to do.

A difference between them should be enough in your case. That has the advantage of interpretability as well.

  • $\begingroup$ Thanks - yes good point! I thought a % change perhaps might make the difference between interventions a bit more interpretable (i.e. x% reduction in cognitive function) - but not if invalid (due to the potential for zero values). $\endgroup$ – user194416 Feb 16 '18 at 0:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.