We are doing a study where we have measured the contractile properties of the lumbar spine muscles. Each individual (N=34) was tested at 12 different sites along their spine. 6 sites (vertebral levels) on the left side of the spine, and 6 sites on the right side of the spine.
We are trying to determine whether there is any significant difference between the contractile properties measured at each site.
So basically we are comparing each individual site to all other vertebral levels on both the left and right side. And we also want to know if the contractile properties at each site vary between male and females.

What would be the best statistical method for this analysis?


1 Answer 1


I would start from hierarchical linear regression for a start. However the design looks complicated and the sample size small for the amount of hypotheses you are trying to address. I am not sure how easy it would be to create a single model in order to test all these hypotheses and I am not sure if a single model would converge.


Your Answer

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

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