I get involved in a study with only 40 subjects, and each subject has repeated measures at 6 months, 12 months, and 24 months. Originally I supposed that the total sample size 40x3 = 120 should be enough to run a mixed model to evaluate the association between two continuous variables. Unfortunately, the reviewers of our manuscript still had a concern in the sample size, so I'd like to use a nonparametric approach to handle this issue. Nonetheless, after searching for several weeks, the only possibility might be the Friedman test, while my study is not a factorial design. I also tried an additive model with both the main effect and the time variable in two separate smoothing functions, but the time variable only has 3 values, which cannot be smoothed enough. I am wondering whether there is a proper nonparametric model to analyze repeated measures with a small sample size data? Any suggestion will be appreciated.

  • $\begingroup$ Bootstrapping or bayes? $\endgroup$ – kjetil b halvorsen Nov 4 '19 at 14:58
  • $\begingroup$ I possibly will use a Bayesian model, but just feel annoyed from choosing priors. Bootstrapping might be another choice, while I need to learn the programming. $\endgroup$ – cchien Nov 5 '19 at 17:31

Your Answer

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

Browse other questions tagged or ask your own question.