I am currently studying the prevalence of a particular disease among eight different subgroups (for example, the prevalence of hepatitis B among HIV+ patients, HIV- patients, etc.). Each subgroup consists of multiple prevalence values from journal articles with particular sample sizes.

I would like to perform a test to see if there are significant differences between the prevalence values of the disease between these subgroups. After doing some research, it appears that a Kruskal-Wallis test would be the best test for the job since the distribution of prevalences within these groups is non-parametric; however, I am not certain of this and would appreciate advice if this seems incorrect.

If I do perform a Kruskal-Wallis test, is there a way to incorporate weights into the analysis? If, in a particular subgroup, there are prevalences that were each calculated from different sample sizes, would it not be beneficial to somehow incorporate the weight from the sample size into that study's respective prevalence?

If it is relevant, I have been using R for my analysis so far. I appreciate any advice and can give more information if it is needed. Thanks.


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.