I want to compare multiple proportions between two samples. There are 60 brands of toys (T1, T2, T3...T60). I would like to know if overall, the proportion of respondents who own each brand of toy is significantly different between the two samples. I believe I can use a chi-squared test here but over 40% of the expected value cell counts are less than 5. Infact, 18% are less than 1. I cannot pool the brand category because there is no meaningful way to do this. What are my alternatives here? I know only basic statistics, so a little more elaboration for more advanced suggestions would be helpful.


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.