I have three groups of people: Group 1, Group 2, Group3. Each group has about 20 people. And I collected data on the proportion of certain features for each person.

Group Person ID feature 1 feature 2 .... feature 20 total of features
A A1 20% 5% 25% 3% 100%
A A2 30% 10% 2% 0% 100%
B B1 34% 0% 45% 25% 100%


So each person is having a different profile of features - but I want to compare whether the feature profile differs on the group level, and if differs, where is the source of the difference. The statistical methods I could think of are:

  • Chi-squared-test on group level (averaging the percentage over people in each group), but then I lose individual data.
  • Generalised mixed effects model glm(Group ~ feature 2 + feature 3 .... + feature 19 + (1|Person ID)), but it does not seem to answer the research question.

Any suggestion?



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.