I'm taking a data analysis course and decided to work on a customer analysis project. In the data, I have three countries:

  1. USA (539 unique users)
  2. BRA(385 unique users)
  3. TUR (129 unique users)

I'm trying to analyze the country that brings in the most income, so I've decided to look at the mean revenue for each country. However when I do that I get the following result:

enter image description here

It's not possible for Turkey to generate the most mean, because it has the lowest number of users and the lowest sum. I think it's showing the highest mean because the denominator when calculating the average is small.

How would you go about solving this issue? Should I randomly sample data from each country and then calculate the mean?

I'd really appreciate any pointers you could give or direction on how this would be solved in a real scenario. Thank you in advance! :)

  • $\begingroup$ Is it actually an issue ? $\endgroup$ Apr 17, 2023 at 5:34

1 Answer 1


You can try Stratified Sampling. You can try SMOTE analysis You can try methods like 10 fold cross validation if you want train the ML model You can also use bagging, boosting or random forest algorithms.


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.