1
$\begingroup$

I am analyzing a survey and came across an interesting pattern. Weekday and month seems to have a very high say in the score given by customers. There can be many reasons for this, though it is unexpected, but how can I test whether it is significant?

From the charts Tuesday is by far the worst day to ask the customers, and February is the worst month. Combined it goes crazy…don’t ask the customer anything on Tuesdays in February.. How can I statistically back up or reject these hypothesis?

Average Customer Score x day and month

$\endgroup$
1
$\begingroup$

It looks like a good case for ANOVA, which can be seen as a generalization of Student t-test for more than two groups.

For visualizing the differences I'd suggest using boxplots (or violin plots) for every group: it's more informative than only plotting the mean which is sensitive to outliers.

| improve this answer | |
$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.