I work with data for a mobile app and the analytics system records the dollar sum of user purchases made on each day. My task is to determine whether there is any statistically significant difference between the sum of purchases made on each day of the week (Sunday to Saturday).
I assume I can use the factual dollar sum total of purchases through the week and let 1/7 of it be the 'ideal' distribution for each day and then compare it with the actual percentage for each day of week. The issue is I'm unsure which test to use for it.
If I understand correctly, I can't use chi-square test because e.g. people who made a purchase on Monday could have done it on Tuesday too, so those are not mutually exclusive categories, which is one of the assumptions for that test?
I'm also curious if any of proposed tests might work in case where we compare the averaged values for the whole year (i.e. mean for Monday's dollar sum total of purchases avereged over the 52 weeks, etc.).
