i have two groups: people who place orders online and people who place orders over the phone. I am comparing order sizes between groups. order size probability distribution is right skewed for both groups (in other words there are a few people who place very expensive orders in each group). My average order size for online group is $90 and for phone group $110. i want to test whether the difference is statistically significant between averages in these two groups - i do not know which test i should use for that?

Question