Plz.. understand my lack of English skills.. T.T
I have a problem dealing with statistical hypothesis testing.
To be specific, we have two special delivery method like
(new) : method "A", (old) : method "B".
In our case, we did A/B test to say that Can (new) special delivery method "A" raise people's special delivery usage? For example,
Group "(a)" can select a delivery service either special method "A" or normal method,
and group "(b)" can select a delivery service either special "B" or normal.
And there are two different-sized sample sets because of experiment costs, so we have sample sets like Group (a) : n=100 , Group(B) : 1000.
We did Fisher's exact test and our conclusion was
(new) special delivery method "A" raise people's special delivery usage.
But now, I have different question about our whole sals perspective.
How can I design the experiment to say that did the "(new) method A" raise our whole sales compared with "old version"?
Because of different sample sizes, Group A's whole sales are always less than Group B's whole sales. How can I perform a reasonable comparison in this case?