These data are from an A/B test of free trials to conversion. Someone can sign up for a 30 day free trial of the product. If they like it they stay on and become paying customers. By default (A), when someone signs up they receive a 6 part email sequence over the 30 days. Group B, the test, received a 16 part sequence.
I would like to know if the difference in conversion rate for group by really is higher, or if there's a good chance it would have happened anyway with the natural ebbs and flows around the mean.
From Googling around and previous posts on this forum I wanted to use a chi-squared test. But my lecturer told me to test for a difference in proportions. Now I'm just confused.
Which method would be appropriate to test if the difference between the two groups is sufficient to arrive at any conclusions about one group being better than the other?