1

Sorry if this is a basic question, I have been learning AB testing for couple of days now. For instance i have a table with visitors and purchases for four different versions of websites and 

visitors|Purchases
21      | 3
180     |30
250     |50
100     |15

Is there a way that I can find the pvalue for this scenario in Python to find the significance of these tests?

TIA

deadvoid
  • 1,270
  • 10
  • 19
apak23
  • 11
  • 1
  • 5
  • Are you trying to check whether more visitors equals more purchases? – user2974951 Oct 22 '18 at 06:52
  • Yes. the four rows are four different versions and I am trying to check which image has more significance for more purchases – apak23 Oct 22 '18 at 14:02
  • Since you only have 4 groups each with 1 record then you are pretty limited. A chi-square test will test whether all these values come from the same distribution, if you reject this null hypothesis then you can use a graph to determine which is likely to be the culprit, as for a p-value... not with data like these. – user2974951 Oct 23 '18 at 08:11

0 Answers0