Hi guys,
I'm trying to create an in-workspace view of statistical significance, using a Z-Score to compare the conversion rates of two dimensions.
Unfortunately, I can't think of how to format a freeform table to provide a single Z-Score to then check for significance, can anyone help or do they have any examples?
Cheers in advance!
Solved! Go to Solution.
Views
Replies
Total Likes
You should be able to find the z-score:
(treatment rate - control rate) / standard deviation of population
There is an SD function there.
Then you create a calculated metric with the Cdf-z function: 1 - Cdf-z(z-score) that should give you the cumulative probability (i.e. Confidence value) that you're looking for.
Views
Replies
Total Likes
You should be able to find the z-score:
(treatment rate - control rate) / standard deviation of population
There is an SD function there.
Then you create a calculated metric with the Cdf-z function: 1 - Cdf-z(z-score) that should give you the cumulative probability (i.e. Confidence value) that you're looking for.
Views
Replies
Total Likes