experiment-model
experiment-model copied to clipboard
experimentcalculator.com
I'd thought it referred to a situation where only X% of the eligible visits are in the experiment, and get either control or treatment, with a 50/50 split for control/treatment.
The calculator is awesome--thanks! I tried inputting 99.99%, and got a really weird answer. I'm pretty sure its because of the way JS processes decimals. Maybe a solution is to...
Hi there. Nice tool - I especially like the clean interface. I ran some simulations based its sample size recommendations, and it seemed like the recommendations were very conservative. I...