Why should an A/B test not be stopped at the first significant result?
- A Peeking inflates the false positive rate well above the stated threshold
- B It costs more to continue
- C The platform prevents it
- D Results are deleted
Answer
Peeking inflates the false positive rate well above the stated threshold
Sample size should be calculated in advance and the test run for whole business cycles to avoid day-of-week bias.





