Explain statistical power and respond to extending a null A/B test

This tests whether you understand statistical power as the probability of detecting a true effect and recognize that extending a test after a null result is p-hacking. A strong answer defines power as one minus beta, explains that it is fixed during design by sample size and minimum detectable effect, refuses to extend since that invalidates p-values, and suggests reporting the effect size with a confidence interval then planning a new study. The interviewer listens for candidates who agree to extend the test or confuse power with sample size.