Hypothesis testing, why alpha cannot be zero

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
1 reply · 3K views
Deathfish
Messages
80
Reaction score
0
Ok someone tell me the official explanation of why alpha cannot be zero in hypothesis testing.
 
Physics news on Phys.org
The UNOFFICIAL explanation (I don't know who could give you the "official" explanation) is that, if you set alpha to zero, you will always accept your null hypothesis, and thus hypothesis testing would be pointless.

Think about it. Testing at P < 0 means that you will call significant only a result that has probability less than 0 of happening if the null hypothesis is true. Obviously, there is never such a result.