Why "bother" with a null hypothesis at all?

Question

*(note: this is a **very basic** probability question, so it is highly probable (heh) that it is a duplicate)* Every time I am trying to get into statistics (again), I am always lost at hypothesis testing. My basic question is - why do we form a null hypothesis as a negation of what we want to prove in the first place, and only then do we prove or disprove the null hypothesis? Why do we do it at all, instead of just proving the original hypothesis?

Answer 1

Good question. The null hypothesis states that there is no effect. The alternative hypothesis contains our prediction of what we expect to happen. Rather than assuming that we're right (HA is true) and trying to figure the probabiity of H_0, we take the cautious route and assume H₀ is true and have to probe how likely it is we got our sample data. Thus, we can only reject or fail to reject the null.

Think of this analogy: in our court system, people are assumed innocent until proven guilty. (Null hypoth = innocence; alternate hypothesis = guilty). We take the cautious route and assume innocence and try to prove the likelihood of being innocent is very slim. Same with research, we take make the cautious assumption (no effect) and have to prove it false.

As a defendant, it would be unfair if we automatically assumed guilt and they had to try to prove their innocence. It would be risky for researchers to start with the assumption that they are correct in their prediction (H_A) .