I have come across a few hypothesis questions where the Q-Bank answer actually did not assume the contrary to the fact being tested. Such as in: “A bottler of iced tea wishes to ensure that an average of 16 ounces of tea is in each bottle…” (Schweser question ID 22960) What is wrong with assuming Ho:=mean NOT equal to 16 and Ha:= mean IS equal to 16? Is Schweser just plain wrong or am I getting my head mixed up?
usually the null hypothesis is = whereas the alternative is !=. The objective of a hypothesis test is to reject the null hypothesis claim, not to support the claim.
I though the null hypothesis Ho is suppose to be what you’re testing for? Could be wrong though…
No in hypothesis testing you are trying to prove Ha.
Exactly, I try to prove that my observation is, in fact, wrong. And if I fail to do that, I must accept that I could not show Ha to be true, that is fail to reject Ho. Now, how does that make a Ho stating the opposite of the observation wrong, according to the question mentioned above?
Is the question “What is wrong with assuming Ho:=mean NOT equal to 16 and Ha:= mean IS equal to 16?”? If so it’s because you can never prove that the mean is exactly equal to 16 as opposed to 16.00001, for instance. In hypothesis testing the strongest conclusion you will ever be able to get with regards to equality is that you do not have sufficient evidence to reject the possibility that mu = 16.
Yes, that was the question. All right, thanks, thats the knock on my head I needed right now. All clear now. Damn, if someone would have told me two weeks ago I would be struggling with such basic stuff I would have him committed 