Discussion in 'Community Discussion' started by njmitchel0, Dec 15, 2008.

1. ### njmitchel0 macrumors member

Joined:
May 21, 2008
#1
I have several problems dealing with hypothesis,
h_o, h_a, and what not...

Couple questions that Im unclear about...

1. How do I determine whether a a hypothesis is a left or right tailed test? Is this determined with the h_o or h_a?

2. Rejecting or failing to reject a hypothesis... How do I know which one to choose...also does it matter if the test is right or left tailed?

3. How is a conclusion drawn?

Help is so very much appreciated!!!!!!

2. ### iMacmatician macrumors 601

Joined:
Jul 20, 2008
#2
It's been a while since I took stats, so I'm a bit hazy on things.

Yes.

For example, if you were testing whether a modification to a rocket would make the rocket go up higher, you'd use a right-tailed test since you're only interested in whether the modification improves the upward distance (in other words, to the right). H0 = modification makes no change to distance, HA = modification increases distance.

If you were testing drag, however, you would use a left-tailed test because a lower drag (numbers to the left) is better. H0 = modification makes no change to drag, HA = modification reduces drag.

You fail to reject the hypothesis when the P-value is greater than your alpha value (usually .05). You reject the hypothesis when it is less than your alpha value. It doesn't matter which side the tail is on (at least with a calculator).