Power = 1 - beta: Determining 'power' of test

J

Jaskaran

Junior Member
Joined
May 5, 2006
Messages
67
Okay, I'm bamboozled on how to measure power, AFAIK, power= 1-Beta

The text is confusing,
A professional basketball player with a poor free throw percentage gets a personal coach to help him increase his percentage. He tells his coach that he has raised his proficiency from 70% to 90%. To see whether the athlete has improved, the coach asks the athlete to take tens hots. He hits 10 out of 10. If the player really can hit 90% now, and it takes at least 10 out of 10 successful shots to convince the coach, what's the power of the test?
 
the power of a test is the probability under the alternative hypothesis of rejecting the null.

It takes 10/10 to convince the coach right? so the probability of rejecting the null hypothesis UNDER the alternative (90%) is .9^10. In otherwords, to choose the alternative hypothesis, he must hit all 10 shots. The prob that he hits all 10 shots if he is in fact a 90% shooter is .9^10.

... i think
 
You must log in or register to reply here.
Top