Grasping the concept of P-value

Asked Dec 29 '14 at 22:18

Active Dec 30 '14 at 02:54

Viewed 62 times

I am having a hard time grasping the concept of the P-value.. I've read some of the posts in here and on internet, but one thing seem to be missing for me.

Usually we look at the probability of values being higher than the sample mean, and if this probability is lower than 0.05($\alpha$) it would be rejected, which don't make sense.

The probability says that values above test-statstics the would occur less time, but why would that lead to rejection of null hypo?

as it is stated a lot of places

In technical terms, a P value is the probability of obtaining an effect at least as extreme as the one in your sample data, assuming the truth of the null hypothesis.

If the probability of getting extreme values are higher than 0.05, then the null hypo will be accepted?.. which makes no sense.

edited Dec 30 '14 at 02:54

Jeremy Miles

13,917
6
30
64

asked Dec 29 '14 at 22:18

Stats

Made it more precise – Stats Dec 29 '14 at 22:31
2

Your final sentence indeed makes no sense, but that's not what the quote above it says. – Glen_b Dec 30 '14 at 01:35
I found a solution for my problem.. What didn't make sense was that the meaning of the probability, but it make more sense, since the probability is seen on a normal dist. If I could answer it myself i Would do it. – Stats Dec 30 '14 at 12:36
The p-value is not in general related to a normal distribution. – Glen_b Dec 30 '14 at 16:50
Well. yeah but i understand how it works now. – Stats Dec 30 '14 at 22:01

Grasping the concept of P-value

0 Answers0

Linked