# p value : Confusion in basics [closed]

If I have to test a hypothesis like:

\begin{align} H_0:\mu\leq10\\H_1:\mu>10\\ Z_{1-p}=\dfrac{\bar{X}-\mu}{\sigma/\sqrt n} \end{align} From which I get p-value and I make a comment about whether it makes sense to reject $H_0$.

What happens for double sided tests? $\mu=10,\mu\neq10$.

Do I simply take $2\times p$ from the above and then make comments?

If yes, why are we doing that?

-

## closed as off-topic by Austin Mohr, Sami Ben Romdhane, TooTone, Shuchang, 6005Feb 28 '14 at 1:14

This question appears to be off-topic. The users who voted to close gave this specific reason:

• "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – Community, TooTone, Shuchang
If this question can be reworded to fit the rules in the help center, please edit the question.

Yes. Ours is not to question why, ours is but to do or die. – Code-Guru Dec 14 '12 at 2:26

## 1 Answer

Now for a more serious answer.

You are correct that you take $2p$ where $p$ is the value from a table of normal distributions. This value represents the area under the curve of one tail of the normal distribution. With a double sided test, we need the area under the curve of both tails and so multiply by 2.

-
But why assume that the area of the tails will be the same? – Inquest Dec 14 '12 at 2:34
@Inquest Because a normal distribution is symmetric about the y-axis. That is we can show that the tails are the same from the definition of the normal distribution; it's not an assumption. – Code-Guru Dec 14 '12 at 2:35
If we were doing a two-tailed test on non-normal data (I've heard such a thing exists :P ), the calculation would be less simple, right? – Ben Millwood Dec 14 '12 at 2:42
@BenMillwood yes. – Code-Guru Dec 14 '12 at 2:50
So, how would I test $\sigma=10, \sigma\neq 10$ using chi-squared. (Two sided of course). – Inquest Dec 14 '12 at 3:18