1. ## hp test

Do not hate me please. It's gonna be over soon.

Let's see if I understood what we did till now. Let's take this excercice.

In a random sample of 160 b-schools students, 72 sample members indicated some measure of their agreement with this statement: "scores on a standardized entrance exam are less important for a student's chances to succedeed academically than his high school GPA" Test th enull hypothesis that one half of all b-schools students would agree with this statement against a two sided alternative. Find and interpret the p-value.

Okay. I found the p value in this way: 72/160= 0.45 so
(0.45-0.50)/ sqroot(0.5*0.5/160) = -1.265
I go to the table and find 89.62%. Then I remembe that the alternative is 2 sided, so I do 1-.8962= 1038 and multiply it by 2. So p is 0.2076.

Now I am unsecure about what to do. Shall I take arbitrary alfa of 5%? Thank you so much for your patiente and help.

2. Originally Posted by 0123
Do not hate me please. It's gonna be over soon.

Let's see if I understood what we did till now. Let's take this excercice.

In a random sample of 160 b-schools students, 72 sample members indicated some measure of their agreement with this statement: "scores on a standardized entrance exam are less important for a student's chances to succedeed academically than his high school GPA" Test th enull hypothesis that one half of all b-schools students would agree with this statement against a two sided alternative. Find and interpret the p-value.

Okay. I found the p value in this way: 72/160= 0.45 so
(0.45-0.50)/ sqroot(0.5*0.5/160) = -1.265
I go to the table and find 89.62%. Then I remembe that the alternative is 2 sided, so I do 1-.8962= 1038 and multiply it by 2. So p is 0.2076.

Now I am unsecure about what to do. Shall I take arbitrary alfa of 5%? Thank you so much for your patiente and help.
Lets assume tha arithmetic is right, then we would expect to see a proportion
this far from 0.5 about 20% of the time, so we expect that we will not reject
the null hypothesis (this is because this means that we would expect to see
a deviation as large as this on about 1 in 5 such surveys)

Choose alpha to be 5% because you are given no other advice and this
value is traditional, and as our p-value > 0.05 we do not reject the null
hypothesis.

RonL

3. Originally Posted by CaptainBlack
Lets assume tha arithmetic is right, then we would expect to see a proportion
this far from 0.5 about 20% of the time, so we expect that we will not reject
the null hypothesis (this is because this means that we would expect to see
a deviation as large as this on about 1 in 5 such surveys)

RonL
So you' re telling me that the p value 0.2076 means that we have 20.76% of probability to find a sample value far from 0.5. But how much far? Because of course we are going to get it different, it's a sample..
I think my problem is that I do not understand what what p value wants to be in its life. My book says "the p value is the smallest significanmce level at which the null hypothesis can be rejected, given the observed sample statistics. " To me it is as clear as thick fog. Might you say it in more basic words? Please, I know I may not be very brilliant(not at all) with statistics, but I am really trying to understand. Thank you very much for all your help Captain.

4. Originally Posted by 0123
So you' re telling me that the p value 0.2076 means that we have 20.76% of probability to find a sample value far from 0.5. But how much far? Because of course we are going to get it different, it's a sample..
As far or further as the value observed

I think my problem is that I do not understand what what p value wants to be in its life. My book says "the p value is the smallest significanmce level at which the null hypothesis can be rejected, given the observed sample statistics. " To me it is as clear as thick fog. Might you say it in more basic words? Please, I know I may not be very brilliant(not at all) with statistics, but I am really trying to understand. Thank you very much for all your help Captain.
The dreadfull truth is that this is classical statistics, and though I not
supposed to tell you I will, it is rubbish, without rational foundation.

The problem is that this approach to statistics is what is taught to most
undergraduated for historical reasons, and what you will be examined on.