00:01
Suppose you're testing someone to see whether she or he can tell pepsi apart from coke.
00:05
And you're using 20 trials, half with coke, half with pepsi.
00:09
The null hypothesis is the person is guessing, which means if they were guessing, then the proportion that they would get right is 50%.
00:20
That's what it means if they were guessing and it was equally likely.
00:24
About how many would you expect the person to get right under the null hypothesis that the person is guessing? so the expected value would be n times p, which is 0 .5 times 20, or you would expect them to get 10 correct and 10 incorrect if they were just guessing.
00:43
Suppose a gets 13 out of 20, right? and person b gets 18 out of 20, which has a smaller p value.
01:00
When it's asking what's the p value it's saying which one of these two is more unusual or which is further from the mean so let's just run this the short answer is going to be b it's more unusual if someone was straight up guessing that they got 18 out of 20 versus 13 out of 20 but let's actually you know run a hypothesis test so we go to statistics, and we're doing a stats test, and we're doing a one -prop z test.
01:38
The null hypothesis is 0 .5.
01:42
The number of successes is 13 out of 20, and then the alternate hypothesis would be that they're not guessing...