00:01
So we have two different driving school instructors and one is a, one is b, and we have 30 out of 50 pass, so that is 60 percent and here we have 22 out of 50, so that's 44 percent and we want to know if we have evidence.
00:18
Do the results give us convincing evidence that a is more effective than b? we want to identify as many possible mistakes as we can.
00:27
Well first of all, we always give the null hypothesis first, and we would be assuming that yes, that's fine, that the proportion difference is zero and then alternately we would have that p1 minus p2 is greater than zero.
00:50
It is not greater than or equal to, never greater than or equal to, and we would have p1 be the proportion for a and p2 be the proportion for b.
01:02
So that's fine.
01:03
Now it says there's no significance level, so i'll use 0 .05.
01:08
That's fine and we have that the number of successes and the number of failures is at least 10, so that's fine and let's see, so and it says the, so you have those things are correct, they're at least 10, and we do have as long as the sample size of is less than or equal to 10 percent of the population, then you're okay.
01:41
We do have your p hats correct.
01:44
The pooled proportion, the pooled proportion does look like they're adding the 2, the 22 and the 30 and the 50 and the 50, so that looks fine, and that is 100 out of, 52 out of 100, so 0 .52, that was fine, and the test statistic, let's see the calculation.
02:08
We take the first proportion, actually we take the first proportion, that is a mistake.
02:13
So we take the 0 .6 minus and then this proportion, the p hat here is not 0 .4, so that is a mistake unless your list is incorrectly given.
02:27
You had 22 successes, so p hat for 2 is 22 out of 50, which is 0 .44, so that was a mistake...