00:01
So we're talking about doing a matched pair, one sample t -test here.
00:05
We have a before and after table with the 12 people in the table.
00:09
And so what i did over here on the right side is i subtracted the after minus the before, because we want more positive answers than negative answers.
00:22
So i did the differences here.
00:26
And so you end up getting mostly positive answers here.
00:28
The first one, you end up with a 1 .1, 7 .8 minus 6 .7.
00:33
And then you get a 0 .5, 5 .9, et cetera, 7 .4 minus 7 .0.
00:40
And what i did was i found the mean and standard deviation for this new list.
00:46
The new list, the mean is 0 .375, and the standard deviation is 0 .372.
00:53
So i take that information now and do a hypothesis test for a matched pair.
00:57
So whenever we do a matched pair, our ho always starts out as mu equals 0, because the null is that there is no difference.
01:09
So the hypothesis would be that mu equals 0.
01:11
The difference between them is 0.
01:13
You can see on the table that there's a couple of them that are 0.
01:16
There's a couple that are negative, and lots of them are positive.
01:18
So we'll see what happens.
01:19
And then we're trying to test if this program helped or improved score or reach.
01:24
So we're going with a greater than 0 there.
01:26
So it's a one -tailed test.
01:27
We're trying to find that p -value over there.
01:31
The test statistic formula, x bar minus the mean over the standard deviation over the square root of n, the degrees of freedom is going to be 11, because i have 12 in my sample.
01:41
So it's just a matter of plugging in the values that i've figured out...