00:01
So for our sign test, we're going to be assuming that there's no difference between the medians and alternately that there is a difference.
00:07
And i looked at these values and i took the post minus the pre and found that there are 12 positive differences and only three negatives.
00:18
And we have 15 1ā2.
00:20
So if the proportion is and they were equal, then you would anticipate having 7 .5 positives and 7 .5 negatives.
00:29
So naturally you can't have exactly that.
00:31
But we want to find what the probability is using binomial probability of getting 12 positives or more.
00:41
And then so the probability of getting if x is the number of positives.
00:45
So the probability of getting greater than or equal to 12 for this situation.
00:52
And then we're going to double it.
00:54
And so i'm going to use my binomial cdf to do this.
00:58
And we could also do the likewise letting x be the number of negatives.
01:03
In fact, why don't we do that and that will make it easier to put in r.
01:07
So x being the number of negatives.
01:09
And so we'll find the likelihood of x being less than or equal to 3 in this distribution if the proportion is actually .5.
01:19
And then we'll double it...