00:01
So we would be assuming that each of these three means is equal and alternately at least two of the means are different.
00:07
And you have a sample size of five for each one.
00:12
And here are the sample means.
00:14
And these are the sample variances.
00:18
And we also have the mean of these means is going to be, or the grand mean is 36.
00:27
So to find the mean square for the group or the mean square for the factor, we're going to take one less than the sample size times the difference in these means squared.
00:36
And we do that for each of the three.
00:38
And then we divide that by i minus 1 or the number of categories less 1, which is the degrees of freedom of the numerator.
00:45
And that gives us 228.
00:47
And the mean square for the error is to take one less than the sample size for the group times its variance open.
00:54
This is a mistake.
00:56
So that was a variance and i don't need to square that.
00:59
So let me quick correct that.
01:01
So that is going to be 4 times 3 .5.
01:07
And those shouldn't be squared.
01:08
Those are not standard deviations.
01:10
So we'll have that 4 times 3 .5.
01:15
And we have three of those.
01:17
And then we're going to divide that by 12.
01:19
So this becomes 3 .5.
01:22
So that's my mistake.
01:23
And i'm glad i just found that.
01:25
So the f statistic, which will have two degrees of freedom in the numerator, 12 in the denominator, is that 228 divided by 3 .5.
01:34
And 228 divided by 3 .5 is a value of a whopping 65 .143.
01:43
And then the p -value for this, the p -value is, and i guess you want that to two decimal places, the p -value is the likelihood of being greater than or equal to that, which we will go to our fcdf...