WEBVTT
00:00:00.290 --> 00:00:05.114
As we briefly showed before, when finding the probability of rolling an average
00:00:05.114 --> 00:00:09.650
of at least three with the tetrahedral die, the central limit theorem is not
00:00:09.650 --> 00:00:14.402
only awesome, but important, because it allows us to know where any sample mean
00:00:14.402 --> 00:00:19.842
falls on the distribution of sample means. In the example of the tetrahedral
00:00:19.842 --> 00:00:25.482
die, we wanted to know the probability of getting at least a three, for an
00:00:25.482 --> 00:00:31.037
average, if we rolled it twice. And we found that when we looked at the
00:00:31.037 --> 00:00:36.400
histogram, rolling at least a 3 was 6 out of 16. And now, we're extending this
00:00:36.400 --> 00:00:42.290
concept to populations. So, if we have the distribution of sample means where
00:00:42.290 --> 00:00:47.580
the samples can be any size. Where does a particular sample mean of that same
00:00:47.580 --> 00:00:52.948
size fall on the distribution? If we know where it falls on the distribution,
00:00:52.948 --> 00:00:58.140
then we can decide if this sample is typical or if something weird is going on.
00:00:58.140 --> 00:01:00.633
So, let's use another example.