1
00:00:00,688 --> 00:00:02,560
- (Instructor) So now that
you've been introduced

2
00:00:02,560 --> 00:00:06,820
to the statistics module
and a little bit with lists,

3
00:00:06,820 --> 00:00:08,860
we'd like you to go
ahead and create a list

4
00:00:08,860 --> 00:00:11,270
using the values that you see here

5
00:00:11,270 --> 00:00:14,160
and use the statistics
module's capabilities

6
00:00:14,160 --> 00:00:17,360
to calculate the mean,
the median, and the mode.

7
00:00:17,360 --> 00:00:19,830
So go ahead and pause the
video and give that a shot

8
00:00:19,830 --> 00:00:21,903
and then come back to see the results.

9
00:00:25,820 --> 00:00:28,920
Okay, let's go ahead and take
a look at the results here.

10
00:00:28,920 --> 00:00:32,680
So, of course in order to
use the statistics module,

11
00:00:32,680 --> 00:00:35,840
we need to first import
it, which we will do

12
00:00:35,840 --> 00:00:38,680
and we then want you to just to find

13
00:00:38,680 --> 00:00:41,950
a list containing those
six values in this case.

14
00:00:41,950 --> 00:00:43,493
So let's go ahead and do that.

15
00:00:44,950 --> 00:00:48,270
Again, comma separated values
within square brackets.

16
00:00:48,270 --> 00:00:51,350
And now we have three
additional cells here,

17
00:00:51,350 --> 00:00:54,050
each of which uses, from
the statistics module,

18
00:00:54,050 --> 00:00:56,650
the mean, the median,
or the mode function.

19
00:00:56,650 --> 00:00:59,610
And by the way, notice
here in Jupyter Notebooks

20
00:00:59,610 --> 00:01:03,750
that it's color coding
the names of the functions

21
00:01:03,750 --> 00:01:06,620
that come from the module to indicate

22
00:01:06,620 --> 00:01:09,840
that they are part of
the module's statistics.

23
00:01:09,840 --> 00:01:12,150
So the color coding that you get

24
00:01:12,150 --> 00:01:15,097
may vary from IPython at the command line

25
00:01:15,097 --> 00:01:19,150
to Jupyter Notebooks
to the different IDE's

26
00:01:19,150 --> 00:01:22,850
that you might use to
implement your Python code.

27
00:01:22,850 --> 00:01:26,310
So we can check the mean of
those values by executing this,

28
00:01:26,310 --> 00:01:28,925
notice that we're getting an
infinitely repeating sequence.

29
00:01:28,925 --> 00:01:32,630
By default, it shows a maximum
of this number of digits

30
00:01:32,630 --> 00:01:33,970
to the right of the decimal point,

31
00:01:33,970 --> 00:01:37,210
rounding to the last
position and of course

32
00:01:37,210 --> 00:01:39,730
we could use format strings to get

33
00:01:39,730 --> 00:01:42,450
a nicer looking format of the data.

34
00:01:42,450 --> 00:01:44,400
The median value in this case,

35
00:01:44,400 --> 00:01:46,908
well, they're not sorted up above here

36
00:01:46,908 --> 00:01:50,520
but we could, of course, add another cell

37
00:01:50,520 --> 00:01:52,250
to see what it would look like sorted.

38
00:01:52,250 --> 00:01:55,740
So let me go ahead and do
that and let's do sorted

39
00:01:57,470 --> 00:02:01,850
with values as an argument
and if I execute that,

40
00:02:01,850 --> 00:02:03,840
you can see these in sorted order

41
00:02:03,840 --> 00:02:07,040
and remember the median
is the middle value

42
00:02:07,040 --> 00:02:10,320
but there's not a single middle value here

43
00:02:10,320 --> 00:02:13,444
because we have an even number of items

44
00:02:13,444 --> 00:02:16,940
so what's happening is it's
taking the two middle values,

45
00:02:16,940 --> 00:02:19,720
calculating the average
of those two values,

46
00:02:19,720 --> 00:02:23,710
and displaying the average so 86

47
00:02:23,710 --> 00:02:27,560
is the average of 84 and 88 of course

48
00:02:27,560 --> 00:02:29,052
and going back up here for a moment,

49
00:02:29,052 --> 00:02:32,650
when we evaluate the mode,
whatever appears the most

50
00:02:32,650 --> 00:02:37,090
times 88, in this case, is
the mode value so again,

51
00:02:37,090 --> 00:02:39,860
in sorted order, that helps us sees things

52
00:02:39,860 --> 00:02:41,800
like the mode as well because we can see

53
00:02:41,800 --> 00:02:43,860
all of the items of the same value,

54
00:02:43,860 --> 00:02:46,350
side by side with one another.

55
00:02:46,350 --> 00:02:50,780
Now, as long as we are looking at this,

56
00:02:50,780 --> 00:02:53,000
I just want to take a moment and make

57
00:02:53,000 --> 00:02:55,370
one other modification here to show you

58
00:02:55,370 --> 00:03:00,370
what happens, if in fact you
don't have a unique mode value.

59
00:03:00,810 --> 00:03:04,160
So we have two 88's, let's go back up here

60
00:03:04,160 --> 00:03:08,300
and add in another 73 and
I'll re-execute that cell

61
00:03:08,300 --> 00:03:11,800
so that I now have an
updated list in memory

62
00:03:11,800 --> 00:03:15,750
and let's now go re-execute
the mode and when we do,

63
00:03:15,750 --> 00:03:18,550
notice we get a statistics error.

64
00:03:18,550 --> 00:03:22,490
The reason is that there
was no unique mode value.

65
00:03:22,490 --> 00:03:24,860
There were two values that had

66
00:03:24,860 --> 00:03:27,990
equally common values
and if we go back above,

67
00:03:27,990 --> 00:03:31,410
we can see we have two
88's and we have two 73's

68
00:03:31,410 --> 00:03:34,950
so the mode is not always the best type

69
00:03:34,950 --> 00:03:38,440
of central statistic to
use because you can't

70
00:03:38,440 --> 00:03:40,250
ever guarantee when you're dealing

71
00:03:40,250 --> 00:03:42,940
with potentially millions of values,

72
00:03:42,940 --> 00:03:45,350
that you have a unique mode value

73
00:03:45,350 --> 00:03:48,743
so you may not see that one
used quite as frequently.