WEBVTT

99:59:59.999 --> 99:59:59.999
1
00:00:04,510 --> 00:00:05,680
Welcome back. This video,

99:59:59.999 --> 99:59:59.999
2
00:00:05,680 --> 00:00:10,480
I'm going to walk you through some of the different types of charts that we're going to be learning how to create outcomes or

99:59:59.999 --> 99:59:59.999
3
00:00:10,480 --> 00:00:16,720
be able to identify the appropriate type of chart for data in a question and understand key rules to avoid common errors.

99:59:59.999 --> 99:59:59.999
4
00:00:16,720 --> 00:00:20,880
I'm not going to be showing the detailed code for these chart types in the video.

99:59:59.999 --> 99:59:59.999
5
00:00:20,880 --> 00:00:24,640
You're going to be able to find that in the documentation link from here. And also,

99:59:59.999 --> 99:59:59.999
6
00:00:24,640 --> 00:00:28,600
I'm going to be preparing a notebook that demonstrates various of these charting

99:59:59.999 --> 99:59:59.999
7
00:00:28,600 --> 00:00:34,330
types with the actual code to create them using the software we discussing.

99:59:59.999 --> 99:59:59.999
8
00:00:34,330 --> 00:00:44,470
So common software for this or Seabourne and matplotlib, those are going to be the primary ones that we're working with this semester.

99:59:59.999 --> 99:59:59.999
9
00:00:44,470 --> 00:00:49,330
When I'm showing the function names, Seabourne is commonly input imported S.A.S.

99:59:59.999 --> 99:59:59.999
10
00:00:49,330 --> 00:00:52,870
So as an ascot function is going to be a seabourne function PLDT,

99:59:59.999 --> 99:59:59.999
11
00:00:52,870 --> 00:00:59,860
the function is going to be a matplotlib of function and also showing the function you can use in plot nine or Ares g.G plot too,

99:59:59.999 --> 99:59:59.999
12
00:00:59,860 --> 00:01:04,600
if you want to use those instead. I often use plot nine for a lot of my graphics.

99:59:59.999 --> 99:59:59.999
13
00:01:04,600 --> 00:01:10,060
That's just for reference though. We're not going to be getting into much detail on Plot nine in the course of this course.

99:59:59.999 --> 99:59:59.999
14
00:01:10,060 --> 00:01:16,130
So there's a variety of different types of charts. Some of them are showing relative proportions.

99:59:59.999 --> 99:59:59.999
15
00:01:16,130 --> 00:01:23,860
Some of them are showing how different amounts relate to each other. Some of them are showing positions and an x y coordinate space.

99:59:59.999 --> 99:59:59.999
16
00:01:23,860 --> 00:01:30,070
A bar chart is a very common type of chart that shows numeric values grouped by a categorical or ordinal variable.

99:59:59.999 --> 99:59:59.999
17
00:01:30,070 --> 00:01:31,780
Sometimes they're grouped by New America as well.

99:59:59.999 --> 99:59:59.999
18
00:01:31,780 --> 00:01:37,600
But usually our x axis is a categorical variable of some kind best with a moderate number of categories.

99:59:59.999 --> 99:59:59.999
19
00:01:37,600 --> 00:01:41,950
We can use a second categorical variable to say color the bars.

99:59:59.999 --> 99:59:59.999
20
00:01:41,950 --> 00:01:50,170
So this chart shows the survival rates of Titanic passengers or the X axis is the passage class for second or third class.

99:59:59.999 --> 99:59:59.999
21
00:01:50,170 --> 00:01:55,660
And then the bars are colored based on the gender of the of the passenger.

99:59:59.999 --> 99:59:59.999
22
00:01:55,660 --> 00:01:58,000
And so we can see the different survival rates.

99:59:59.999 --> 99:59:59.999
23
00:01:58,000 --> 00:02:06,250
The y axis on a bar chart is often a mean or a sum or a count within the cap of the group determined by our categorical variables.

99:59:59.999 --> 99:59:59.999
24
00:02:06,250 --> 00:02:12,970
Sometimes these will be horizontal. So the horizontal bar chart, the categorical is on the Y and the bars run horizontally.

99:59:59.999 --> 99:59:59.999
25
00:02:12,970 --> 00:02:17,080
This also shows some whiskers that come from a confidence interval.

99:59:59.999 --> 99:59:59.999
26
00:02:17,080 --> 00:02:24,760
It's very easy to generate a default, relatively good confidence interval with Seabourne so tough to pluck to plot

99:59:59.999 --> 99:59:59.999
27
00:02:24,760 --> 00:02:30,730
these Seabourne has the count plot function which lets which does a quick,

99:59:59.999 --> 99:59:59.999
28
00:02:30,730 --> 00:02:36,970
basically categorical histogram. How many observations are in each are in each category.

99:59:59.999 --> 99:59:59.999
29
00:02:36,970 --> 00:02:41,830
The cap plot variable will plot by default a mean value for each category.

99:59:59.999 --> 99:59:59.999
30
00:02:41,830 --> 00:02:47,860
And if you have it, do the mean plotting. It will also compute. Ninety five percent confidence intervals.

99:59:59.999 --> 99:59:59.999
31
00:02:47,860 --> 00:02:53,350
That's what's being shown in this in this plot here.

99:59:59.999 --> 99:59:59.999
32
00:02:53,350 --> 00:03:00,050
And then you can also use the bat, the bar function or the plot nine Geon Bar.

99:59:59.999 --> 99:59:59.999
33
00:03:00,050 --> 00:03:04,340
So if you rules about bar charts first is never start the Y axis on a bar chart.

99:59:59.999 --> 99:59:59.999
34
00:03:04,340 --> 00:03:13,320
Anything but zero. And so the reason for this we can see here is that.

99:59:59.999 --> 99:59:59.999
35
00:03:13,320 --> 00:03:17,270
So the top one. So these are these are looking at the mean average ratings.

99:59:59.999 --> 99:59:59.999
36
00:03:17,270 --> 00:03:22,770
We take you to movies, mean rating, and then we compute the mean of the average ratings within a genre.

99:59:59.999 --> 99:59:59.999
37
00:03:22,770 --> 00:03:33,570
What is that? So if we look here, the difference between horror and IMAX, it's a notable difference, but it's a difference of about point five or so.

99:59:59.999 --> 99:59:59.999
38
00:03:33,570 --> 00:03:40,080
The difference between sci fi and short is a difference of a little under one, probably.

99:59:59.999 --> 99:59:59.999
39
00:03:40,080 --> 00:03:46,170
But when we start the Y axis at 2.5 instead of zero,

99:59:59.999 --> 99:59:59.999
40
00:03:46,170 --> 00:03:51,840
what happens is the differences look much larger than they are because the human eye, naturally it's not.

99:59:59.999 --> 99:59:59.999
41
00:03:51,840 --> 00:04:01,050
We not only want to see the difference, but we want to it's very natural for us to compare the difference to the bar length because these are bars.

99:59:59.999 --> 99:59:59.999
42
00:04:01,050 --> 00:04:06,620
They have length, they have an area since they're all the same with the length is proportional to the area.

99:59:59.999 --> 99:59:59.999
43
00:04:06,620 --> 00:04:12,660
Braking length area. Proportionality is a good way to confuse your readers,

99:59:59.999 --> 99:59:59.999
44
00:04:12,660 --> 00:04:20,910
but it looks like IMAX movies have twice as high an average rating as horror movies because the bar is twice as high, but they don't.

99:59:59.999 --> 99:59:59.999
45
00:04:20,910 --> 00:04:25,770
It's really a shift from about 2.8 to three point three or three point four.

99:59:59.999 --> 99:59:59.999
46
00:04:25,770 --> 00:04:33,090
And so it creates a distortion that makes the different like it highlights the differences, but it makes the differences look larger than they are.

99:59:59.999 --> 99:59:59.999
47
00:04:33,090 --> 00:04:39,210
So when I talked about integrity and avoiding deception, when I was introducing statistical graphics,

99:59:59.999 --> 99:59:59.999
48
00:04:39,210 --> 00:04:45,750
this is what I was talking about, the differences there. It's just not as big as it looks like it is.

99:59:59.999 --> 99:59:59.999
49
00:04:45,750 --> 00:04:47,460
And we truncate our bar charts.

99:59:59.999 --> 99:59:59.999
50
00:04:47,460 --> 00:04:55,050
So if you have the general rule here to generalize beyond bar charts is if something has a length that varies based on the data,

99:59:59.999 --> 99:59:59.999
51
00:04:55,050 --> 00:05:04,470
that length needs to actually represent the value, not the value, minus something because you started the axis somewhere else.

99:59:59.999 --> 99:59:59.999
52
00:05:04,470 --> 00:05:06,960
So if you're including Whiskers, like I did in the previous chart,

99:59:59.999 --> 99:59:59.999
53
00:05:06,960 --> 00:05:14,040
define how they're computed and also as one thing to just be careful of seaboard's cat platen count plot.

99:59:59.999 --> 99:59:59.999
54
00:05:14,040 --> 00:05:21,270
If you aren't using the color for a second variable, they will just make every bar a different color for no particular reason,

99:59:59.999 --> 99:59:59.999
55
00:05:21,270 --> 00:05:25,010
which it creates something that's different when it doesn't need to be so.

99:59:59.999 --> 99:59:59.999
56
00:05:25,010 --> 00:05:29,520
It causes the reader to look for a difference that isn't actually their best avoided.

99:59:59.999 --> 99:59:59.999
57
00:05:29,520 --> 00:05:32,640
You can fix that by just specifying the color.

99:59:59.999 --> 99:59:59.999
58
00:05:32,640 --> 00:05:40,050
We saw histograms last week in a histogram as a bar chart, but a categorical was Binz or ranges of a numerical value.

99:59:59.999 --> 99:59:59.999
59
00:05:40,050 --> 00:05:46,800
Also, though, if we have a bar chart that's showing the relative frequency of categorical variables that can also be called a histogram,

99:59:59.999 --> 99:59:59.999
60
00:05:46,800 --> 00:05:50,730
the Y axis is either the number or the fraction of occurrences in this case.

99:59:59.999 --> 99:59:59.999
61
00:05:50,730 --> 00:05:58,170
So we can that. The key thing, though, is the different heights of the bars that I see visually, the relative frequency of different values.

99:59:59.999 --> 99:59:59.999
62
00:05:58,170 --> 00:06:01,680
So it really makes it visually clear how the data is shaped.

99:59:59.999 --> 99:59:59.999
63
00:06:01,680 --> 00:06:07,950
We can see Skewes and things like that. Is there one way to graphically describe a distribution?

99:59:59.999 --> 99:59:59.999
64
00:06:07,950 --> 00:06:12,120
A scatterplot shows two numeric variables. So each observation is a dot.

99:59:59.999 --> 99:59:59.999
65
00:06:12,120 --> 00:06:16,800
Each observation has two numeric variables. And we put the one variable on the x axis.

99:59:59.999 --> 99:59:59.999
66
00:06:16,800 --> 00:06:22,290
The other variable on the Y axis and put the dot at where its variable values would intersect.

99:59:59.999 --> 99:59:59.999
67
00:06:22,290 --> 00:06:26,700
This is really useful for seeing how two variables relate. Does one increase with the other Duplin?

99:59:59.999 --> 99:59:59.999
68
00:06:26,700 --> 00:06:30,840
Do points clump or cluster in an interesting way? Other interesting patterns.

99:59:59.999 --> 99:59:59.999
69
00:06:30,840 --> 00:06:43,050
It helps us find outliers. So this this is scatterplot is showing the tip versus the total bill for a bunch of restaurant bills.

99:59:59.999 --> 99:59:59.999
70
00:06:43,050 --> 00:06:45,540
And each each observation is a bill.

99:59:59.999 --> 99:59:59.999
71
00:06:45,540 --> 00:06:55,380
And then the x axis is the the under the total bill on the Y axis is the tip that the that the the customer added to the bill.

99:59:59.999 --> 99:59:59.999
72
00:06:55,380 --> 00:07:02,790
And we a couple of refinements we can do here. We can color or change to the point tight by a categorical variable.

99:59:59.999 --> 99:59:59.999
73
00:07:02,790 --> 00:07:08,140
So on this one, we've changed it so that the points are different color.

99:59:59.999 --> 99:59:59.999
74
00:07:08,140 --> 00:07:13,310
So those dinners are blue circles and the lunches are orange AXA's.

99:59:59.999 --> 99:59:59.999
75
00:07:13,310 --> 00:07:19,130
We could also add a trend line or some other kind of a line to show some context, for example, on this chart,

99:59:59.999 --> 99:59:59.999
76
00:07:19,130 --> 00:07:27,170
we might want to plot a line that shows that the 20 percent point and that let us easily see where we're going over 20 percent,

99:59:59.999 --> 99:59:59.999
77
00:07:27,170 --> 00:07:33,650
how that the tips are distributed relative to it to a 20 percent mark.

99:59:59.999 --> 99:59:59.999
78
00:07:33,650 --> 00:07:41,170
We can also X can be a categorical variable when that happens. We call this a point plot or a strip plot.

99:59:59.999 --> 99:59:59.999
79
00:07:41,170 --> 00:07:47,020
Functions for doing this are scatter scatterplot and then plotlines John Point,

99:59:59.999 --> 99:59:59.999
80
00:07:47,020 --> 00:07:50,800
the Seabourne documentation has some examples of more of these align plot.

99:59:59.999 --> 99:59:59.999
81
00:07:50,800 --> 00:07:59,350
It's like a scatterplot that we have to numeric variables. However, we it emphasizes the progression or continuity from one variable to the next.

99:59:59.999 --> 99:59:59.999
82
00:07:59,350 --> 00:08:04,930
By combining them with a line, it really works best. We have one Y per X value that we want to plot.

99:59:59.999 --> 99:59:59.999
83
00:08:04,930 --> 00:08:11,260
If we've got more than one, it really starts getting very, very jagged. It's very common for Time series.

99:59:59.999 --> 99:59:59.999
84
00:08:11,260 --> 00:08:15,430
So this is another example from the Seabourne tutorial not labeled super well.

99:59:59.999 --> 99:59:59.999
85
00:08:15,430 --> 00:08:22,510
I don't know what the value actually is, but it shows that we have some kind of a value that's changing over time and it's going negative.

99:59:59.999 --> 99:59:59.999
86
00:08:22,510 --> 00:08:29,710
That was zero. The Y axis is at the top and the values otherwise our negative functions to create.

99:59:59.999 --> 99:59:59.999
87
00:08:29,710 --> 00:08:36,370
These are line plot from seabourne, line from a matplotlib and Gyeom line from plot nine.

99:59:59.999 --> 99:59:59.999
88
00:08:36,370 --> 00:08:41,380
A box plot shows the distribution of a numeric variable grouped by a categorical.

99:59:59.999 --> 99:59:59.999
89
00:08:41,380 --> 00:08:46,630
So the bar chart just showed us, say, the average value, maybe with confidence interval.

99:59:59.999 --> 99:59:59.999
90
00:08:46,630 --> 00:08:51,820
The box plot actually shows us the distribution and it does so in a way that's based on the median.

99:59:59.999 --> 99:59:59.999
91
00:08:51,820 --> 00:09:01,050
So the median, the the horizontal line in the middle of the box is the median value, the top and bottom of the box.

99:59:59.999 --> 99:59:59.999
92
00:09:01,050 --> 00:09:06,750
Are the first and third quarter close to the bottom of the first quartile and the top as the third quartile.

99:59:59.999 --> 99:59:59.999
93
00:09:06,750 --> 00:09:12,630
And what that means is twenty five percent of the values are below the bottom of the box.

99:59:59.999 --> 99:59:59.999
94
00:09:12,630 --> 00:09:17,700
Twenty five percent in the bottom half. Twenty five percent here and then twenty five percent above.

99:59:59.999 --> 99:59:59.999
95
00:09:17,700 --> 00:09:21,390
We then show these these whiskers that extend out to the minimum,

99:59:59.999 --> 99:59:59.999
96
00:09:21,390 --> 00:09:27,390
a maximum of the data and a number of plotting packages will do some kind of an outlier detection.

99:59:59.999 --> 99:59:59.999
97
00:09:27,390 --> 00:09:37,470
This is using seabourne default outlier detection. So if the max is very high and what the rule it uses by default is it allows the whisker to be.

99:59:59.999 --> 99:59:59.999
98
00:09:37,470 --> 00:09:44,460
So you've got the IQ are the inter quartile range. That's the height of the box. It allows the whisker to be one point five times that tall.

99:59:59.999 --> 99:59:59.999
99
00:09:44,460 --> 00:09:50,880
And if you have any data points that are further away than that, it plots them as individual points, makes it easy to see outliers.

99:59:59.999 --> 99:59:59.999
100
00:09:50,880 --> 00:09:57,930
You can change. It's that the whisker goes all the way up to the max, but it lets you quickly see and compare between different groups.

99:59:59.999 --> 99:59:59.999
101
00:09:57,930 --> 00:10:03,210
The median, the first and third quartiles and the men in the max to the data.

99:59:59.999 --> 99:59:59.999
102
00:10:03,210 --> 00:10:11,520
Very useful for comparing observations of a variable when you're grouped by some categorical functions

99:59:59.999 --> 99:59:59.999
103
00:10:11,520 --> 00:10:19,680
for doing this or box plot from both Seabourne and matplotlib and then Gyeom block box from plot nine.

99:59:59.999 --> 99:59:59.999
104
00:10:19,680 --> 00:10:25,710
A few more plots, a violin plot. It's like a box plot, except it's based around the mean and has curved sides.

99:59:59.999 --> 99:59:59.999
105
00:10:25,710 --> 00:10:30,120
The swarm plot is a kind of another kind of a categorical scatterplot.

99:59:59.999 --> 99:59:59.999
106
00:10:30,120 --> 00:10:38,860
It's usually best to avoid pie charts, especially 3D pie charts, or a lot of the of our software is not going to produce 3D charts very easily.

99:59:59.999 --> 99:59:59.999
107
00:10:38,860 --> 00:10:46,620
Don't try to go make a 3D chart. They're almost always more confusing, especially like the 3D bars that you have from vintage PowerPoint.

99:59:59.999 --> 99:59:59.999
108
00:10:46,620 --> 00:10:57,750
But even a pie chart, just because the human perception is not super great at accurately comparing angular areas.

99:59:59.999 --> 99:59:59.999
109
00:10:57,750 --> 00:11:00,120
So usually a bar chart,

99:59:59.999 --> 99:59:59.999
110
00:11:00,120 --> 00:11:07,680
restacked bar chart is going to be a better option than a pie chart or a donut chart is sometimes a better option where you've got to circle.

99:59:59.999 --> 99:59:59.999
111
00:11:07,680 --> 00:11:14,250
This is one place where I disagree with the reading. The reading that I gave you recommends pie charts for showing relative proportions.

99:59:59.999 --> 99:59:59.999
112
00:11:14,250 --> 00:11:19,170
I recommend usually avoiding those use a bar chart is a stacked bar chart if you need to show you

99:59:59.999 --> 99:59:59.999
113
00:11:19,170 --> 00:11:25,230
want to show multiple proportions of different or relative proportions within different categories.

99:59:59.999 --> 99:59:59.999
114
00:11:25,230 --> 00:11:29,670
There's another kind of plot that's not a plot on its own, but it's combined with other kinds of plots.

99:59:59.999 --> 99:59:59.999
115
00:11:29,670 --> 00:11:34,830
That's a rug plot useful for just displaying distributions at a margin.

99:59:59.999 --> 99:59:59.999
116
00:11:34,830 --> 00:11:40,500
So to learn more, I've gone I've taken a whirlwind tour through a number of different plot types, the class readings.

99:59:59.999 --> 99:59:59.999
117
00:11:40,500 --> 00:11:46,170
So the paper that I assigned you to read, it talks through the use cases for a number of different plot types.

99:59:59.999 --> 99:59:59.999
118
00:11:46,170 --> 00:11:50,160
I'm going to be providing tutorial notebooks that walk you through different plot types.

99:59:59.999 --> 99:59:59.999
119
00:11:50,160 --> 00:11:54,660
The textbook talks about graph plotting and data visualization.

99:59:59.999 --> 99:59:59.999
120
00:11:54,660 --> 00:11:58,190
The Seabourne and matplotlib docs are extensive. And for what?

99:59:59.999 --> 99:59:59.999
121
00:11:58,190 --> 00:12:04,230
If you're using another plodding library, its documentation as well. Most plotting libraries also have a gallery student.

99:59:59.999 --> 99:59:59.999
122
00:12:04,230 --> 00:12:10,740
Go through the gallery, look for a plot that has a feature you want in your plot or that you think might be useful for displaying your data.

99:59:59.999 --> 99:59:59.999
123
00:12:10,740 --> 00:12:14,580
Click on it and they'll give you the code to show you how they made that plot.

99:59:59.999 --> 99:59:59.999
124
00:12:14,580 --> 00:12:22,650
You might want to combine pieces from multiple plots. In practice, it takes a lot of trial and error to really get the hang of your plot and library

99:59:59.999 --> 99:59:59.999
125
00:12:22,650 --> 00:12:27,630
and figure out how to make it show you the data in the way you really want it to.

99:59:59.999 --> 99:59:59.999
126
00:12:27,630 --> 00:12:32,730
Learning one plotting library really deep is useful for a lot of the a lot of the python ones,

99:59:59.999 --> 99:59:59.999
127
00:12:32,730 --> 00:12:37,350
especially the ones that are oriented towards static charts. They're built on top of matplotlib.

99:59:59.999 --> 99:59:59.999
128
00:12:37,350 --> 00:12:41,910
So Seabourne is a convenience API on top of matplotlib. If you're using Seabourne,

99:59:59.999 --> 99:59:59.999
129
00:12:41,910 --> 00:12:49,500
you're also going to need to use matplotlib calls a lot of the time when the seabourne gets you 90 percent of the way there,

99:59:59.999 --> 99:59:59.999
130
00:12:49,500 --> 00:12:55,470
but not quite all the way. So to wrap up, there are many different types of charts that have different use cases.

99:59:59.999 --> 99:59:59.999
131
00:12:55,470 --> 00:13:01,020
Learning graphics techniques takes time and practice takes some of the example notebooks that I'm providing.

99:59:59.999 --> 99:59:59.999
132
00:13:01,020 --> 00:13:05,730
Take some of the galleries from the examples from, say, the Seabourne Gallery.

99:59:59.999 --> 99:59:59.999
133
00:13:05,730 --> 00:13:11,670
Play with them, play with them with some data that I'm giving you, play with them with some data that you have elsewhere.

99:59:59.999 --> 99:59:59.999
134
00:13:11,670 --> 00:13:30,957
But it takes time and practice and spend some time with the galleries of the of the the plotting libraries you're using.

99:59:59.999 --> 99:59:59.999