1
00:00:01,000 --> 00:00:05,868
So far, the visualizations
we've covered have mostly taken

2
00:00:05,868 --> 00:00:09,524
place on a grid with a 
labeled x axis and y axis.

3
00:00:09,524 --> 00:00:14,900
The exception is the pie chart, but
even the pie chart has a fixed shape and

4
00:00:14,900 --> 00:00:18,297
strict parameters on how
it can be presented.

5
00:00:18,297 --> 00:00:23,440
However, sometimes the story you're
telling involves geographic data.

6
00:00:23,440 --> 00:00:28,514
It might be possible to convey election
results using a bar chart, for

7
00:00:28,514 --> 00:00:34,190
example, but users will likely find
election information more meaningful

8
00:00:34,190 --> 00:00:38,409
when results are visualized by city,
state, or county.

9
00:00:38,409 --> 00:00:43,488
Let's take a look at some different
ways that data visualization and

10
00:00:43,488 --> 00:00:45,687
geographic maps intersect.

11
00:00:45,687 --> 00:00:49,946
It's important to note that maps
themselves are data visualizations,

12
00:00:49,946 --> 00:00:53,030
since they help us make sense
of the world around us.

13
00:00:53,030 --> 00:00:57,990
In fact, the earliest known data
visualization is a map of the town

14
00:00:57,990 --> 00:01:01,897
of Konya in Turkey created
over 8000 years ago.

15
00:01:01,897 --> 00:01:07,211
Of course, these days sophisticated
apps like Google Maps provide more than

16
00:01:07,211 --> 00:01:13,014
just geographic information, offering
features like up to date traffic reports,

17
00:01:13,014 --> 00:01:17,282
and suggesting alternate routes
to reach your destination.

18
00:01:17,282 --> 00:01:21,399
When it comes to plotting data
points on a geographic map,

19
00:01:21,399 --> 00:01:26,020
the most common methods you'll
encounter are choropleth maps,

20
00:01:26,020 --> 00:01:29,982
proportional symbol maps,
and geographic heat maps.

21
00:01:29,982 --> 00:01:35,185
Choropleth maps are created by coloring
in existing geographic regions

22
00:01:35,185 --> 00:01:40,142
such as counties within a state based
on the frequency of a variable.

23
00:01:40,142 --> 00:01:44,575
For example, a choropleth map could
demonstrate unemployment numbers in

24
00:01:44,575 --> 00:01:47,890
various counties throughout
the state of California.

25
00:01:47,890 --> 00:01:49,568
When plotting this data,

26
00:01:49,568 --> 00:01:54,616
note that choropleth maps are only
effective when showing relative numbers.

27
00:01:54,616 --> 00:01:58,042
The first map on the left
shows absolute numbers, but

28
00:01:58,042 --> 00:02:03,081
it's not surprising that Los Angeles
County has the most unemployed people.

29
00:02:03,081 --> 00:02:07,489
After all, Los Angeles County has
more than 10 million residents,

30
00:02:07,489 --> 00:02:12,584
while some counties in Northeastern
California have fewer than 10,000.

31
00:02:12,584 --> 00:02:16,774
This second map colors in
the counties of California based on

32
00:02:16,774 --> 00:02:21,963
the unemployment rates per 100 people,
which is far more meaningful.

33
00:02:21,963 --> 00:02:26,204
Absolute numbers can be shown
using a proportional symbol map,

34
00:02:26,204 --> 00:02:28,491
sometimes called a bubble map.

35
00:02:28,491 --> 00:02:33,868
Proportional symbol maps are still based
on geographic regions such as counties,

36
00:02:33,868 --> 00:02:38,723
but the frequency of the variable is
represented by the size of the symbol.

37
00:02:38,723 --> 00:02:44,052
This map of Oklahoma uses large circles
to represent counties with at least

38
00:02:44,052 --> 00:02:49,300
40,000 people who identify as
American Indian or Alaska Native and

39
00:02:49,300 --> 00:02:53,629
small circles to represent
counties with smaller numbers.

40
00:02:53,629 --> 00:02:58,607
Challenges with proportional symbol
maps include the difficulty users

41
00:02:58,607 --> 00:03:01,145
might have comparing circle sizes.

42
00:03:01,145 --> 00:03:06,976
And also that circles tend to overlap
when plotting densely populated areas,

43
00:03:06,976 --> 00:03:10,080
such as the Northeastern United States.

44
00:03:10,080 --> 00:03:15,580
Sometimes however, your geographic data
doesn't need to be constrained by region.

45
00:03:15,580 --> 00:03:19,938
Geographic heat maps demonstrate
the frequency of data points

46
00:03:19,938 --> 00:03:23,318
on a map while ignoring
geographic boundaries.

47
00:03:23,318 --> 00:03:28,049
For example, the collaborative
data science platform called Mode

48
00:03:28,049 --> 00:03:32,780
offers an interactive geographic
heat map showing the locations of

49
00:03:32,780 --> 00:03:36,692
Carl's Jr fast food restaurants
in the United States.

50
00:03:36,692 --> 00:03:40,981
Red spots on the map represent
areas with a high volume of

51
00:03:40,981 --> 00:03:44,449
Carl's Jr restaurants in close proximity,

52
00:03:44,449 --> 00:03:50,035
while cooler colors indicate areas
with fewer Carl's Jr restaurants.

53
00:03:50,035 --> 00:03:54,132
So that covers a number of basic
types of data visualizations.

54
00:03:54,132 --> 00:03:59,116
In the final stage of this course, we'll
learn a few more rules to guide you as you

55
00:03:59,116 --> 00:04:01,550
put your new knowledge into practice.