DATA VISUALIZATION WITH GGPLOT2

Choropleths
 Data Visualization with ggplot2

Chapter Contents
● Maps
● GIS = Geographic Information System
● Choropleths
● Cartographic maps
● Animations
 Data Visualization with ggplot2

Choropleths > library(ggplot2)

> usa <- map_data("usa")
> ggplot(usa, aes(long, lat, group = group)) +
geom_polygon() +
Bunch of polygons coord_map()
50

45

40
lat

35

30

25
−120 −100 −80
long
 Data Visualization with ggplot2

Choropleths >
>
>
library(ggplot2)
library(ggalt)
usa <- map_data("usa")
> ggplot(usa, aes(long, lat, group = group)) +
geom_polygon() +
50 coord_proj("+proj=wintri")

45

40
lat

35

30

25

−120 −100 −80

long
 Data Visualization with ggplot2

Many polygons > states <- map_data("state")

> ggplot(states,
aes(long, lat,
fill = region, group = group)) +
geom_polygon(color = "white") +
coord_map()
 Data Visualization with ggplot2

Weed prices

value

350

300

250
 Data Visualization with ggplot2

Weed prices

value
400

350

300

250

200
 Data Visualization with ggplot2

Alternatives
oregon ●
washington ●
colorado ●
california ●
montana ●
nevada ●
idaho ●
new mexico ●
utah ●
michigan ●
alaska ●
arizona ●
florida ●
hawaii ●
maine ●
rhode island ●
georgia ●
indiana ●
ohio ●
texas ●
wyoming ●
mississippi ●
kentucky ●
alabama ●
region

nebraska ●
connecticut ●
south carolina ●
new york ●
arkansas ●
district of columbia ●
massachusetts ●
new jersey ●
north carolina ●
kansas ●
tennessee ●
wisconsin ●
illinois ●
minnesota ●
west virginia ●
missouri ●
new hampshire ●
oklahoma ●
pennsylvania ●
delaware ●
louisiana ●
iowa ●
virginia ●
maryland ●
vermont ●
south dakota ●
north dakota ●

200 250 300 350 400

value
 Data Visualization with ggplot2

Alternatives
michigan ●
indiana ●
ohio ●
nebraska ●

Midwest
kansas ●
wisconsin ●
illinois ●
minnesota ●
missouri ●
iowa ●
south dakota ●
north dakota ●

maine ●

rhode island ●

connecticut ●

Northeast
new york ●

massachusetts ●

new jersey ●

new hampshire ●

pennsylvania ●

vermont
state

florida ●
georgia ●
texas ●
mississippi ●
kentucky ●
alabama ●
south carolina

South
●
arkansas ●
north carolina ●
tennessee ●
west virginia ●
oklahoma ●
delaware ●
louisiana ●
virginia ●
maryland ●

oregon ●
washington ●
colorado ●
california ●
montana ●
nevada

West
●
idaho ●
new mexico ●
utah ●
alaska ●
arizona ●
hawaii ●
wyoming ●

200 250 300 350 400

value
 Data Visualization with ggplot2

Alternatives
michigan ●

indiana ●

ohio ● Midwest East North Central

wisconsin ●

illinois ●

nebraska ●
kansas ●
minnesota ●
missouri ● Midwest West North Central
iowa ●
south dakota ●
north dakota ●

new york ●

new jersey ● Northeast Mid−Atlantic

pennsylvania ●

maine ●
rhode island ●
connecticut ●
Northeast New England
massachusetts ●
new hampshire ●
vermont ●

mississippi ●
state

kentucky ●
South East South Central
alabama ●

tennessee ●

florida ●
georgia ●
south carolina ●
north carolina ●
South South Atlantic
west virginia ●
delaware ●
virginia ●
maryland ●

texas ●

arkansas ●
South West South Central
oklahoma ●

louisiana ●

colorado ●
montana ●
nevada ●
idaho ●
West Mountain
new mexico ●
utah ●
arizona ●
wyoming ●

oregon ●

washington ●

california ● West Pacific

alaska ●

hawaii ●

200 250 300 350 400

value
DATA VISUALIZATION WITH GGPLOT2

Let’s practice!
 DATA VISUALIZATION WITH GGPLOT2

Cartographic Maps
 Data Visualization with ggplot2

Cartographic map
● Drawn
● Topographical maps
● Altitude, infrastructure ...
● Photographic
● Satellite images
● Hybrid
● ggmap
 Data Visualization with ggplot2

> # Default style - zoom = 3

> library(ggmap)
> def_03 <- get_map(location = "Berlin, Germany", zoom = 3)
> ggmap(def_03, extent = "device")
 Data Visualization with ggplot2

> # Default style - zoom = 13

> library(ggmap)
> def_13 <- get_map(location = "Berlin, Germany", zoom = 13)
> ggmap(def_13, extent = "device")
 Data Visualization with ggplot2

> # Default style - zoom = 20

> library(ggmap)
> def_20 <- get_map(location = "Berlin, Germany", zoom = 20)
> ggmap(def_20, extent = "device")
 Data Visualization with ggplot2

> # stamen/watercolor - zoom = 13

> library(ggmap)
> wc_13 <- get_map(location = "Berlin, Germany", zoom = 13,
source = "stamen", maptype = "watercolor")
> ggmap(wc_13, extent = "device")
 Data Visualization with ggplot2

> # stamen/toner - zoom = 13

> library(ggmap)
> ton_13 <- get_map(location = "Berlin, Germany", zoom = 13,
source = "stamen", maptype = "toner")
> ggmap(ton_13, extent = "device")
 Data Visualization with ggplot2

> # stamen/hybrid - zoom = 13

> library(ggmap)
> hyb_13 <- get_map(location = "Berlin, Germany", zoom = 13,
source = "stamen", maptype = "hybrid")
> ggmap(ton_13, extent = "device")
 Data Visualization with ggplot2

> # google/satellite - zoom = 13

> library(ggmap)
> sat_13 <- get_map(location = "Berlin, Germany", zoom = 13,
source = "google", maptype = "satellite")
> ggmap(ton_13, extent = "device")
 Data Visualization with ggplot2

Get coordinates
> berlin_sites <- c("Brandenburger Tor", "Potsdamer Platz", 

"Victory Column Berlin", "Checkpoint Charlie", 

"Reichstag Berlin", "Alexander Platz")

> xx <- geocode(berlin_sites)

Information from URL : http://maps.googleapis.com/maps/...
Information from URL : ...

> # Add column with cleaned up names

> xx$location <- sub(" Berlin", "", berlin_sites)

> str(xx)
'data.frame': 6 obs. of 3 variables:
$ lon : num 13.4 13.4 13.4 13.4 13.4 ...
$ lat : num 52.5 52.5 52.5 52.5 52.5 ...
$ location: chr "Brandenburger Tor" "Potsdamer Platz" ...
 Data Visualization with ggplot2

> # google/roadmap - zoom = 13

> road_13 <- get_map(location = "Berlin, Germany", zoom = 13, 

source = "google", maptype = "roadmap")
> ggmap(road_13, extent = "device")
 Data Visualization with ggplot2

> # google/roadmap - zoom = 13

> road_13 <- get_map(location = "Berlin, Germany", zoom = 13, 

source = "google", maptype = "roadmap")
> ggmap(road_13, extent = "device") +
geom_point(data = xx, col = "red")

●
 Data Visualization with ggplot2

> bbox <- make_bbox(lon = xx$lon, lat = xx$lat, f = .1)

> boxed_14 <- get_map(location = bbox, zoom = 14,
source = "google", maptype = "roadmap")
> ggmap(boxed_14, extent = "device") +
geom_point(data = xx, col = "red")

●
 Data Visualization with ggplot2

> bbox <- make_bbox(lon = xx$lon, lat = xx$lat, f = .1)

> boxed_14 <- get_map(location = bbox, zoom = 14,
source = "google", maptype = "roadmap")
> ggmap(boxed_14, extent = "device") +
geom_point(data = xx, aes(col = location), size = 3) +
scale_colour_brewer(palette = "Set1")

location
● Alexander Platz
●
● Brandenburger Tor
●
● Checkpoint Charlie
● Potsdamer Platz
● Reichstag
● ● Victory Column
●
 Data Visualization with ggplot2

Final Plot

Alexander
Platz

Reichstag
Brandenburger
Tor
Victory
Column

Potsdamer
Platz
Checkpoint
Charlie
DATA VISUALIZATION WITH GGPLOT2

Let’s practice!
DATA VISUALIZATION WITH GGPLOT2

Animations
 Data Visualization with ggplot2

Animations
● Dense temporal data
● Great exploratory tool
● Several ways
● for loop to produce gif
● animation
● gganimate
 Data Visualization with ggplot2

Motion Chart
● Hans Rosling
● Karolinska Institute in Stockholm
● Founder of Gapminder
● UN data
● Life expectancy, GDP ...
 Data Visualization with ggplot2

Gapminder data
> # import tab-delimited data
> gapminder <- read.delim("gapminder.tsv",
stringsAsFactors = FALSE)

> str(gapminder)
'data.frame': 1704 obs. of 6 variables:
$ country : chr "Afghanistan" "Afghanistan" "Afghanistan" ...
$ year : int 1952 1957 1962 1967 1972 1977 1982 1987 ...
$ pop : num 8425333 9240934 10267083 11537966 13079460 ...
$ continent: chr "Asia" "Asia" "Asia" "Asia" ...
$ lifeExp : num 28.8 30.3 32 34 36.1 ...
$ gdpPercap: num 779 821 853 836 740 ...
 Data Visualization with ggplot2

Static plot > ggplot(gapminder, aes(x = gdpPercap,

y = lifeExp,
colour = continent,
size = pop)) +
geom_point(alpha = 0.6) # details omitted
85
80 Population
2.50e+08
5.00e+08
7.50e+08
Life Expectancy

1.00e+09
60
1.25e+09

Continent
Africa
40 Americas
Asia
Europe
Oceania
20

1e+03 1e+04 1e+05

GDP (per capita, log10)
 Data Visualization with ggplot2

Motion chart (1) > p <- ggplot(gapminder, aes(x = gdpPercap,

y = lifeExp,
colour = continent,
size = pop,
frame = year)) +
geom_point(alpha = 0.6) # details omitted
> gg_animate(p, "chart1.gif")
 Data Visualization with ggplot2

Motion chart (2) > p <- ggplot(gapminder, aes(x = gdpPercap,

y = lifeExp,
colour = continent,
size = pop,
frame = country)) +
geom_point(alpha = 0.6) # details omitted
> gg_animate(p, "chart2.gif")
 Data Visualization with ggplot2

Motion chart (3) > p <- ggplot(gapminder, aes(x = gdpPercap,

y = lifeExp,
colour = continent,
size = pop,
frame = country)) +
geom_point(alpha = 0.6) # details omitted
> gg_animate(p, "chart3.gif", interval = 3.0)
 Data Visualization with ggplot2

Population size

Original visualization by Kyle Walker

DATA VISUALIZATION WITH GGPLOT2

Let’s practice!