Graph Catalogue V3 – Accompanies Choosing the Right Graph V3
Kelly Goedert
February 2023
Each of the following graphs uses different themes from the ggthemes package so you can get a sense for what the different themes look like. See also: https://mran.microsoft.com/snapshot/2017-02-04/web/packages/ggthemes/vignettes/ggthemes.html
QUERY 1 CATEGORICAL DATA: What is the Distribution or Composition of a Single Categorical Variable?
For Query 1, we will be using data from the diamonds package, which comes with ggplot2.
Vertical Bar Chart
Represent frequency distribution when have few categories. First, generating a frequency table to see how many catgories in the ‘cut’ variable of the diamonds package.
library(tidyverse)
library(ggthemes)
#using diamonds dataset that comes with ggplot2
table(diamonds$cut) # few categories, so do vertical bar chart##
## Fair Good Very Good Premium Ideal
## 1610 4906 12082 13791 21551#a vertical freqeuncy bar chart
ggplot(diamonds, aes(x = cut)) +
geom_bar() +
labs(x = "Cut",
y = "Count") +
theme_hc() +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Horizontal Bar Chart
Represent frequency distribution when have many categories.
Note: 3 ways to re-order bar charts by count: https://www.roelpeters.be/reorder-ggplot2-bar-chart-by-count/
#a horizontal frequency bar chart
diamonds %>%
group_by(clarity) %>%
summarise(count = n()) %>%
ggplot(aes(x = reorder(clarity,(-count)), y = count)) +
geom_bar(stat = 'identity') + # stat = identity plots the values present in cells
coord_flip() + #function that turns vertical bars to horizontal bars
labs(x = "Clarity",
y = "Count") +
theme_minimal() +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Vertical Cleveland Dot Plot
Represent frequency distribution for few categories.
#a vertical freqeuncy Cleveland Dot Plot
ggplot(diamonds, aes(x = cut)) +
geom_point(stat = "count", color = "dark blue", size = 5) +
labs(x = "Cut",
y = "Count") +
theme_minimal() +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Horiztonal Cleveland Dot Plot
Represent frequency distribution for many categories. Note: 3 ways to re-order bar charts by count: https://www.roelpeters.be/reorder-ggplot2-bar-chart-by-count/
#a horizontal cleveland dot plot
diamonds %>%
group_by(clarity) %>%
summarise(count = n()) %>%
ggplot(aes(x = reorder(clarity,(-count)), y = count)) +
geom_point(stat = "identity", color = "dark blue", size = 5) +
coord_flip() + #function that turns vertical bars to horizontal bars
labs(x = "Clarity",
y = "Count") +
theme_bw() +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
QUERY 1 NUMERIC DATA: What is the Distribution or Composition of a Single Numeric Variable?
Histogram
head(diamonds) #identify depth as a continous variable we could use here## # A tibble: 6 × 10
## carat cut color clarity depth table price x y z
## <dbl> <ord> <ord> <ord> <dbl> <dbl> <int> <dbl> <dbl> <dbl>
## 1 0.23 Ideal E SI2 61.5 55 326 3.95 3.98 2.43
## 2 0.21 Premium E SI1 59.8 61 326 3.89 3.84 2.31
## 3 0.23 Good E VS1 56.9 65 327 4.05 4.07 2.31
## 4 0.29 Premium I VS2 62.4 58 334 4.2 4.23 2.63
## 5 0.31 Good J SI2 63.3 58 335 4.34 4.35 2.75
## 6 0.24 Very Good J VVS2 62.8 57 336 3.94 3.96 2.48summary(diamonds)## carat cut color clarity depth
## Min. :0.2000 Fair : 1610 D: 6775 SI1 :13065 Min. :43.00
## 1st Qu.:0.4000 Good : 4906 E: 9797 VS2 :12258 1st Qu.:61.00
## Median :0.7000 Very Good:12082 F: 9542 SI2 : 9194 Median :61.80
## Mean :0.7979 Premium :13791 G:11292 VS1 : 8171 Mean :61.75
## 3rd Qu.:1.0400 Ideal :21551 H: 8304 VVS2 : 5066 3rd Qu.:62.50
## Max. :5.0100 I: 5422 VVS1 : 3655 Max. :79.00
## J: 2808 (Other): 2531
## table price x y
## Min. :43.00 Min. : 326 Min. : 0.000 Min. : 0.000
## 1st Qu.:56.00 1st Qu.: 950 1st Qu.: 4.710 1st Qu.: 4.720
## Median :57.00 Median : 2401 Median : 5.700 Median : 5.710
## Mean :57.46 Mean : 3933 Mean : 5.731 Mean : 5.735
## 3rd Qu.:59.00 3rd Qu.: 5324 3rd Qu.: 6.540 3rd Qu.: 6.540
## Max. :95.00 Max. :18823 Max. :10.740 Max. :58.900
##
## z
## Min. : 0.000
## 1st Qu.: 2.910
## Median : 3.530
## Mean : 3.539
## 3rd Qu.: 4.040
## Max. :31.800
## ggplot(diamonds, aes(x = price)) +
geom_histogram(bins = 70) +
labs(x = "Price",
y = "Frequency") +
theme_minimal() +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Frequency Polygon
ggplot(diamonds, aes(x = price)) +
geom_freqpoly(bins = 100)+
labs(x = "Price",
y = "Count") +
theme_minimal() +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Histodot (quantile dotplot using package ggdist)
library(ggdist)
head(diamonds) #identify depth as a continous variable we could use here## # A tibble: 6 × 10
## carat cut color clarity depth table price x y z
## <dbl> <ord> <ord> <ord> <dbl> <dbl> <int> <dbl> <dbl> <dbl>
## 1 0.23 Ideal E SI2 61.5 55 326 3.95 3.98 2.43
## 2 0.21 Premium E SI1 59.8 61 326 3.89 3.84 2.31
## 3 0.23 Good E VS1 56.9 65 327 4.05 4.07 2.31
## 4 0.29 Premium I VS2 62.4 58 334 4.2 4.23 2.63
## 5 0.31 Good J SI2 63.3 58 335 4.34 4.35 2.75
## 6 0.24 Very Good J VVS2 62.8 57 336 3.94 3.96 2.48ggplot(diamonds, aes(x = price)) +
stat_dots(position ="dodge", quantiles = 100)+
labs(x = "Price",
y = "Count") +
theme_minimal() +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Density Plot
ggplot(diamonds, aes(x = price)) +
geom_density() +
labs(x = "Price",
y = "Density") +
theme_minimal() +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
QUERY 2 CATEGORICAL DATA: Comparing Composition
The data for these graphs was taken from the starwars datset taht comes with dplyr. The original dataset was modified (see code below) to produce the variables graphed here.
Stacked column chart
head(starwars)## # A tibble: 6 × 14
## name height mass hair_…¹ skin_…² eye_c…³ birth…⁴ sex gender homew…⁵
## <chr> <int> <dbl> <chr> <chr> <chr> <dbl> <chr> <chr> <chr>
## 1 Luke Skywal… 172 77 blond fair blue 19 male mascu… Tatooi…
## 2 C-3PO 167 75 <NA> gold yellow 112 none mascu… Tatooi…
## 3 R2-D2 96 32 <NA> white,… red 33 none mascu… Naboo
## 4 Darth Vader 202 136 none white yellow 41.9 male mascu… Tatooi…
## 5 Leia Organa 150 49 brown light brown 19 fema… femin… Aldera…
## 6 Owen Lars 178 120 brown,… light blue 52 male mascu… Tatooi…
## # … with 4 more variables: species <chr>, films <list>, vehicles <list>,
## # starships <list>, and abbreviated variable names ¹hair_color, ²skin_color,
## # ³eye_color, ⁴birth_year, ⁵homeworldtable(starwars$gender)##
## feminine masculine
## 17 66table(starwars$species)##
## Aleena Besalisk Cerean Chagrian Clawdite
## 1 1 1 1 1
## Droid Dug Ewok Geonosian Gungan
## 6 1 1 1 3
## Human Hutt Iktotchi Kaleesh Kaminoan
## 35 1 1 1 2
## Kel Dor Mirialan Mon Calamari Muun Nautolan
## 1 2 1 1 1
## Neimodian Pau'an Quermian Rodian Skakoan
## 1 1 1 1 1
## Sullustan Tholothian Togruta Toong Toydarian
## 1 1 1 1 1
## Trandoshan Twi'lek Vulptereen Wookiee Xexto
## 1 2 1 2 1
## Yoda's species Zabrak
## 1 2starwarsModified <- starwars %>%
mutate(
species2 = ifelse(species=="Human", "Human",
ifelse(species=="Droid", "Droid", "Other"))
) %>%
filter(!is.na(species) & !is.na(gender))
ggplot(starwarsModified, aes(x = species2, fill = gender)) +
geom_bar()+
labs(x = "Species",
y = "Count") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Proportional Stacked column chart
ggplot(starwarsModified, aes(x = species2, fill = gender)) +
geom_bar(position = "fill")+
labs(x = "Species",
y = "Proportion") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Vertical bar chart for counts of two groups
ggplot(starwarsModified, aes(x = species2, fill = gender)) +
geom_bar(position = "dodge")+
labs(x = "Species",
y = "Count") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
QUERY 2 NUMERIC DATA: Comparing Distribution
Overlapping histograms
table(msleep$vore)##
## carni herbi insecti omni
## 19 32 5 20msleepModified <- msleep %>%
filter(!is.na(sleep_total) & !is.na(vore))
use <- c("carni", "herbi")
ggplot(data = subset(msleepModified, subset = vore %in% use), aes(x = sleep_total, fill = vore)) +
geom_histogram(alpha = 0.4, bins = 15) +
labs(x = "Average Amount of Sleep",
y = "Count") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
frequency polygon
ggplot(data = subset(msleepModified, subset = vore %in% use), aes(x = sleep_total, color = vore)) +
geom_freqpoly(alpha = 0.4, bins = 15, size = 2) +
labs(x = "Average Amount of Sleep",
y = "Count") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
Overlapping density
ggplot(data = subset(msleepModified, subset = vore %in% use), aes(x = sleep_total, fill = vore)) +
geom_density(alpha = 0.3) +
labs(x = "Average Amount of Sleep",
y = "Count") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Overlapping density – scaled
ggplot(data = subset(msleepModified, subset = vore %in% use), aes(x = sleep_total, fill = vore)) +
geom_density(alpha = 0.3, aes(y = ..scaled..)) +
labs(x = "Average Amount of Sleep",
y = "Count") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))## Warning: The dot-dot notation (`..scaled..`) was deprecated in ggplot2 3.4.0.
## ℹ Please use `after_stat(scaled)` instead.
Overlapping density – more groups – scaled
ggplot(msleepModified, aes(x = sleep_total, fill = vore)) +
geom_density(alpha = 0.3, aes(y = ..scaled..)) +
labs(x = "Average Amount of Sleep",
y = "Count") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))
Ridgeline Plot
library(ggridges)##
## Attaching package: 'ggridges'## The following objects are masked from 'package:ggdist':
##
## scale_point_color_continuous, scale_point_color_discrete,
## scale_point_colour_continuous, scale_point_colour_discrete,
## scale_point_fill_continuous, scale_point_fill_discrete,
## scale_point_size_continuousggplot(msleepModified, aes(x = sleep_total, y = vore, fill = vore)) +
geom_density_ridges() +
theme_ridges() +
labs(x = "Average Amount of Sleep", y = "Feeder\nType") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5)) +
theme(legend.position = "none") # supresses the legend## Picking joint bandwidth of 2.27
Violin Plots
ggplot(msleepModified, aes(x = vore, y = sleep_total, fill = vore)) +
geom_violin() +
labs(x = "Feeder Type", y = "Average\nAmount of\nSleep") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5)) +
theme(legend.position = "none") # supresses the legend
Histodot plots / Strip plots
Also called Wilkinson dot plots or strip plots. They show every point.
ggplot(msleepModified, aes(x = vore, y = sleep_total, fill = vore)) +
geom_dotplot(binaxis = "y", binwidth = .5, stackdir = "center") +
labs(x = "Feeder Type", y = "Average\nAmount of\nSleep") +
theme_minimal(base_size = 16) +
theme(axis.title.x = element_text(face="bold", colour="#000000", size=14),
axis.text.x = element_text(size=16),
axis.text.y = element_text(angle = 0, size = 14),
axis.title.y = element_text(angle=0, size = 16, face = "bold", vjust = .5))+
theme(legend.position = "none") # supresses the legend
Strip plot with large number of observations
Histodot plots or strip plots can be very useful for visualizing hte distribution of a large number of data points and particularly useful for comparing the distributions across a number of categories. The example here uses Gapminder data from 1952 to 2007 that was already cleaned and presented at http://bit.ly/2cLzoxH.
Note in the code below, the use of geom_jitter. The jitter allows dots that would otherwise be plotted in the exact same location to be jittered slightly so that they are visible.
strip <- ggplot(df, aes(x = continent, y = lifeExp)) +
geom_jitter(position = position_jitter(height = 0, width = .25),
alpha = .17,
size = 3,
color = "lightseagreen") +
theme_minimal(base_size = 14)+
ylim(0, 90)+
scale_color_few("Medium") +
xlab("\nGeographic region")+
ylab("Life\nexpectancy\n(years)") +
theme(axis.title.y = element_text(angle = 0, vjust = .5),
legend.position = "none")
strip