Agenda

Visualizing data

• the grammar of graphics
• working with ggplot2
• color consciousness

graphics in R

Three common ways to generate graphics in R are via

• base R
• lattice (little development in recent years though)
• ggplot2

We will skip base R graphics since ggplot2 will be the graphics package for this class.

Let us see how we use it, starting with a simple bar-chart

Remember the basic options...

• one qualitative/categorical variables: bar-chart

• one quantitative/continuous variables: histogram/box-plot/area-chart

• two quantitative/continuous variables: scatter-plot/hex-bin

I will use two data-sets, the first being this IMDB data-set

The internet movie database, http://imdb.com/, is a website devoted to collecting movie data supplied by studios and fans. It claims to be the biggest movie database on the web and is run by amazon. More about information imdb.com can be found online, http://imdb.com/help/show_ leaf?about, including information about the data collection process, http://imdb.com/help/show_leaf?infosource.

library(ggplot2movies)

A data frame with 28819 rows and 24 variables

• title. Title of the movie.
• year. Year of release.
• budget. Total budget (if known) in US dollars
• length. Length in minutes.
• rating. Average IMDB user rating.
• votes. Number of IMDB users who rated this movie.
• r1-10. Multiplying by ten gives percentile (to nearest 10%) of users who rated this movie a 1.
• mpaa. MPAA rating (missing for a lot of movies)
• action, animation, comedy, drama, documentary, romance, short. Binary variables representing if movie was classified as belonging to that genre.

The second data-set is the Star Wars dataset, a tibble with 87 rows and 13 variables:

library(dplyr)
data(starwars)

• name: Name of the character
• height: Height (cm)
• mass: Weight (kg)
• hair_color,skin_color,eye_color: Hair, skin, and eye colors
• birth_year: Year born (BBY = Before Battle of Yavin)
• gender: male, female, hermaphrodite, or none.
• homeworld: Name of homeworld
• species: Name of species
• films: List of films the character appeared in
• vehicles: List of vehicles the character has piloted
• starships: List of starships the character has piloted

a tibble you say?

data frames vs. tibbles

data.frame(
  `Some Letters` = c("A", "B", "C"), 
  `Some Numbers` = c(1, 2, 3)
  ) -> adf
str(adf)

## 'data.frame':    3 obs. of  2 variables:
##  $ Some.Letters: Factor w/ 3 levels "A","B","C": 1 2 3
##  $ Some.Numbers: num  1 2 3

print(adf)

##   Some.Letters Some.Numbers
## 1            A            1
## 2            B            2
## 3            C            3

tibble(
  `Some Letters` = c("A", "B", "C"), 
  `Some Numbers` = c(1, 2, 3)
  ) -> atib
glimpse(atib)

## Observations: 3
## Variables: 2
## $ `Some Letters` <chr> "A", "B", "C"
## $ `Some Numbers` <dbl> 1, 2, 3

print(atib)

## # A tibble: 3 x 2
##   `Some Letters` `Some Numbers`
##   <chr>                   <dbl>
## 1 A                           1
## 2 B                           2
## 3 C                           3

and the grammar of graphics

library(ggplot2)

qplot will generate a quick plot but ggplot2 is the way to go so we build with it

ggplot(data = starwars)

ggplot(data = starwars,
        mapping = aes(x = eye_color)
       )

ggplot(data = starwars, 
       mapping = aes(x = eye_color, 
                     colour = eye_color)) +
  geom_bar() +
  labs(x = "Eye Color", 
       y = "Frequency", 
       title = "Bar-chart of Eye Color", 
       subtitle = "(of Star Wars characters)")

This is not a very good plot of course. Why is that?

c("black", "blue", "slategray",
  "brown", "gray34", "gold",
  "greenyellow", "navajowhite1",
  "orange", "pink", "red",
  "magenta", "thistle3", "white",
  "yellow") -> mycolors
ggplot(data = starwars, 
       mapping = aes(x = eye_color)) +
  geom_bar(fill = mycolors) +
  labs(x = "Eye Color", 
       y = "Frequency", 
       title = "Bar-chart of Eye Color", 
       subtitle = "(of Star Wars characters)",
       caption = "Much better!")

We can still do better. How can we improve this?

think in terms of the frequencies of eye-color

I'll switch to a different variable and show you how to use prebuilt color palettes

ggplot(data = starwars, 
       mapping = aes(x = gender)) +
  geom_bar(aes(fill = gender)) +
  labs(x = "Gender", y = "Frequency",
       title = "Bar-chart of Gender", 
       subtitle = "(of Star Wars characters)", 
       caption = "(Source: The dplyr package)") +
  scale_fill_brewer(palette = "Pastel1")

library(wesanderson)
ggplot(data = starwars, mapping = aes(x = gender)) +
  geom_bar(aes(fill = gender)) +
  labs(x = "Gender", y = "Frequency",
       title = "Bar-chart of Gender", 
       subtitle = "(of Star Wars characters)",
       caption = "(Source: The dplyr package)") +
  scale_fill_manual(values = 
                      wes_palette("Darjeeling1"))

• scale_fill_brewer in the plot on the left, calling on built-in color palettes. You can review them here
• scale_fill_manual in the plot on the right and a specific palette is being invoked from the wesanderson package

One can also lean on various plotting themes as shown below

library(ggthemes)
ggplot(data = starwars, aes(x = eye_color)) +
  geom_bar() +
  theme_tufte() +
  theme(axis.text.x = element_text(size = 6)) -> p1
ggplot(data = starwars, aes(x = eye_color)) +
  geom_bar() +
  theme_solarized() +
  theme(axis.text.x = element_text(size = 6)) -> p2
ggplot(data = starwars, aes(x = eye_color)) +
  geom_bar() +
  theme_economist() +
  theme(axis.text.x = element_text(size = 6)) -> p3
ggplot(data = starwars, aes(x = eye_color)) +
  geom_bar() +
  theme_fivethirtyeight() +
  theme(axis.text.x = element_text(size = 6)) -> p4
library(patchwork)
p1 + p2 + p3 + p4 + plot_layout(ncol = 2)

Note: I am not using mapping = aes(...) any longer

Later on you will learn these & other ways to build advanced visualizations ...for now we get to work more with ggplot2

library(ggplot2)
ggplot(data = movies, aes(x = mpaa)) +
  geom_bar() +
  theme_minimal()

library(ggplot2)
ggplot(data = movies) +
  geom_bar(aes(x = mpaa)) +
  theme_minimal()

Notice how the aes(x = mpaa) is placed differently in the two commands and yet yields identical plots

MPAA ratings are missing for a lot of movies so we eliminate these from the plot via subset(mpa != "")

str(movies$mpaa)

##  chr [1:58788] "" "" "" "" "" "" "R" "" "" "" "" "" "" "" "PG-13" "PG-13" "" "" "" "" ...

ggplot(subset(movies, mpaa != ""), aes(x = mpaa)) + geom_bar() + theme_minimal()

The order of the bars is fortuitous in that it goes from the smallest frequency to the highest frequency.

I said fortuitous because ggplot2's default is to order the bars in an ascending alphabetic/alphanumeric ssequence if the variable is a character. See below for an example.

df = tibble(x = c(rep("A", 2), rep("B", 4), rep("C", 1)))
ggplot(data = df, aes(x = x)) + geom_bar() + theme_minimal()

Later on we'll learn how to order the bars with ascending/descending frequencies or by some other logic

What about plotting the relative frequencies on the y-axis rather than the frequencies?

ggplot(data = subset(movies, mpaa != ""), 
       aes(x = mpaa, 
           y = (..count..)/sum(..count..))) +
  geom_bar() +
  scale_y_continuous(labels = scales::percent) +
  labs(x = "MPAA Rating", 
       y = "Relative Frequency (%)") +
  theme_minimal()

We could also add a second or even a third/fourth categorical variable. Let us see this with our hsb2 data-set

library(here)
load("data/hsb2.RData")
ggplot(data = hsb2, aes(x = ses, group = female)) +
  geom_bar(aes(fill = female)) +
  theme_minimal()

This is not very useful since the viewer has to estimate the relative sizes of the two colors within any given bar.

ggplot(data = hsb2, aes(x = ses, group = female)) +
  geom_bar(aes(fill = female), position = "dodge") +
  theme_minimal()

position = "dodge" will juxtapose the bars, and this is much better.

This is fine if you want to know what percent of the 200 students are low SES males, low SES females, etc. What if you wanted to calculate percentages within each sex?

ggplot(data = hsb2, aes(x = ses, y = female)) +
  geom_bar(aes(group = female, fill = female, y = ..prop..), position = "dodge") +
  scale_y_continuous(labels = scales::percent) +
  labs(y = "Relative Frequency (%)", x = "Socioeconomic Status Groups") +
  theme_minimal()

What about within each ses?

ggplot(data = hsb2, aes(x = female, y = ses)) + 
  geom_bar(aes(group = ses, fill = ses, y = ..prop..), position = "dodge") +
  scale_y_continuous(labels = scales::percent) +
  labs(y = "Relative Frequency (%)", x = "Socioeconomic Status Groups") +
  theme_minimal()

geom_histogram() does the trick but note that the default number of bins is not very useful and can be tweaked, along with other embellishments

ggplot(data = hsb2, aes(x = read)) +
  geom_histogram(fill = "cornflowerblue", color = "white") +
  ggtitle("Histogram of Reading Scores") +
  labs(x = "Reading Score", y = "Frequency") +
  theme_minimal()

See bins = 5 and also experiment with binwidth =

ggplot(data = hsb2, aes(x = read)) +
  geom_histogram(fill = "cornflowerblue", color = "white", bins = 5) +
  ggtitle("Histogram of Reading Scores") +
  labs(x = "Reading Score", y = "Frequency") +
  theme_minimal()

If we wanted to break out the histogram by one or more categorical variables, we could do so

ggplot(hsb2, aes(x = read)) +
  geom_histogram(fill = "cornflowerblue", bins = 5, color = "white") +
  ggtitle("Histogram of Reading Scores") +
  labs(x = "Reading Score", y = "Frequency") +
  facet_wrap(~ female) +
  theme_minimal()

Or better yet,

ggplot(hsb2, aes(x = read)) +
  geom_histogram(fill="cornflowerblue", 
                 bins = 10, 
                 color = "white") +
  ggtitle("Histogram of Reading Scores") +
  labs(x = "Reading Score", y = "Frequency") +
  facet_wrap(~ female, ncol = 1) +
  theme_minimal()

One useful design element with breakouts is placing in relief the consolidated data (i.e., the distribution without break-outs)

ggplot(data = hsb2, aes(x = read, fill = female)) +
  geom_histogram(bins = 10, color = "white") +
  ggtitle("Histogram of Reading Scores") +
  labs(x = "Reading Score", y = "Frequency") +
  facet_wrap(~ female, ncol = 1) +
  geom_histogram(data = hsb2[, -2],
                 bins = 10, 
                 fill = "grey",
                 alpha = .5) +
  theme_minimal()

Here it is obvious that the distribution of readings scores of any one sex are similar to the overall distribution so perhaps the groups are not really that different in terms of reading scores

For breakouts with two categorical variables we could do

ggplot(data = hsb2, aes(x = read)) +
  geom_histogram(fill="cornflowerblue",
                 bins = 10, 
                 color = "white") +
  ggtitle("Histogramx of Reading Scores") +
  labs(x = "Reading Score", 
       y = "Frequency") +
  facet_wrap(~ female + schtyp, 
             ncol = 2) +
  theme_minimal()

... which is the same as ...

ggplot(data = hsb2, aes(x = read)) +
  geom_histogram(fill = "cornflowerblue", bins = 10, color = "white") +
  ggtitle("Histogramx of Reading Scores") +
  labs(x = "Reading Score", y = "Frequency") +
  facet_wrap(~ schtyp + female, ncol = 2) +
  theme_minimal()

These were all the rage in the summer of 2017, and initially named joy plots but the unfortunate connection with the source of the plots led the name to be revised to ridge-plots. If you are curious, see why not joy?. You need to have installed the ggridges package but other than that, they are easy to craft.

library(viridis)
library(ggridges)
library(ggthemes)
ggplot(lincoln_weather, 
       aes(x = `Mean Temperature [F]`, y = `Month`)) +
  geom_density_ridges(scale = 3, alpha = 0.3, 
                      aes(fill = Month)) +
  labs(x = "Mean Temperature",
       y = "Month", 
       title = 'Temperatures in Lincoln NE', 
       subtitle = 'Mean temperatures (Fahrenheit) 
       by month for 2016',
       caption = "Data: The Weather Underground") +
  theme_ridges() +
  theme(axis.title.y = element_blank(), 
        legend.position = "none")

Here is another one, mapping the distribution of hemoglobin in four populations (the US being the reference group) as part of a study looking at the impacy of altitude on hemoglobin concentration (courtesy Whitlock and Schluter).

hemo <- read.csv(url("http://whitlockschluter.zoology.ubc.ca/wp-content/data/chapter02/chap02e3cHumanHemoglobinElevation.csv"))
ggplot(hemo, aes(x = hemoglobin, y = population)) +
  geom_density_ridges(scale = 3, alpha = 0.3, 
                      aes(fill = population)) +
  labs(title = 'Hemoglobin Concentration Levels',
       subtitle = 'in Four populations') +
  theme_ridges() +
  theme(axis.title.y = element_blank(),
        legend.position = "none")

As should be evident, they are visually appealing when comparing a large number of groups on a single continuous variable and using simple facet-wrap or other options would be suboptimal

These can be useful to look at the distribution of a continuous variable

ggplot(hemo, aes(y = hemoglobin, x = "")) +
  geom_boxplot(fill = "cornflowerblue") +
  coord_flip() + 
  labs(x = "", y = "Hemoglobin Concentration") +
  theme_minimal()

Ideal use: time series data

library(plotly)
data(economics)
# names(economics)
ggplot(economics, aes(x = date, y = uempmed)) +
  geom_line() + labs(x = "Date", 
                     y = "Unemployment Rate") +
  theme_minimal()

load("data/gap.df.RData")
ggplot(gap.df, aes(x = year, y = LifeExp,
                   group = continent, 
                   color = continent)) +
  geom_line() +
  geom_point() +
  labs(x = "Year", 
       y = "Median Life Expectancy (in years)",
       color = "") +
  theme_minimal() +
  theme(legend.position = "bottom")

Pretty ugly!!

but we can dress these up by using plotly

library(plotly)
myplot <- plot_ly(economics, x = ~date) %>%
  add_trace(y = ~uempmed, line = list(color = 'black'),
            mode = "lines", 
            name = "Unemployment Rate") %>%
  add_trace(y = ~psavert, line = list(color = 'red'),
            mode = "lines", 
            name = "Personal Savings Rate") %>%
  layout(autosize = F, width = 600, height = 300)
library(shiny)
div(myplot, align = "center")

scatter-plots ...

load("data/hsb2.RData")
ggplot(hsb2, aes(x = write, y = science)) +
  geom_point() +
  labs(x = "Writing Scores", y = "Science Scores") +
  theme_minimal()

color by ses

ggplot(hsb2, aes(x = write, y = science)) +
  geom_point(aes(color = ses)) +
  labs(x = "Writing Scores", y = "Science Scores") +
  theme_minimal() +
  theme(legend.position = "bottom")

breakout ses for ease of interpretation

ggplot(hsb2, aes(x = write, y = science)) +
  geom_point() +
  labs(x = "Writing Scores", y = "Science Scores") +
  facet_wrap(~ ses) +
  theme_minimal()

make it interactive with plotly

p = plot_ly(hsb2, x = ~write, y = ~science, color = ~ses) %>%
  layout(autosize = FALSE, width = 600, height = 300) 
div(p, align = "center")

count plots: see the frequency of given pairs of values by varying sizes of the points. The more the frequency of a pair the greater the size of these points.

data(mpg, package = "ggplot2")
ggplot(mpg, aes(x = cty, y = hwy)) +
  geom_count(col = "firebrick", show.legend = FALSE) +
  labs(subtitle = "City vs Highway mileage",
       y = "Highway mileage", 
       x = "City mileage")  +
  theme_minimal()

Boston Marathon data

boston = read.csv("./data/BostonMarathon.csv")
boston2 = boston[sample(nrow(boston), 200), ]
ggplot(boston2, aes(x = Age, y = finishtime, group = M.F)) +
  geom_count(aes(color = M.F), show.legend = FALSE) +
  labs(subtitle = "", y = "Finishing Times (in seconds)",
       x = "Age (in years)") +
  facet_wrap(~ M.F, ncol = 1) +
  theme_minimal()

• scatter-plots not helpful when data points overlap
• hex-bins carve up the plotting grid into hexagons of equal size, count how many $x,y$ pairs fall in each hexagon, and use a color scheme (like a heatmap) to show where hexagons have more data versus less.

ggplot(data = diamonds, aes(y = price, x = carat)) +
  geom_hex() +
  labs(x = "Weight in Carats", y = "Price") +
  theme_minimal()

• adding a third variable

ggplot(data = diamonds, aes(y = price, x = carat)) +
  geom_hex() +
  labs(x = "Weight in Carats", y = "Price") +
  facet_wrap(~ color) +
  theme_minimal()

Some helpful resources

The Data Visualization Catalogue developed by Severino Ribecca to create a library of different information visualisation types

The R Graph Catalog maintained by Joanna Zhao and Jenny Bryan is always useful to see code-and-resulting-figure

Ferdio's Data Visualization Project, a website trying to present all relevant data visualizations, so you can find the right visualization and get inspired how to make them

The Chartmaker Directory will offer an answer to one of the most common questions in data visualisation: which tool do you need to make that chart?

Emery's Essentials focusing on the charts that give you the best bang for your buck

The Data Visualization Checklist by Ann K. Emery & Stephanie Evergreen

Data Visualization: A Practical Introduction by Kieran Healy

Some relevant RStudio webinars

• The grammar of graphics
• The boxplot
• The histogram

Check out Data Camp and the blogosphere for plenty of examples