The Grammar of Graphics: Designing and Transforming Data

“R is a language and environment for statistical computing and graphics”

Source: R-project.org

• ggplot2 is a data visualization package for R
  • developed by Hadley Wickham
  • inspired by Leland Wilkinson's “The Grammar of Graphics”
  • provides an organizing philosophy for building graphs – a structured approach to graphing
    

“The emphasis in ggplot2 is reducing the amount of thinking time by making it easier to go from the plot in your brain to the plot on the page.”

Wickham, 2012

• A ggplot2 graph is built up from a few basic elements:
  • Data: the raw data you want to plot
  • Aesthetics: including Mapping e.g., which variable is on the x-axis? the y-axis? Should the color/size/position of the plotted data that be mapped to some variable?
  • Geometries: the geometric shapes that represent the data
  • Statistics: statistical transformations that are used to summarize the data

Source: Hopper (2014)

install.packages('ggplot2')
library('ggplot2')

Make sure you have the most recent version of R to get the most recent version of ggplot2

data(Seatbelts)
s <- as.data.frame(Seatbelts)

ts <- data.frame(Year=floor(time(Seatbelts)),
Month=factor(cycle(Seatbelts),
labels=month.abb), Seatbelts)

• DriversKilled: car drivers killed
• drivers: monthly totals of car drivers in Great Britain killed or seriously injured Jan 1969 to Dec 1984
• front: front-seat passengers killed or seriously injured
• rear: rear-seat passengers killed or seriously injured
• kms: distance driven
• PetrolPrice: price of petrol
• VanKilled: number of van (‘light goods vehicle’) drivers
• law: 0/1: was the law in effect that month?
  
  Source: UK Driver Deaths via R datasets

head(ts)

qplot( data = ts,x= Year,y= DriversKilled, main= "Drivers Killed by Year")

• Scatter Plot
• Bar Graph
• Line Graph

ggplot(data = ts, 
       aes(x = Year, 
           y = DriversKilled)) + 
  geom_point() +
  ggtitle("Drivers killed by Year")

qplot( data = ts,
       x= Year,
       y= DriversKilled,
       main= "Drivers Killed by Year")

ggplot(data = ts, 
       aes(x = Year, 
           y = DriversKilled)) + 
  geom_point() +
  ggtitle("Drivers killed by Year")

ggplot(data = ts, 
       aes(x = Year, 
           y = VanKilled)) +
  geom_bar(stat = 'identity')

ggplot(data = ts, 
       aes(x = Year, 
           y = front)) +
  geom_point() +
  geom_line()

ggplot(data = ts, 
       aes(x = Year, 
           y = DriversKilled)) + 
  geom_point() +
  scale_y_continuous(limits = c(50,200))

ggplot(data = ts, 
       aes(x = Year, 
           y = DriversKilled, 
           color = Month)) + 
  geom_point() 

ggplot(data = ts, 
       aes(x = Year, 
           y = VanKilled)) +
  geom_bar(stat = 'identity') +
  theme(panel.background = element_blank())

• A ggplot2 graph is built up from a few basic elements:
  • Data: the raw data you want to plot
  • Aesthetics: including Mapping e.g., which variable is on the x-axis? the y-axis? Should the color/size/position of the plotted data that be mapped to some variable?
  • Geometries: the geometric shapes that represent the data
  • Statistics: statistical transformations that are used to summarize the data

Source: Hopper (2014)

• “A Layered Grammar of Graphics” by Hadley Wickham

• “A Simple Introduction to the Graphing Philosophy of ggplot2” by Tom Hopper

• ggplot2 Cheat Sheet by RStudio

• ggplot2 Documentation by Hadley Wickham & Winston Chang

• Resources for Doing Data Journalism by Timo Grossenbacher

• Shapes and Line Types in R by Winston Chang

• Text Labels
• Facetting
• Fitted Lines

ggplot(data = ts, 
       aes(x = Month, 
           y = DriversKilled)) +
        geom_text(aes(label = Year)) 

• Facetting allows you to split up your data by one or two variables
• facet_grid() places one or two variables in either vertical or horizontal directions
• facet_wrap() places facets next to each other, wrapping with a certain # of rows and/or columns

facet_grid(vertical ~ horizontal)

facet_wrap(~ variable, nrow = ___, ncol = ___)

ggplot(data = ts, 
       aes(x = Year, 
           y = DriversKilled)) + 
  geom_line() +
  facet_wrap(~ Month)

• How do we draw a fitted line through these points?

ggplot(data = ts, 
       aes(x = Year, 
           y = DriversKilled)) + 
  geom_point() +
  stat_smooth(method = 'lm')