Graphic Design with ggplot2

How to Create Engaging and
Complex Visualizations in R

Cédric Scherer // rstudio::conf // July 2022

Welcome!

Conversation Starters

• What’s your name?
• Where do you feel home?
• When did you use R for the first time?
• What’s your favorite animal / plant / color / typeface?
• Whom would you like to meet during the rstudio::conf?
• Which R package are you looking forward to try?

Announcements

WiFi

• Username: conf22
• Password: together!
  
  

Course Materials

• Webpage: rstudio-conf-2022.github.io/ggplot2-graphic-design
• Rstudio Cloud: rstudio.cloud/spaces/235465

Announcements

• gender neutral bathrooms by the National Harbor rooms
• meditation room is located at National Harbor 9, open 8am-5pm, Monday-Thursday
• there is also has a dedicated room behind the reception
• lactation room is located at Potomac Dressing Room, open 8am-5pm, Monday-Thursday.
• participants with red lanyards wish to be not photographed, please note everyone’s lanyard colors before taking a photo
• The Code of Conduct and COVID policies can be found at rstudio.com/conference/2022/2022-conf-code-of-conduct,
  please review them carefully

Communication Strategy

• Green sticky note — I am done with the exercise
  
  
• Pink sticky note — I need help or support
  
  
• You can ask questions / comment at any time during the course.
• Please avoid questions during the breaks to give us a chance to recover and reflect.
• Team up with your neighbors for exercises — and lunch!
• We will collect feedback twice during the day (more later).

Introduction

{ggplot2} is a system for declaratively creating graphics,
based on “The Grammar of Graphics” (Wilkinson, 2005).

You provide the data, tell {ggplot2} how to map variables to aesthetics,
what graphical primitives to use, and it takes care of the details.

Advantages of {ggplot2}

• consistent underlying “grammar of graphics” (Wilkinson 2005)
  
• very flexible, layered plot specification
  
• theme system for polishing plot appearance
  
• lots of additional functionality thanks to extensions
  
• active and helpful community

Illustration by Allison Horst

ggplot2 Examples featured on ggplot2.tidyverse.org

Illustration by Allison Horst

The {ggplot2} Showcase

Collection of Graphics from the BBC R Cookbook

Contribution to #TidyTuesday Week 2020/31

My reinterpreted The Economist graphic

“European Elections” by Torsten Sprenger

Contribution to #TidyTuesday Week 2020/08

Contribution to the #SWDchallenge “Small Multiples”

Contribution to the #TidyTuesday Week 2021/15 x #30DayChartChallenge 2021 by Jake Kaupp

Moon Charts as a Tile Grid Map showing the 2nd Vote Results from the German Election 2021

#TidyTuesday Contribution by Jack Davison

Our Winning Contribution to the BES MoveMap Contest

Bivariate Choropleth x Hillshade Map by Timo Gossenbacher

Pixel Art by Georgios Karamanis

Generative Art by Thomas Lin Pedersen

A Motivational Example

• Graphic
• Code

## scatter plot of plot bikes$count versus bikes$count
ggplot(bikes, aes(temp_feel, count)) +
  geom_point(size = 2.2)

• Graphic
• Code

ggplot(bikes, aes(temp_feel, count)) +
  ## colored, semi-transparent points
  geom_point(
    aes(color = season),
    size = 2.2, alpha = .55
  )

• Graphic
• Code

ggplot(bikes, aes(temp_feel, count)) +
  geom_point(
    aes(color = season),
    size = 2.2, alpha = .55
  ) +
  ## add a linear fitting for each time of the day
  geom_smooth(
    aes(group = day_night),
    method = "lm", color = "black"
  )

• Graphic
• Code

ggplot(bikes, aes(temp_feel, count)) +
  geom_point(
    aes(color = season),
    size = 2.2, alpha = .55
  ) +
  geom_smooth(
    aes(group = day_night),
    method = "lm", color = "black"
  ) +
  ## create small multiples of time of day versus workday
  facet_grid(day_night ~ is_workday)

• Graphic
• Code

ggplot(bikes, aes(temp_feel, count)) +
  geom_point(
    aes(color = season),
    size = 2.2, alpha = .55
  ) +
  geom_smooth(
    aes(group = day_night),
    method = "lm", color = "black"
  ) +
  ## create free-ranging, proportionally sized small multiples
  facet_grid(
    day_night ~ is_workday,
    scales = "free_y", space = "free_y"
  )

• Graphic
• Code

ggplot(bikes, aes(temp_feel, count)) +
  geom_point(
    aes(color = season),
    size = 2.2, alpha = .55
  ) +
  geom_smooth(
    aes(group = day_night),
    method = "lm", color = "black"
  ) +
  facet_grid(
    day_night ~ is_workday,
    scales = "free_y", space = "free_y"
  ) +
  ## add labels + titles
  labs(
    x = "Feels-Like Temperature", y = NULL,
    caption = "Data: TfL (Transport for London), Jan 2015 — Dec 2016",
    title = "Reported bike rents versus feels-like temperature in London per time of day, period, and season."
  )

• Graphic
• Code

ggplot(bikes, aes(temp_feel, count)) +
  geom_point(
    aes(color = season),
    size = 2.2, alpha = .55
  ) +
  geom_smooth(
    aes(group = day_night),
    method = "lm", color = "black"
  ) +
  facet_grid(
    day_night ~ is_workday,
    scales = "free_y", space = "free_y"
  ) +
  ## add custom colors + legend styling
  scale_color_manual(
    values = c("#3c89d9", "#1ec99b", "#F7B01B", "#a26e7c"), name = "Season:",
    guide = guide_legend(override.aes = list(size = 5))
  ) +
  labs(
    x = "Feels-Like Temperature", y = NULL,
    caption = "Data: TfL (Transport for London), Jan 2015 — Dec 2016",
    title = "Reported bike rents versus feels-like temperature in London per time of day, period, and season."
  ) +
  ## use different theme and typeface
  theme_light(base_size = 18, base_family = "Cabinet Grotesk")

• Graphic
• Code

## coding for facet strip text
codes <- c(
  workday = "Workday",
  weekend_or_holiday = "Weekend or Holiday"
)

ggplot(bikes, aes(temp_feel, count)) +
  ## format seasons
  geom_point(
    aes(color = forcats::fct_relabel(season, stringr::str_to_title)),
    size = 2.2, alpha = .55
  ) +
  geom_smooth(
    aes(group = day_night),
    method = "lm", color = "black"
  ) +
  ## format facet strip text
  facet_grid(
    day_night ~ is_workday,
    scales = "free_y", space = "free_y",
    labeller = labeller(
      day_night = stringr::str_to_title,
      is_workday = codes
    )
  ) +
  ## customize x axis
  scale_x_continuous(
    expand = c(.02, .02),
    breaks = 0:6*5, labels = function(x) paste0(x, "°C")
  ) +
  ## customize y axis
  scale_y_continuous(
    expand = c(.1, .1), limits = c(0, NA),
    breaks = 0:5*10000, labels = scales::comma_format()
  ) +
  scale_color_manual(
    values = c("#3c89d9", "#1ec99b", "#F7B01B", "#a26e7c"), name = NULL,
    guide = guide_legend(override.aes = list(size = 5))
  ) +
  labs(
    x = "Feels-Like Temperature", y = NULL,
    caption = "Data: TfL (Transport for London), Jan 2015 — Dec 2016",
    title = "Reported bike rents versus feels-like temperature in London per time of day, period, and season."
  ) +
  theme_light(
    base_size = 18, base_family = "Cabinet Grotesk"
  ) +
  ## theme adjustments
  theme(
    plot.title.position = "plot",
    plot.caption.position = "plot",
    plot.title = element_text(face = "bold"),
    strip.text = element_text(face = "bold"),
    legend.position = "top"
  )

• Graphic
• Code

codes <- c(
  workday = "Workday",
  weekend_or_holiday = "Weekend or Holiday"
)

ggplot(bikes, aes(temp_feel, count)) +
  ## point outline
  geom_point(
    color = "black", fill = "white",
    shape = 21, size = 2.8
  ) +
  ## opaque point background
  geom_point(
    color = "white", size = 2.2
  ) +
  ## colored, semi-transparent points
  geom_point(
    aes(color = forcats::fct_relabel(season, stringr::str_to_title)),
    size = 2.2, alpha = .55
  ) +
  geom_smooth(
    aes(group = day_night), method = "lm", color = "black"
  ) +
  facet_grid(
    day_night ~ is_workday,
    scales = "free_y", space = "free_y",
    labeller = labeller(
      day_night = stringr::str_to_title,
      is_workday = codes
    )
  ) +
  scale_x_continuous(
    expand = c(.02, .02),
    breaks = 0:6*5, labels = function(x) paste0(x, "°C")
  ) +
  scale_y_continuous(
    expand = c(.1, .1), limits = c(0, NA),
    breaks = 0:5*10000, labels = scales::comma_format()
  ) +
  scale_color_manual(
    values = c("#3c89d9", "#1ec99b", "#F7B01B", "#a26e7c"), name = NULL,
    guide = guide_legend(override.aes = list(size = 5))
  ) +
  labs(
    x = "Feels-Like Temperature", y = NULL,
    caption = "Data: TfL (Transport for London), Jan 2015 — Dec 2016",
    title = "Reported bike rents versus feels-like temperature in London per time of day, period, and season."
  ) +
  theme_light(
    base_size = 18, base_family = "Cabinet Grotesk"
  ) +
  ## more theme adjustments
  theme(
    plot.title.position = "plot",
    plot.caption.position = "plot",
    plot.title = element_text(face = "bold", size = rel(1.3)),
    axis.text = element_text(family = "Tabular"),
    axis.title.x = element_text(hjust = 0, color = "grey30", margin = margin(t = 12)),
    strip.text = element_text(face = "bold" , size = rel(1.15)),
    panel.grid.major.x = element_blank(),
    panel.grid.minor = element_blank(),
    panel.spacing = unit(1.2, "lines"),
    legend.position = "top",
    legend.text = element_text(size = rel(1)),
    ## for fitting my slide background
    legend.key = element_rect(color = "#f8f8f8", fill = "#f8f8f8"),
    legend.background = element_rect(color = "#f8f8f8", fill = "#f8f8f8"),
    plot.background = element_rect(color = "#f8f8f8", fill = "#f8f8f8")
  )