Visualization with ggplot2

visualization
ggplot2

Using ggplot2

Author

Josef Fruehwald

Published

September 9, 2024

Data Packages 
install.packages(
  c(
    "phonTools",
    "ngramr"
  )
)
library(ggplot2)
library(ngramr)

sneak_past <- ngram(
  phrases = c(
    "sneaked",
    "snuck"
  )
)
rmarkdown::paged_table(
  head(sneak_past)
)

Anatomy of a ggplot

Data Layer

ggplot(
  data = sneak_past
)

Map data to “aesthetics”

ggplot(
  data = sneak_past,
  aes(
    x = Year,
    y = Frequency
  )
)
1
Think of aes() as a special quoting function

Geometries

ggplot(
  data = sneak_past,
  aes(
    x = Year,
    y = Frequency
  )
) +
  geom_point()
1
Points

Adjusting Geometries

ggplot(
  data = sneak_past,
  aes(
    x = Year,
    y = Frequency
  )
) +
  geom_point(
    size = 0.1
  )      

ggplot(
  data = sneak_past,
  aes(
    x = Year,
    y = Frequency
  )
) +
  geom_point(
    shape = "square"
  )      

ggplot(
  data = sneak_past,
  aes(
    x = Year,
    y = Frequency
  )
) +
  geom_point(
    color = "steelblue"
  )      

ggplot(
  data = subset(sneak_past, Phrase == "sneaked"),
  aes(
    x = Year,
    y = Frequency
  )
) +
  geom_line(
    linewidth = 2
  )

Exercise

Plot the ngram data as large red triangles

ggplot(
  data = sneak_past,
  aes(
    x = Year,
    y = Frequency
  )
)+
  geom_point(
    size = 5, 
    shape = "triangle",
    color = "red"
  )

ggplot(
  data = sneak_past,
  aes(
    x = Year,
    y = Frequency
  )
) +
  geom_line(
    aes(
      color = Phrase
    )
  )
1
Mapping in geom 

ggplot(
  data = sneak_past,
  aes(
    x = Year,
    y = Frequency,
    color = Phrase
  )
) +
  geom_line()
1
Mapping in data layer

Exercise

We’ve been told by a journal editor that we can’t have color figures in our paper in the year 2024 CE. Instead, we need to map the Phrase data to linetype.

ggplot(
  data = sneak_past,
  aes(
    x = Year,
    y = Frequency
  )
) +
  geom_line(
    aes(linetype = Phrase)
  )

Statistical layers

library(phonTools)

data("pb52")

head(pb52) |> 
  rmarkdown::paged_table()
ggplot(
  pb52,
  aes(f0, f1)
)+
  geom_point()

ggplot(
  pb52,
  aes(f0, f1)
)+
  geom_point()+
  stat_smooth(
    method = lm
  )

Exercise

Let’s partially recreate the Peterson & Barney plot. One addition to the code below will draw a data ellipse for each separate vowel categpory.

ggplot(
  pb52,
  aes(
    f1, 
    f2, 
    color = vowel
  )
)+
  geom_point()+
  stat_ellipse()

ggplot(
  pb52,
  aes(
    f1,
    f2
  )
)+
  stat_bin_2d()

Scales

ggplot(
  pb52,
  aes(
    x = f2,
    y = f1,
    color = vowel
  )
)+
  geom_text(
    aes(label = vowel)
  )

Issues

  • F1 and F2 dimensions are upside down and backwards.

  • The default color scale is problematic

Grey Scale Printing and Color Vision Deficiency simulations

Grey Scale Printing and Color Vision Deficiency simulations
library(ggthemes)

ggplot(
  pb52,
  aes(
    x = f2,
    y = f1,
    color = vowel
  )
)+
  geom_text(
    aes(label = vowel)
  )+
  scale_color_ptol()

Exercise

Look over the list of x/y scales to find a way to flip the x and y axes.

ggplot(
  pb52,
  aes(
    x = f2,
    y = f1,
    color = vowel
  )
)+
  geom_text(
    aes(
      label = vowel
    )
  ) +
  scale_color_ptol()+
  scale_x_reverse()+
  scale_y_reverse()

Labels and guides

ggplot(
  pb52,
  aes(
    x = f2,
    y = f1,
    color = vowel
  )
)+
  geom_text(
    aes(
      label = vowel
    )
  ) +
  scale_color_ptol()+
  scale_x_reverse()+
  scale_y_reverse()+
  labs(
    x = "F2 (front-back)",
    y = "F1 (high-low)"
  )

ggplot(
  pb52,
  aes(
    x = f2,
    y = f1,
    color = vowel
  )
)+
  geom_text(
    aes(
      label = vowel
    )
  ) +
  scale_color_ptol()+
  scale_x_reverse()+
  scale_y_reverse()+
  labs(
    x = "F2 (front-back)",
    y = "F1 (high-low)"
  ) +
  guides(
    color = "none"
  )

Exercise

Give the plot a title and a subtitle

ggplot(
  pb52,
  aes(
    x = f2,
    y = f1,
    color = vowel
  )
)+
  geom_text(
    aes(
      label = vowel
    )
  ) +
  scale_color_ptol()+
  scale_x_reverse()+
  scale_y_reverse()+
  labs(
    x = "F2 (front-back)",
    y = "F1 (high-low)",
    title = "Vowels",
    subtitle = "From Peterson & Barney (1952)"
  )+
  guides(
    color = "none"
  )

Faceting

ggplot(
  pb52,
  aes(
    x = f2,
    y = f1,
    color = vowel
  )
)+
  geom_text(
    aes(
      label = vowel
    )
  ) +
  scale_color_ptol()+
  scale_x_reverse()+
  scale_y_reverse()+
  labs(
    x = "F2 (front-back)",
    y = "F1 (high-low)"
  )+
  guides(
    color = "none"
  ) +
  facet_wrap(~type)

Theming

ggplot(
  pb52,
  aes(
    x = f2,
    y = f1,
    color = vowel
  )
)+
  geom_text(
    aes(
      label = vowel
    )
  ) +
  scale_color_ptol()+
  scale_x_reverse()+
  scale_y_reverse()+
  labs(
    x = "F2 (front-back)",
    y = "F1 (high-low)"
  )+
  guides(
    color = "none"
  ) +
  facet_wrap(~type)+
  theme_minimal()

p + 
  theme_bw()

p + 
  theme_void()

p +
  ggdist::theme_ggdist()

p+
  ggthemes::theme_economist()+
  ggthemes::scale_color_economist()

p + 
  ggthemes::theme_excel()+
  ggthemes::scale_color_excel()

All together

ggplot(
  pb52,
  aes(
    x = f2,
    y = f1,
    color = vowel
  )
)+
  geom_text(
    aes(
      label = vowel
    )
  ) +
  stat_ellipse()+
  scale_color_ptol()+
  scale_x_reverse()+
  scale_y_reverse()+
  labs(
    x = "F2 (front-back)",
    y = "F1 (high-low)",
    title = "Vowels",
    subtitle = "Peterson & Barney (1952)",
    caption = "Source: {phonTools}"
  )+
  guides(
    color = "none"
  )+
  facet_wrap(
    ~type,
    labeller = "label_both"
  )+
  theme_minimal(
    base_size = 16
  )+
  theme(
    panel.grid.minor = element_blank(),
    text = element_text(family = "Georgia")
  )
1
The basic data layer
2
A geometry layer
3
A statistic layer
4
Scale adjustments
5
Label adjustments
6
Guide adjustment
7
Faceting
8
A built-in theme, adjusting the base font size
9
Some custom themeing (no minor breaks grid, changing the font family)

Getting fancier

You can set the plot font to any google font with the showtext package.

library(showtext)

font_add_google("Noto Sans", "Noto Sans")
font_add_google("Lobster", "Lobster")
showtext_auto()

p + 
  theme(
    text = element_text(family = "Noto Sans")
  )

p + 
  theme(
    text = element_text(family = "Lobster")
  )

Reuse

CC-BY 4.0

Citation

BibTeX citation:
@online{fruehwald2024,
  author = {Fruehwald, Josef},
  title = {Visualization with Ggplot2},
  date = {2024-09-09},
  url = {https://lin611-2024.github.io/notes/meetings/2024-09-09_ggplot2.html},
  langid = {en}
}
For attribution, please cite this work as:
Fruehwald, Josef. 2024. “Visualization with Ggplot2.” September 9, 2024. https://lin611-2024.github.io/notes/meetings/2024-09-09_ggplot2.html.