Bob’s Burgers

Exploring dialogue trends in Bob’s Burgers with the power of {ggplot2} and {gghalves}, showcasing half-violin and haf-beeswarm scatter plots for storytelling.

About the Data

This dataset explores the dialogue metrics of episodes from the animated TV series Bob’s Burgers. Compiled by Steven Ponce, the data is made available through the {bobsburgersR} R package, which includes episode transcripts and additional metadata. A blog post by Steven Ponce provides insights into visualizing this dataset. Curated for this week’s Tidy Tuesday challenge by Jon Harmon, the dataset offers an opportunity to analyze dialogue trends and patterns across seasons. Key variables include dialogue density, average dialogue length, sentiment variance, unique word counts, and punctuation ratios (questions and exclamations). You can load the dataset using the tidytuesdayR package or directly from GitHub.

The visualization explores trends in dialogue punctuation in Bob’s Burgers episodes across seasons, focusing on the proportion of dialogue lines containing question marks (question_ratio) and exclamation points (exclamation_ratio). The graphic features a split representation of data: a scatter plot highlighting individual episode ratios and a violin plot illustrating the overall distribution of these proportions by season. The analysis reveals a noticeable decline in both ratios over the latter seasons, suggesting a decrease in the use of questions and exclamatory expressions in dialogue. This shift could indicate changes in the show’s tone, writing style, or character interactions as the series progresses.

Figure 1: This graphic explores how punctuation in Bob’s Burgers dialogue has evolved across seasons, focusing on the proportion of lines containing question marks and exclamation points. Each dot represents an episode, while the violin plots show the distribution of these ratios within each season. The analysis highlights a clear decline in both questions and exclamations in later seasons, suggesting a subtle shift in the show’s tone or dialogue style.

How I made this graphic?

To create this graphic, I used the {ggplot2} package, a foundational tool for data visualization in R, in combination with the {gghalves} package authored by Frederik Tiedemann and hosted on GitHub. The {gghalves} package simplifies the creation of “half-half” plots by extending {ggplot2} with geoms such as gghalves::geom_half_violin() and gghalves::geom_half_point(). In this graphic, I used geom_half_violin() to represent the distribution of punctuation ratios across seasons and geom_half_point() to plot individual episode data points, combining them to show detailed insights. To highlight trends across seasons, I added smoothed lines using ggplot2::geom_smooth(). {gghalves} is a versatile package that makes it easy to compose compact visualizations by combining plot types, and more details can be found on its GitHub repository.

Steps

Loading required libraries, data import & creating custom functions.

# Data Import and Wrangling Tools
library(tidyverse)            # All things tidy

# Final plot tools
library(scales)               # Nice Scales for ggplot2
library(fontawesome)          # Icons display in ggplot2
library(ggtext)               # Markdown text support for ggplot2
library(showtext)             # Display fonts in ggplot2
library(colorspace)           # Lighten and Darken colours
library(patchwork)            # Compiling Plots


# devtools::install_github('erocoar/gghalves')

# Geocomputation
library(gghalves)             # Half violin plot

# Load the data
episode_metrics <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2024/2024-11-19/episode_metrics.csv')

Visualization Parameters

# Font for titles
font_add_google("Rampart One",
  family = "title_font"
) 

# Font for the caption
font_add_google("Saira Extra Condensed",
  family = "caption_font"
) 

# Font for plot text
font_add_google("Atma",
  family = "body_font"
) 

showtext_auto()

bobs_burgers_palette <- c(
 "#f0262a", 
 "#f172a9", 
 "#95d244",
 "#fcdd60",
 "#9ac7e8"
)
 
# A base Colour
bg_col <- "#FEDD00FF"
seecolor::print_color(bg_col)

# Colour for highlighted text
text_hil <- "#BA0C2FFF"
seecolor::print_color(text_hil)

# Colour for the text
text_col <- "#000000FF"
seecolor::print_color(text_col)


# Define Base Text Size
bts <- 90 

# Caption stuff for the plot
sysfonts::font_add(
  family = "Font Awesome 6 Brands",
  regular = here::here("docs", "Font Awesome 6 Brands-Regular-400.otf")
)
github <- "&#xf09b"
github_username <- "aditya-dahiya"
xtwitter <- "&#xe61b"
xtwitter_username <- "@adityadahiyaias"
social_caption_1 <- glue::glue("<span style='font-family:\"Font Awesome 6 Brands\";'>{github};</span> <span style='color: {text_col}'>{github_username}  </span>")
social_caption_2 <- glue::glue("<span style='font-family:\"Font Awesome 6 Brands\";'>{xtwitter};</span> <span style='color: {text_col}'>{xtwitter_username}</span>")

# Add text to plot--------------------------------------------------------------
plot_title <- "Fading Punctuation:\nThe Changing Voice of Bob's Burgers"

plot_subtitle <- "Each dot represents an Episode of Bob's Burgers. Over the seasons, there is a decline in the use of both questions and exclamations in its dialogue, suggesting a subtle shift in the show's tone: moving toward less dramatic and inquisitive interactions."

plot_caption <- paste0(
  "**Data:** {bobsburgersR} by Steven Ponce", 
  " |  **Code:** ", 
  social_caption_1, 
  " |  **Graphics:** ", 
  social_caption_2
  )

rm(github, github_username, xtwitter, 
   xtwitter_username, social_caption_1, 
   social_caption_2)

Exploratory Data Analysis and Wrangling

# summarytools::dfSummary(episode_metrics) |> 
#   summarytools::view()

df <- episode_metrics |> 
  select(season, episode, question_ratio, exclamation_ratio) |> 
  pivot_longer(
    cols = c(question_ratio, exclamation_ratio),
    names_to = "facet_var",
    values_to = "y_var"
  )

The Base Plot

facet_labeller <- c(
  "% Dialogue ending in Questions (?)",
  "% Dialogue ending in Exclamations (!)"
)
names(facet_labeller) <- unique(df$facet_var)

g <- df |> 
  ggplot(
    mapping = aes(
      y = y_var,
      x = season,
      group = season
    )
  ) +
  
  # Plotting the data
  gghalves::geom_half_violin(
    colour = text_col,
    fill = alpha(text_col, 0.5),
    trim = FALSE,
    scale = "width",
    width = 0.7
  ) +
  gghalves::geom_half_point(
    size = 2,
    alpha = 0.9,
    colour = text_col
  ) +
  geom_smooth(
    mapping = aes(
      group = 1
    ),
    se = FALSE,
    colour = alpha(text_hil, 0.4),
    linewidth = 4
  ) +
  
  # Scales
  scale_x_continuous(
    breaks = 1:14,
    labels = 1:14,
    name = "Season",
    expand = expansion(c(0, 0.05))
  ) +
  scale_y_continuous(
    labels = scales::label_percent(),
    name = NULL
  ) +
  # Facets
  facet_wrap(
    ~facet_var, 
    ncol = 1,
    scales = "free_y",
    labeller = labeller(
      facet_var = facet_labeller
    )
  ) +
  labs(
    title = plot_title,
    subtitle = plot_subtitle |> str_wrap(90) |> str_view(),
    caption = plot_caption
  ) +
  theme_minimal(
    base_family = "body_font",
    base_size = bts
  ) +
  theme(
    # Overall
    plot.margin = margin(5,5,5,5, "mm"),
    plot.title.position = "plot",
    text = element_text(
      colour = text_col,
      lineheight = 0.3,
      margin = margin(0,0,0,0, "mm"),
      hjust = 0.5
    ),
    panel.background = element_rect(
      fill = "transparent",
      colour = "transparent"
    ),
    plot.background = element_rect(
      fill = "transparent",
      colour = "transparent"
    ),
    
    # Grid
    panel.grid = element_line(
      colour = alpha(text_hil, 0.7),
      linewidth = 0.3,
      linetype = 3
    ),
    panel.grid.minor = element_blank(),
    
    # Axes
    axis.ticks = element_blank(),
    axis.ticks.length = unit(0, "mm"),
    axis.text = element_text(
      colour = text_col,
      margin = margin(0,0,0,0, "mm"),
      face = "bold"
    ),
    axis.title.x = element_text(
      colour = text_col,
      size = 1.5 * bts, 
      face = "bold"
    ),
    
    
    # Labels
    plot.title = element_text(
      size = 2 * bts,
      family = "title_font",
      colour = text_hil,
      lineheight = 0.3,
      margin = margin(10,0,5,0, "mm"),
      hjust = 0.5
    ),
    plot.caption = element_textbox(
      family = "caption_font",
      margin = margin(5,0,0,0, "mm"),
      colour = text_col,
      size = 0.6 * bts,
      hjust = 0.5
    ),
    plot.subtitle = element_text(
      colour = text_col,
      hjust = 0.5,
      lineheight = 0.3,
      margin = margin(0,0,0,0, "mm")
    ),
    strip.text = element_text(
      hjust = 0.5,
      colour = text_hil,
      family = "title_font",
      size = 1.5 * bts,
      face = "bold",
      margin = margin(10,0,2,0, "mm")
    )
  )
  


ggsave(
  filename = here::here(
    "data_vizs",
    "tidy_bobs_burgers.png"
  ),
  plot = g,
  width = 400,
  height = 500,
  units = "mm",
  bg = bg_col
)

Savings the thumbnail for the webpage

# Saving a thumbnail

library(magick)
# Saving a thumbnail for the webpage
image_read(here::here("data_vizs", 
                      "tidy_bobs_burgers.png")) |> 
  image_resize(geometry = "400") |> 
  image_write(
    here::here(
      "data_vizs", 
      "thumbnails", 
      "tidy_bobs_burgers.png"
    )
  )

Session Info

# Data Import and Wrangling Tools
library(tidyverse)            # All things tidy

# Final plot tools
library(scales)               # Nice Scales for ggplot2
library(fontawesome)          # Icons display in ggplot2
library(ggtext)               # Markdown text support for ggplot2
library(showtext)             # Display fonts in ggplot2
library(colorspace)           # Lighten and Darken colours
library(patchwork)            # Compiling Plots


# devtools::install_github('erocoar/gghalves')

# Geocomputation
library(gghalves)             # Half violin plot

sessioninfo::session_info()$packages |> 
  as_tibble() |> 
  select(package, 
         version = loadedversion, 
         date, source) |> 
  arrange(package) |> 
  janitor::clean_names(
    case = "title"
  ) |> 
  gt::gt() |> 
  gt::opt_interactive(
    use_search = TRUE
  ) |> 
  gtExtras::gt_theme_espn()

Table 1: R Packages and their versions used in the creation of this page and graphics