Lecture 7

Refine Your Plots

Byeong-Hak Choe
bchoe@geneseo.edu

SUNY Geneseo

February 19, 2025

Refine your plots

Refine your plots

Use color to your advantage

  • We should choose a color palette based on the types of the data we are plotting.
    • An unordered categorical variable (e.g., ‘Sex’ or ‘Country’) requires distinct colors that won’t be easily confused with one another.
    • An ordered categorical variable (e.g., ‘Level of Education’) requires a graded color scheme of some kind running from less to more or earlier to later.
  • We choose color palettes for mappings through one of the scale_ functions for color or fill.

Refine your plots

RColorBrewer

# install.pacakges("RColorBrewer")
library(RColorBrewer)
  • RColorBrewer package provides a wide range of named color palettes.
    • We can access these colors by specifying the scale_color_brewer() or scale_fill_brewer() functions with palette parameter.

Refine your plots

Sequential

Refine your plots

Diverging

Refine your plots

Qualitative

Refine your plots

Use color to your advantage

p <- ggplot(data = organdata,
            mapping = 
              aes(x = roads, 
                  y = donors, 
                  color = world))

p + geom_point(size = 2) + 
  scale_color_brewer(
    palette = "Set2") +
  theme(legend.position = "top")

  • Let’s use the named palettes with scale_color_brewer(palette = ...).

Refine your plots

Pastel2

p + geom_point(size = 2) + 
  scale_color_brewer(
    palette = "Pastel2") +
  theme(legend.position = "top")

Refine your plots

Dark2

p + geom_point(size = 2) + 
  scale_color_brewer(
    palette = "Dark2") +
  theme(legend.position = "top")

Refine your plots

Use color to your advantage

  • We can also specify colors manually, via scale_color_manual() or scale_fill_manual().
    • These functions take values argument that can be specified as vector of color names or hex colors.
    • Check https://www.color-hex.com

Refine your plots

Use color to your advantage

p + geom_point(size = 2) + 
  scale_color_manual(
    values = c("#3c6ff8", "#afd68d", 
               "#8467ad", "#82857f")) +
  theme_ipsum() + 
  theme(legend.position = "top")

Refine your plots

Color blindness

brewer.pal.info
  • RColorBrewer provides the color-blind friendly pallets.

Refine your plots

Layer color and text together

  • We can use color to highlight some aspect of our data.
    • Let’s do a simple exploratory analysis on the relationship between politics and race using color.

Refine your plots

Layer color and text together

# DEM Blue and REP Red
party_colors <- 
  c("#2E74C0", "#CB454A") 

p0 <- ggplot(
  data = county_data |> 
    filter(flipped == "No"),
  mapping = 
    aes(x = pop, 
        y = black/100) )

p1 <- p0 + 
  geom_point(alpha = 0.15, 
             color = "gray50") +
  scale_x_log10(labels=scales::comma) 
p1

Refine your plots

Layer color and text together

p2 <- p1 + 
  geom_point(
    data = county_data |> 
      filter(flipped == "Yes"),
    mapping = 
      aes(x = pop, y = black/100,
          color = partywinner16)) +
    scale_color_manual(
      values = party_colors)
p2

Refine your plots

Layer color and text together

p3 <- p2 + 
  scale_y_continuous(
    labels=scales::percent) +
  labs(
    color = 
      "County flipped to ... ",
    x = 
      "County Population (log scale)",
    y = 
      "Percent Black Population",
    title = 
      "Flipped counties, 2016",
    caption = 
      "Counties in gray did not flip.")
p3

Refine your plots

Layer color and text together

p4 <- p3 + 
  geom_text_repel(
    data = 
      filter(
       county_data,
       flipped == "Yes" & black >25),
    mapping = 
      aes(x = pop, y = black/100,
          label = state), 
    size = 2)

p4 + theme_minimal() + 
  theme(legend.position="top")

Refine your plots

Layer color and text together

p4 + theme_economist() +
  theme(legend.position = "top")

Refine your plots

Layer color and text together

p4 + theme_wsj() +
  theme(
    plot.title = 
      element_text(size = rel(0.6)),
    legend.title = 
      element_text(size = rel(0.35)),
    plot.caption = 
      element_text(size = rel(0.35)),
    legend.position = "top")

Refine your plots

Change the appearance of plots with theme()

p4 + theme(
  legend.position = "top",
  plot.title = element_text(
      size = rel(2),
      lineheight = .5,
      family = "Times",
      face = "bold.italic",
      color = "orange"),
  axis.text.x = element_text(
      size = rel(1.1),
      family = "Courier",
      face = "bold",
      color = "purple")
  )

  • The theme() function allows us to exert very fine-grained control over the appearance of all kinds of text and graphical elements in a plot.

Refine your plots

Change the appearance of plots with theme()

p4 + theme(
  legend.position = "top",
  plot.title = element_text(
    size = rel(2),
    lineheight = .5,
    family = "Times",
    face = "bold.italic",
    colour = "orange"),
  axis.text.x = element_blank()
  )

  • We can also use element_blank() to remove a number of elements by naming them and making them disappear.

Refine your plots

Change the appearance of plots with theme()

  • Let’s create an effective small multiple of the age distribution of General Social Survey (GSS) respondents over the years using gss_lon data.
yrs <- c(seq(1972, 1988, 4), 
         1993, 
         seq(1996, 2016, 4))

mean_age <- gss_lon |>
    filter( !is.na(age), 
            year %in% yrs) |>
    group_by(year) |>
    summarize(
      xbar = round(
        mean(age, na.rm = TRUE), 0)
      )
mean_age <- mean_age |> 
  mutate(y = 0.3)

yr_labs <- data.frame(
  x = 85, y = 0.8, 
  year = yrs)  # to position the age as a text label

Refine your plots

Change the appearance of plots with theme()

p <- ggplot(
  data = gss_lon |> 
    filter(year %in% yrs),
  mapping = 
    aes(x = age))

p1 <- p + 
  geom_density(
    fill = "black", color = FALSE,
    alpha = 0.9, 
    mapping = aes(y = ..scaled..))
p1

Refine your plots

Change the appearance of plots with theme()

p2 <- p1 + 
  geom_vline(
    data = mean_age |> 
      filter(year %in% yrs),
    aes(xintercept = xbar), 
    color = "white", size = 0.5) + 
  geom_text(
    data = mean_age |> 
      filter(year %in% yrs),
    aes(x = xbar, y = y, label = xbar), 
    nudge_x = 7.5, color = "white", 
    size = 3.5, hjust = 1) +
  geom_text(data = yr_labs |> 
              filter(year %in% yrs),
    aes(x = x, y = y, label = year)) 
p2

  • The nudge_x argument to push the label slightly to the right of its x-value.

Refine your plots

Change the appearance of plots with theme()

p3 <- p2  + 
  facet_grid(year ~ ., switch = "y")
p3

  • In the facet_grid() we use the switch argument to move the labels to the left.

Refine your plots

Change the appearance of plots with theme()

p2a <- p3 + 
  theme(
    plot.title = 
      element_text(size = 16),
    axis.text.x= 
      element_text(size = 12),
    axis.title.y=element_blank(),
    axis.text.y=element_blank(),
    axis.ticks.y = element_blank(),
    strip.background = element_blank(),
    strip.text.y = element_blank(),
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank()) +
  labs(x = "Age", y = NULL,
       title = 
         "Age Distribution of\nGSS Respondents")
p2a

Refine your plots

ggridges

# install.packages("ggridges")
library(ggridges)
  • ggridges allows the distributions to overlap vertically.
    • It is especially useful for repeated distributional measures that change in a clear direction.

Refine your plots

ggridges

p <- ggplot(
  data = gss_lon,
  mapping = 
    aes(x = age, 
        y = factor(year, 
                   levels = rev(unique(year)), 
                   ordered = TRUE)))
  • factor() converts a variable to a factor variable.
    • With levels parameter, we can set the categories of a categorical variable.

Refine your plots

ggridges

p2b <- p + 
  geom_density_ridges(
    alpha = 0.6, fill = "lightblue", 
    scale = 1.5) +  
    scale_x_continuous(
      breaks = c(25, 50, 75)) +
    scale_y_discrete(
      expand = c(0.01, 0)) + 
    labs(x = "Age", y = NULL, 
         title = 
           "Age Distribution of\nGSS Respondents") +
    theme_ridges() +  # make labels aligned properly
    theme(
      title = 
        element_text(
          size = 16, face = "bold"))
p2b

  • The expand argument in scale_y_discrete() adjusts the scaling of the y-axis slightly.

Refine your plots

Arrange plots with gridExtra::grid.arrange()

# install.packages("gridExtra")
grid.arrange(p2a, p2b, nrow = 1)   # sub-figures
  • gridExtra package provides grid.arrange() function.
    • grid.arrange() arranges multiple ggplot objects on a page, and draw tables of figures.

Refine your plots

Arrange plots with gridExtra::grid.arrange()

Refine your plots

Advanced Bar Chart 1

  • Use data.frame studebt to describe how the distribution of Debt pct varies by type.
studebt

p_xlab <- "Amount Owed, in thousands of Dollars"
p_title <- "Outstanding Student Loans"
p_subtitle <- "44 million borrowers owe a total of $1.3 trillion"
p_caption <- "Source: FRB NY"

f_labs <- c(`Borrowers` = "Percent of\nall Borrowers",
            `Balances` = "Percent of\nall Balances")

Refine your plots

Advanced Bar Chart 1

p <- ggplot(
  data = studebt,
  mapping = 
    aes(x = pct/100, y = Debt,
        fill = type))
p1 <- p + geom_col()
p1

Refine your plots

Advanced Bar Chart 1

p2 <- p1 +
  scale_fill_brewer(
    type = "qual", palette = "Dark2")
p2

Refine your plots

Advanced Bar Chart 1

p3 <- p2 +
  scale_x_continuous(
    labels = scales::percent)
p3

Refine your plots

Advanced Bar Chart 1

p4 <- p3 +
  guides(fill = FALSE)
p4

Refine your plots

Advanced Bar Chart 1

p5 <- p4 +
  facet_grid(
    .~ type, 
    labeller = as_labeller(f_labs))
p5

Refine your plots

Advanced Bar Chart 1

p6 <- p5 +
  labs(y = NULL, x = p_xlab, 
       caption = p_caption,
       title = p_title,
       subtitle = p_subtitle)
p6

Refine your plots

Advanced Bar Chart 1

p7 <- p6 +
  theme(strip.text.x = 
          element_text(face = "bold"))
p7

Refine your plots

Advanced Bar Chart 2

  • Instead of having separate bars distinguished by heights, we can array the percentages for each distribution proportionally within a single bar.

  • Let’s make a stacked bar chart with just two main bars, and lie them on their side for comparison.

Refine your plots

Advanced Bar Chart 2

p <- ggplot(
  studebt, 
  aes(x = pct/100, y = type, 
      fill = Debtrc))
p1 <- p + 
  geom_col(color = "gray80")
p1

Refine your plots

Advanced Bar Chart 2

p2 <- p1 +
  scale_y_discrete(
    labels = as_labeller(f_labs))
p2

Refine your plots

Advanced Bar Chart 2

p3 <- p2 +
  scale_x_continuous(
    labels = scales::percent)
p3

Refine your plots

Advanced Bar Chart 2

p4 <- p3 +
  scale_fill_viridis_d(
    option = "B")
p4

Refine your plots

Advanced Bar Chart 2

p5 <- p4 +
  guides(
    fill = 
      guide_legend(
        reverse = TRUE,
        title.position = "top",
        label.position = "bottom",
        keywidth = 3,
        nrow = 1))
p5

Refine your plots

Advanced Bar Chart 2

p6 <- p5 +
  labs(x = NULL, y = NULL,
       fill = "Amount Owed, in thousands of dollars",
       caption = p_caption,
       title = p_title,
       subtitle = p_subtitle)
p6

Refine your plots

Advanced Bar Chart 2

p7 <- p6 +
  theme(legend.position = "top",
        axis.text.y = 
          element_text(
            face = "bold",
            hjust = 1, 
            size = 12),
        axis.ticks.length = 
          unit(0, "cm"),
        panel.grid.major.y = 
          element_blank())
p7