The Cleveland dot plot

A dot plot is useful when you want to compare values in a narrow range. Dots show position on a scale, so the axis can exclude zero without changing what the length of a mark represents.

Usage

geom_cleveland_dot(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  orientation = c("y", "x"),
  leader = c("axis", "full", "none"),
  leader_colour = "grey80",
  leader_linetype = "dotted",
  leader_linewidth = 0.3,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

Arguments

mapping, data, stat, position, na.rm, show.legend, inherit.aes, ...

Standard ggplot2 layer arguments. See layer().

orientation

Which axis holds the categories. "y", the default, puts categories down the left and values across. This leaves room for category names. "x" transposes the layout.

leader

One of "axis" (the default), which draws the leader from the axis to the dot, "full", which runs it the whole width of the panel, or "none".

leader_colour, leader_linetype, leader_linewidth

Appearance of the leader line. Keep it light enough to distinguish from the dots.

Value

A ggplot2 layer.

Details

Cleveland's leader lines help readers follow each row to its value. You can extend them across the panel or turn them off.

If rank is the comparison you want, sort the categories by value before plotting. Use stats::reorder() or forcats::fct_reorder().

Examples

library(ggplot2)
d <- data.frame(
  country = c("Japan", "Korea", "Australia", "New Zealand", "Singapore"),
  value = c(41.2, 32.9, 38.4, 29.7, 22.5)
)

ggplot(d, aes(value, stats::reorder(country, value))) +
  geom_cleveland_dot() +
  labs(x = "Share (%)", y = NULL) +
  theme_tufte()