How to read a boxplot (and make one in R)

What the box, whiskers, line and dots in a boxplot actually mean — median, quartiles, IQR and outliers — with a labelled ggplot2 example in R.

Boxplots are everywhere, yet often half-understood. In one compact picture they show a distribution’s centre, spread and outliers. Here’s exactly what each part means — and how to build one in R. (Need the numbers behind it? Use our descriptive statistics calculator.)

The anatomy

• The line inside the box — the median (50th percentile).
• The box — from Q1 (25th) to Q3 (75th) percentile; its height is the interquartile range (IQR), the spread of the middle half of the data.
• The whiskers — extend to the most extreme points within 1.5 × IQR of the box.
• The dots — points beyond the whiskers, flagged as potential outliers.

library(ggplot2)
set.seed(6)
d <- data.frame(y = c(rnorm(60, 50, 8), 78, 82))  # two outliers

ggplot(d, aes(x = "", y = y)) +
  geom_boxplot(width = 0.35, fill = "#2f6fed", alpha = 0.15,
               colour = "#1b2a4a", outlier.colour = "#b02a37", outlier.size = 2) +
  annotate("text", x = 1.28, y = median(d$y), label = "median", hjust = 0, colour = "#1b2a4a") +
  annotate("text", x = 1.28, y = quantile(d$y, .75), label = "Q3", hjust = 0, colour = "#5a6478") +
  annotate("text", x = 1.28, y = quantile(d$y, .25), label = "Q1", hjust = 0, colour = "#5a6478") +
  labs(title = "Anatomy of a boxplot", x = NULL, y = "Value") +
  coord_cartesian(xlim = c(0.6, 1.7)) +
  theme_minimal(base_family = "sans") +
  theme(plot.title = element_text(face = "bold", colour = "#1b2a4a"))

Figure 1

Reading it at a glance

• Skew: if the median sits off-centre in the box, or one whisker is much longer, the distribution is skewed.
• Spread: a taller box means more variable data.
• Comparisons: side-by-side boxplots make group differences (and different spreads) obvious — which is why they pair so well with a t-test or ANOVA.

One caution

A boxplot hides the shape within each region — a bimodal distribution can look identical to a uniform one. For small datasets, overlay the points (geom_jitter()) or use a violin plot to see the full picture.

---

Clear figures are half of good analysis. When you need publication-ready ones, that’s our work.