Coloring boxplot outlier points in ggplot2?

Question

How can I color the outlier points in ggplot2? I want them to be the same color as the boxplot itself. colour= is not enough to do this.

Example:

<

Answer 1

Just in case you really want to keep the boxes black:

Another possible solution is to

1. store in a variable whether a point is an outlier,
2. suppress outlier plotting with outlier.color = NA, and
3. plot only those points that are outliers and use the col aesthetic in this call to geom_point()

Step 1: Define a function to determine whether a point is an outlier:

is.outlier <- function (x) {
  x < quantile(x, .25) - 1.5 * IQR(x) |
    x > quantile(x, .75) + 1.5 * IQR(x)
}

Step 2: Group data and calculate outliers with this function for the groups

diamonds %>% group_by(cut) %>%
  mutate(outlier.p = is.outlier(price)) %>%
  ungroup() -> diamonds

Step 3: Create plot

ggplot(diamonds, aes(x = cut, y = price, fill = cut)) +
  geom_boxplot(outlier.color = NA) +
  geom_point(data = diamonds[diamonds$outlier.p,], aes(col = cut))

Please note that you have to subset your dataset in the geom_point() call (last line) lest all points are plotted.