I'm a new stackoverflow user and can't comment currently on the original post to ask a question. I came across a previous stackoverflow answer (https://stackoverflow.com/a/34045068/11799491) and I was wondering how you would add two vertical lines (mean of the group and median of the group) to this graph here.
My attempt: I don't know how to add in the group variable "type"
geom_vline(aes(xintercept = mean(diff), ), color="black") +
geom_vline(aes(xintercept = median(diff), ), color="red")
There are a few different ways to do this, but I like creating a separate summarized data frame and then passing that into the geom_vline call. This lets you analyze the results and makes it easy to add multiple lines that are automatically sorted and colored by type:
library(tidyverse)
df <-
tibble(
x = rnorm(40),
category = rep(c(0, 1), each = 20)
)
df_stats <-
df %>%
group_by(category) %>%
summarize(
mean = mean(x),
median = median(x)
) %>%
gather(key = key, value = value, mean:median)
df %>%
ggplot(aes(x = x)) +
geom_histogram(bins = 20) +
facet_wrap(~ category) +
geom_vline(data = df_stats, aes(xintercept = value, color = key))
The easiest way is to pre-compute the means and the medians by groups of type. I will do it with aggregate.
agg <- aggregate(diff ~ type, data, function(x) {
c(mean = mean(x), median = median(x))
})
agg <- cbind(agg[1], agg[[2]])
agg <- reshape2::melt(agg, id.vars = "type")
library(ggplot2)
ggplot(data, aes(x = diff)) +
geom_histogram() +
geom_vline(data = agg, mapping = aes(xintercept = value,
color = variable)) +
facet_grid(~type) +
theme_bw()
