CODEWITHSUNDEEP
rggplot2
tjebo
tjebo

Reputation: 23767

Cutting axis ticks at the correct limits when setting coord_cartesian limits on factorised continuous x

I need to combine continuous and categorical data and need to factorise my continuous variable. I struggle to understand how to cut off the axis ticks on the right limits.

I understand that the x limits are based on the new factor levels, but why do the ticks extend beyond the lower limit?

Background: I'd like to combine a linear regression line on a continuous variable (with geom_smooth) with summarising stats on binned data of the same variable (e.g., geom_boxplot). I need to create a factor with all levels because otherwise those plots don't overlay. But this then creates a lot of blank space to both sides, therefore I tried to set the limits with coord_cartesian, which I am aware is always continuous.

library(ggplot2)

foo <- data.frame(x = 20:50, y = rnorm(31))

ggplot(foo) + 
  geom_col(aes(factor(x), y)) +
  coord_cartesian(xlim = c(10,50))

Created on 2020-03-02 by the reprex package (v0.3.0)

Upvotes: 2

Views: 362

Answers (1)

Gregor Thomas
Gregor Thomas

Reputation: 145965

Use a group aesthetic to define which bar is which, rather than using a discrete scale.

ggplot(foo, aes(x, y)) + 
  stat_smooth(method = "lm", se = FALSE) +
  geom_col(aes(group = x))

enter image description here

Upvotes: 5

Related Questions