Remove n legend from ggplot

Question

I'm trying to remove the 'n' legend from the following plot. I'm guessing it's related to the stat part of geom_bar() but am not entirely sure what it is showing and hence am not sure how to remove it. I do want the fill legend so show.legends=FALSE isn't the right option. Sorry if this is a duplicate but after a lot of looking I can't find the answer, changing the legend on a scale_x_x doesn't cover it.

ggplot(iris,aes(x=Sepal.Length,y=Sepal.Width,fill=Species))+
   geom_bar(stat="sum")

Possible duplicate of [Turning off some legends in a ggplot](https://stackoverflow.com/questions/14604435/turning-off-some-legends-in-a-ggplot) — kath, Aug 22 '18 at 11:13
You can add `+ guides(size = F)` to you plot, so yes it is not exactly there only `guides(color = F)` — kath, Aug 22 '18 at 11:21
Thanks kath that's perfect. The issue I was having was identifying it was the size element that was causing the issue, which James has helped with below — Morgan Ball, Aug 22 '18 at 11:22

score 11 · Accepted Answer · answered Aug 22 '18 at 11:08

11

You can control legends with the show.legend parameter, with fine control by using a named vector:

show.legend

logical. Should this layer be included in the legends? NA, the default, includes if any aesthetics are mapped. FALSE never includes, and TRUE always includes. It can also be a named logical vector to finely select the aesthetics to display.

The trick is to recognise that the n part of the legend comes from the size aesthetic.

ggplot(iris,aes(x=Sepal.Length,y=Sepal.Width,fill=Species))+
+    geom_bar(stat="sum", show.legend=c(size=FALSE))

answered Aug 22 '18 at 11:08

James

65,548
14
155
193

4

"The trick is to recognise that the n part of the legend comes from the size aesthetic", how did you notice it? (+1) – markus Aug 22 '18 at 11:10
5

I used `str` on the plot object to see what was in it. – James Aug 22 '18 at 11:13
Alternatively, knowing the aes name you can use `+ guides(size = FALSE)` – CSJCampbell Mar 25 '22 at 10:46

score 3 · Answer 2 · answered Aug 22 '18 at 11:01

I would do the following

library(tidyverse)
iris %>%
    group_by(Species, Sepal.Length) %>%
    summarise(Sepal.Width = sum(Sepal.Width)) %>%
    ggplot(aes(Sepal.Length, Sepal.Width, fill = Species)) +
    geom_col()

Explanation: Summarise Sepal.Width per Species per Sepal.Length first, then plot. Generally (I think) it is advisable and tidier to keep data manipulations/summarisations and plotting separate.

Remove n legend from ggplot

2 Answers2

Linked