My barplot with errorbars in ggplot2 is wrong

Question

Minimal reproducible example is here:

library(ggplot2)

vals <- c(10, 12, 13, 20, 21, 28)
err <- c(0.85, 1.2, 0.9, 1.35, 2.2, 0.98)
cat <- c(rep("A",3),rep("B",3))

df <- data.frame(vals, err, cat)
df$cat<- as.factor(df$cat)

ggplot(df, aes(x=cat, y=vals, fill=vals)) +
  geom_bar(stat='identity', position='dodge') +
  geom_errorbar(aes(ymin=vals-err, ymax=vals+err), width=.1)

This code returns a plot like this:

however I expected two columns next to each other per cat. Closer to this:

What am I doing wrong here?

I believe the continuous fill is confusing "dodge". One option is to switch to using "dodge2". Alternatively add group = factor(vals) into your global aes() so dodge knows the categories to dodge on. I think the latter might be easiest if you are going to added the dodged bars — aosmith
– aosmith, Commented Apr 27, 2021 at 14:42

Kieran · Accepted Answer · 2021-04-27 14:42:22Z

1

If each row is a separate observation, you need to make that explicit. Make an index variable, turn it into a factor, and put that on the x-axis.

If obs is repeated across categories (e.g. if both A and B have obs numbered 1, 2 and 3), then facet on cat with facet_wrap(~ cat).


vals <- c(10, 12, 13, 20, 21, 28)
err <- c(0.85, 1.2, 0.9, 1.35, 2.2, 0.98)
cat <- c(rep("A",3),rep("B",3))

df <- data.frame(vals, err, cat)
df$cat<- as.factor(df$cat)
df$obs <- factor(1:nrow(df))

ggplot(df, aes(x=obs, y=vals, fill=cat)) +
  geom_col(position='dodge') +
  geom_errorbar(aes(ymin=vals-err, ymax=vals+err))

edited Apr 27, 2021 at 14:42

answered Apr 27, 2021 at 14:36

Kieran

1,22310 silver badges9 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

My barplot with errorbars in ggplot2 is wrong

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related