How to define a degree of bar overlap by group in ggplot2?-CodePudding

In the following plot, I have two variables plotted side-by-side in a bar plot:

# required libraries
library(ggplot2)
library(reshape2)

# Make data
data(cars)
cars_scaled <- as.data.frame(scale(cars[1:10,]))
cars_scaled$id <- seq(nrow(cars_scaled))
cars_scaled <- melt(cars_scaled, id.vars = "id")
cars_scaled

# plot
ggplot(cars_scaled)   aes(x = id, y = value, group = variable, fill = variable)  
  geom_bar(stat="identity", position = "dodge", width = 0.8, col = 1, alpha = 0.5)

Is it possible to have the two bars of a group partially overlap? i.e. the pink extends to the right partially over the blue, and vice-versa.

CodePudding user response：

What you could do is add position_dodge to your geom_bar like this:

# required libraries
library(ggplot2)
library(reshape2)

# Make data
data(cars)
cars_scaled <- as.data.frame(scale(cars[1:10,]))
cars_scaled$id <- seq(nrow(cars_scaled))
cars_scaled <- melt(cars_scaled, id.vars = "id")
cars_scaled
#>    id variable       value
#> 1   1    speed -1.60356745
#> 2   2    speed -1.60356745
#> 3   3    speed -0.40089186
#> 4   4    speed -0.40089186
#> 5   5    speed  0.00000000
#> 6   6    speed  0.40089186
#> 7   7    speed  0.80178373
#> 8   8    speed  0.80178373
#> 9   9    speed  0.80178373
#> 10 10    speed  1.20267559
#> 11  1     dist -1.40818923
#> 12  2     dist -0.59772061
#> 13  3     dist -1.20557208
#> 14  4     dist  0.61798233
#> 15  5     dist  0.01013086
#> 16  6     dist -0.59772061
#> 17  7     dist  0.21274801
#> 18  8     dist  1.02321664
#> 19  9     dist  1.83368526
#> 20 10     dist  0.11143944

# plot
ggplot(cars_scaled, aes(x=id, y=value, fill=variable, group = variable))   
  geom_bar(position=position_dodge(.7), stat="identity", width = 0.8, col = 1, alpha = 0.5)

^{Created on 2022-07-19 by the reprex package (v2.0.1)}