R : ggplot2 plot several data frames in one plot

Question

I'm little bit stuck on ggplot2 trying to plot several data frame in one plot.

I have several data frame here I'll present just two exemples.

The data frame have the same Header but are different. Let say that I want to count balls that I have in 2 boxes.

name=c('red','blue','green','purple','white','black')
value1=c(2,3,4,2,6,8)
value2=c(1,5,7,3,4,2)

test1=data.frame("Color"=name,"Count"=value1)
test2=data.frame("Color"=name,"Count"=value2)

What I'm trying to do it's to make a bar plot of my count. At the moment what I did it's :

(plot_test=ggplot(NULL, aes(x= Color, y=Count)) + 
    geom_bar(data=test1,stat = "identity",color='green')+
    geom_bar(data=test2,stat = "identity",color='blue')
)

I want to have x=Color and y=Count, and barplot of test2 data frame next to test1. Here there are overlapping themselves. So I'll have same name twice in x but I want to plot the data frames in several color and got in legend the name.

For example "Green bar" = test1 "Blue bar" = test2

Thank you for your time and your help.

Best regards

Answer 1

Try this:

name=c('red','blue','green','purple','white','black')
value1=c(2,3,4,2,6,8)
value2=c(1,5,7,3,4,2)

test1=data.frame("Color"=name,"Count"=value1)
test2=data.frame("Color"=name,"Count"=value2)

test1$var <- 'test1'
test2$var <- 'test2'

test_all <- rbind(test1,test2)


(plot_test=ggplot(data=test_all) + 
  geom_bar(aes(x=Color,y=Count,color=var),
           stat = "identity", position=position_dodge(1))+
  scale_color_manual(values = c('green', 'blue'))
)

Answer 2

You have two options here:

Either tweak the size and position of the bars

ggplot(NULL, aes(x= Color, y=Count)) + 
geom_bar(data=test1, aes(color='test1'), stat = "identity",
         width=.4, position=position_nudge(x = -0.2)) +
geom_bar(data=test2, aes(color='test2'), stat = "identity", 
         width=.4, position=position_nudge(x = 0.2))

or what I recommend is join the two data frames together and then plot

library(dplyr)
test1 %>% 
  full_join(test2, by = 'Color') %>% 
  data.table::melt(id.vars = 'Color') %>% 
  ggplot(aes(x= Color, y=value, fill = variable)) + 
  geom_bar(stat = "identity", position = 'dodge')

Answer 3

This will do what you were trying to do:

balls <- data.frame(
  count = c(c(2,3,4,2,6,8),c(1,5,7,3,4,2)),
  colour = c(c('red','blue','green','purple','white','black'),c('red','blue','green','purple','white','black')),
  box = c(rep("1", times = 6), rep("2", times = 6))
)

ggplot(balls, aes(x = colour, y = count, fill = box)) +
  geom_col() +
  scale_fill_manual(values = c("green","blue"))

This is better because it facilitates comparisons between the box counts:

ggplot(balls, aes(x = colour, y = count)) +
  geom_col() +
  facet_wrap(~ box, ncol = 1, labeller = as_labeller(c("1" = "Box #1", "2" = "Box #2")))