Search code examples
rggplot2bar-chartgeom-barstacked-chart

How to make a stacked barchart in ggplot from a long dataframe format?


I have a long dataframe format which is created like this:

testAppVectorJG <- c(17, 155, 200, 200, 382, 499, 548, 548, 642, 642, 699, 699)
testVectorJG <- c(testAppVectorJG[1], diff(testAppVectorJG))

testAppVectorJW <- c(145, 209, 366, 548, 548, 613, 746, 928, 1064, 1266, 1371, 1573)
testVectorJW <- c(testAppVectorJW[1], diff(testAppVectorJW))

test_df <- data.frame(puntenvector = c(testVectorJG, testVectorJW),
                      team = c(rep("Jasper & Gijs", length(testAppVectorJG)),
                               rep("Jaap & Wil", length(testAppVectorJW))),
                      Rondenummer = as.factor(1:length(testVectorJG)))

I want to make a stacked bar chart with a bar per 'Rondenummer' (i.e. number of the round played). I want to see the percentage/distribution of points per round per team.

So far I have tried:

ggplot(data = test_df, aes(Rondenummer)) +
    geom_bar(aes(x = puntenvector, fill = team))

But then I get:

Warning message:
position_stack requires non-overlapping x intervals

And not at all the plot I want. How do I achieve this fairly simple plot?


Solution

  • Maybe something like this ?

    library(ggplot2)
    
    ggplot(data = test_df, aes(Rondenummer, puntenvector, fill = team)) +
       geom_bar(stat='identity')
    

    enter image description here

    If we want to include the values in the plot we can use label with geom_text

    ggplot(data = test_df, 
          aes(Rondenummer, puntenvector, fill = team, label = puntenvector)) +
          geom_bar(stat='identity') +
          geom_text(position = position_stack(vjust = 0.5))
    

    enter image description here

    And finally if we want to reverse the order of Rondenummer after coord_flip() we can add scale_x_discrete and reverse the levels.

     ggplot(data = test_df, 
       aes(Rondenummer, puntenvector, fill = team, label = puntenvector)) +
       geom_bar(stat='identity') +
       geom_text(position = position_stack(vjust = 0.5)) +
       coord_flip() + 
       scale_x_discrete(limits = rev(levels(test_df$Rondenummer))) 
    

    enter image description here