Rstudio reorder stacked ggplot geom_bar

Question

So I've made a stacked Barplot with help of @Axeman . So far so good.. Now I would love to swap the Blue plot with the Red plot. I've literally tried everything using the order function in ggplot and I cant get it right. Here is the Plot:

Stacked barplot

Below, you can find an extract of the dataframe that was used for this plot (first row containts the header). Basically its a division name, followed by two types of savings (Repetitive and MDF) and the the last column contains the sum of both (NewRepetitive + Anrechenbare_MDF)

    "DivisionName", "NewRepetitive", "Anrechenbare_MDF", "Total_CR"
    "Electrification Products" 0 0 0
    "Electrification Products" 39.762 0 39.762
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 1641.348852 0 1641.348852
    "Electrification Products" 6367.725252 0 6367.725252
    "Electrification Products" 24299.24741 0 24299.24741
    "Electrification Products" 32910.87486 0 32910.87486
    "Electrification Products" 8637.313704 0 8637.313704
    "Electrification Products" 9220.489704 0 9220.489704
    "Electrification Products" -1038.609948 0 -1038.609948
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 6346.518852 0 6346.518852
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 4675.189452 0 4675.189452
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 0 0 0
    "Electrification Products" 8495.893524 0 8495.893524
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 2439.849336 0 2439.849336
    "Power Grids" 0 0 0
    "Power Grids" -3909.837222 0 -3909.837222
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" -974.473842 0 -974.473842
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" -448.740678 0 -448.740678
    "Power Grids" -886.997442 0 -886.997442
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 263.767854 0 263.767854
    "Power Grids" -28.4961 0 -28.4961
    "Power Grids" 122.00307 0 122.00307
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 3910.778256 0 3910.778256
    "Power Grids" 0 0 0
    "Power Grids" -1246.87005 0 -1246.87005
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" -61.047924 0 -61.047924
    "Power Grids" -61.047924 0 -61.047924
    "Power Grids" 0 0 0
    "Power Grids" -6.348666 0 -6.348666
    "Power Grids" -12.697332 0 -12.697332
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" -240.798672 0 -240.798672
    "Power Grids" 0 0 0
    "Power Grids" -65.395236 0 -65.395236
    "Power Grids" 24.678948 0 24.678948
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 0 0 0
    "Power Grids" 215.669088 0 215.669088
    "Power Grids" -61.047924 0 -61.047924
    "Power Grids" 14.605908 0 14.605908
    "Power Grids" -42.61161 0 -42.61161
    "Power Grids" -553.540056 0 -553.540056
    "Power Grids" 0 0 0
    "Power Grids" 113.507256 0 113.507256
    "Power Grids" -21.365448 0 -21.365448
    "Power Grids" 0 0 0
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 603.25581 0 603.25581
    "Power Grids" -61.047924 0 -61.047924
    "Power Grids" 0 0 0
    "Power Grids" -224.429982 0 -224.429982
    "Power Grids" 0 0 0
    "Power Grids" 44.904552 0 44.904552
    "Power Grids" -240.798672 0 -240.798672
    "Power Grids" 0 0 0
    "Power Grids" -361.198008 0 -361.198008
    "Power Grids" 0 0 0
    "Power Grids" -240.798672 0 -240.798672
    "Power Grids" -115.349562 0 -115.349562
    "Power Grids" 0 0 0
    "Power Grids" 263.767854 0 263.767854
    "Power Grids" -1948.93443 0 -1948.93443
    "Power Grids" 16.36869 0 16.36869
    "Power Grids" 111.280584 0 111.280584
    "Power Grids" 924.254436 0 924.254436
    "Power Grids" -1122.136656 0 -1122.136656
    "Power Grids" 0 0 0
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 5.977554 0 5.977554
    "Power Grids" 0 0 0
    "Electrification Products" 5542.133592 -559.362777 4982.770815
    "Electrification Products" 3038.638548 -260.690669 2777.947879
    "Electrification Products" 2507.378466 -279.681388 2227.697078
    "Electrification Products" -2241.940608 -334.661491 -2576.602099
    "Electrification Products" -4535.744118 -931.275287 -5467.019405
    "Electrification Products" 808.003602 -156.175362 651.82824
    "Electrification Products" 958.873884 -209.163422 749.710462
    "Electrification Products" -830.588418 -144.621569 -975.209987
    "Electrification Products" -999.404616 -184.794228 -1184.198844
    "Electrification Products" 342.112248 -225.099694 117.012554
    "Electrification Products" -237.193584 -92.629517 -329.823101
    "Electrification Products" 238.572 -236.653473 1.918527
    "Electrification Products" 145.674714 -71.925657 73.749057
    "Electrification Products" 0 -40.637464 -40.637464
    "Electrification Products" -437.382 -111.553826 -548.935826
    "Electrification Products" 1852.074198 -643.426551 1208.647647
    "Electrification Products" 1318.680222 -290.106352 1028.57387
    "Electrification Products" 395.89698 -81.009327 314.887653
    "Electrification Products" 314.106546 -118.725138 195.381408
    "Electrification Products" -711.580752 -277.88855 -989.469302
    "Electrification Products" -897.22953 -196.215213 -1093.444743
    "Electrification Products" 228.074832 -150.066454 78.008378
    "Electrification Products" -657.517686 -188.844696 -846.362382
    "Electrification Products" -986.269902 -283.267038 -1269.53694
    "Electrification Products" -474.638994 -99.203216 -573.84221
    "Electrification Products" -711.951864 -148.804831 -860.756695
    "Electrification Products" 286.577988 -66.863233 219.714755

Here is the R-Code that was used to edit the dataframe the right way (using tidyr package), so I could make the stacked barplot, since ggplot needs the data in a specific form.

  library(ggplot2)
  library(tidyr) 

  pd <- gather(stack, 'key', 'value', -DivisionName)
  pd <- pd[pd$key != 'Total_CR', ]

And finally the ggplot code:

ggplot(pd, aes(DivisionName,  value, fill = key)) +
  geom_bar(stat = "summary", fun.y = "sum") +
  scale_fill_discrete(name="Saving-Typen",labels=c("MDF-Savings","Repetitive Savings")) +
  ylab("Total Savings") +  scale_y_continuous(breaks=c(seq(0,15000000,1000000)), expand = c(0,0))

The barplot looks exactly how I want it to, except the order.. I need to have the MDF Savings above of Repetitive-Savings. I'm sitting here for 5 hours and can't get the right solution.. I've tried everything with the "order" option.

Can someone help me?

Answer 1

Try to manually select the colors that you want:

+ scale_fill_manual(values=c("#999999", "#E69F00"), 
                       name="Legend title",
                       breaks=c("Electrical", "Power grids"),
                       labels=c("Electrical", "Power grids"))

as found on http://www.cookbook-r.com/Graphs/Legends_(ggplot2)/

Not tested.

Answer 2

The problem is not in your plot code, but in the data frame pd . You have pd$key as a character vector instead of a factor. Convert this to a factor with the correct ordering, and ggplot will do the rest for you.

After

pd <- gather(stack, 'key', 'value', -DivisionName)
pd <- pd[pd$key != 'Total_CR', ]

You do:

pd$key <- factor(pd$key,
                 levels = c("NewRepetitive",
                            "Anrechenbare_MDF"))

Don't forget to change your labels again to match the correct values.

ggplot(pd, aes(DivisionName,  abs(value), fill = key)) +
  geom_bar(stat = "summary", fun.y = "sum") +
  scale_fill_discrete(name="Saving-Typen",labels=c("Repetitive Savings","MDF_Savings")) +
  ylab("Total Savings") +  scale_y_continuous(breaks=c(seq(0,15000000,1000000)), expand = c(0,0))

gives

Note that I had to use abs(value) due to your dodgy example dataset. Please take a look at How to make a great R reproducible example? to get an idea of how to ask a question in the future.