I have a very annoying problem with a stacked bar plot created using ggplot2
.
There are a couple of similar questions previously asked but after going through the example code I cannot figure out what I am doing wrong.
I would like to make the graph so that the bars are stacked in the following order based on their Biogeographic.affinity
: (Top to Bottom= Bassian,Widespread,Torresian, and Eyrean). The colors to the bars should be: (Bassian=drakgrey, Widespread=lightgrey, Torresian=white, and Eyrean=black).
This is what the dataset looks like:
biogeo
Site Biogeographic.affinity Rank Number.of.species Total.Species Percent
1 A Bassian 1 1 121 0.8264463
2 A Eyrean 4 39 121 32.2314050
3 A Torresian 3 62 121 51.2396694
4 A Widespread 2 19 121 15.7024793
5 DD Bassian 1 1 128 0.7812500
6 DD Eyrean 4 46 128 35.9375000
7 DD Torresian 3 63 128 49.2187500
8 DD Widespread 2 18 128 14.0625000
9 E_W Bassian 1 1 136 0.7352941
10 E_W Eyrean 4 54 136 39.7058824
11 E_W Torresian 3 65 136 47.7941176
12 E_W Widespread 2 16 136 11.7647059
13 KS Bassian 1 2 145 1.3793103
14 KS Eyrean 4 63 145 43.4482759
15 KS Torresian 3 62 145 42.7586207
16 KS Widespread 2 18 145 12.4137931
17 Z_Ka Bassian 1 1 110 0.9090909
18 Z_Ka Eyrean 4 64 110 58.1818182
19 Z_Ka Torresian 3 31 110 28.1818182
20 Z_Ka Widespread 2 14 110 12.7272727
This is the code I have written so far (including some of my failed attempts to correct the problem).
ggplot(data=biogeo, aes(x=Site, y=Percent, fill=Biogeographic.affinity)) + geom_bar(stat="identity", colour="black")+
scale_fill_grey() + ylab("Percent") + xlab("Location") +
theme_bw()+ theme(panel.grid.minor = element_blank())
This gives the basic graph but the colors and order are still wrong. To correct the order I tried, but that did not change anything (FRUSTRATED)!:
newone <- transform(biogeo, Biogeographic.affinity = factor(Biogeographic.affinity ), Rank = factor(Rank, levels = 1:4))
As for color changing I have tried and seems to work but it all looks like the order is still wrong!
cols<- c("Bassian"="darkgrey","Widespread"="lightgrey", "Torresian"="white", "Eyrean"="black") #designates the colors of the bars
ggplot(data=newone, aes(x=Site, y=Percent, fill=Biogeographic.affinity)) + geom_bar(stat="identity", colour="black")+
scale_fill_manual(values = cols) + ylab("Percent") + xlab("Location") +
theme_bw()+ theme(panel.grid.minor = element_blank())
please help.
To reorder bars manually, you have to pass stat=”identity” in the geom_bar() function. Example: R.
To set colors for bars in Bar Plot drawn using barplot() function, pass the required color value(s) for col parameter in the function call. col parameter can accept a single value for color, or a vector of color values to set color(s) for bars in the bar plot.
Rearranging Results in Basic R Then draw the bar graph of the new object. If you want the bar graph to go in descending order, put a negative sign on the target vector and rename the object. Then draw the bar graph of the new object.
The order that bars are drawn (bottom to top) in a stacked barplot in ggplot2 is based on the ordering of the factor which defines the groups. So the Biogeographic.affinity factor must be reordered. Generally we use reorder
(if we want to order the factor according to a continuous levels) but here I will just create a new ordered factor similar to what you tried to do.
biogeo <- transform(biogeo,
Biog.aff.ord = factor(
Biogeographic.affinity ,
levels=c( 'Bassian','Widespread','Torresian', 'Eyrean'),
ordered =TRUE))
Now if you fill your barplot using Biog.aff.ord rather than the original factor and overriding the default grouping order by defining aes_group_order as Biog.aff.ord order you get the expected result:
cols <- c(Bassian="darkgrey",Widespread="lightgrey",
Torresian="white", Eyrean="black")
ggplot(data=biogeo, aes(x=Site, y=Percent,
order=Biog.aff.ord)) + ##!! aes_group_order
geom_bar(stat="identity", colour="black",
aes(fill=Biog.aff.ord)) +
scale_fill_manual(values = cols)
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With