Hey I'd like to access the exact positions of the boxplot when dodging is applied as in the case below:
testPlot <- ggplot(iris, aes(x=Species, y=Sepal.Width, fill=Petal.Width > 1)) +
geom_boxplot(position="dodge")
testPlot
This is important for instance so that one can add annotation directly above one box without fiddling. I can't find anything in str(testPlot).
Inspecting the output of ggplot_build(testPlot) can give you the data used to plot the boxplot. Here is the output.
# $data[[1]]
# fill ymin lower middle upper ymax outliers notchupper notchlower x PANEL group ymin_final ymax_final xmin xmax weight colour size alpha shape linetype
# 1 #F8766D 2.9 3.20 3.4 3.675 4.2 4.4, 2.3 3.506137 3.293863 1.0000 1 1 2.3 4.4 0.625 1.375 1 grey20 0.5 NA 19 solid
# 2 #00BFC4 2.3 2.70 2.9 3.000 3.4 2.2 2.972284 2.827716 2.1875 1 3 2.2 3.4 2.000 2.375 1 grey20 0.5 NA 19 solid
# 3 #F8766D 2.0 2.25 2.4 2.500 2.7 2.549296 2.250704 1.8125 1 2 2.0 2.7 1.625 2.000 1 grey20 0.5 NA 19 solid
# 4 #00BFC4 2.5 2.80 3.0 3.175 3.6 3.8, 2.2, 3.8 3.083792 2.916208 3.0000 1 4 2.2 3.8 2.625 3.375 1 grey20 0.5 NA 19 solid
# -------------------------------------------------------------------------
Using such information for example, we can annotate a text on top of the [1] boxplot at a particular location as follows
testPlot + annotate(geom="text", x=1, y=3.3, label="Sample Text here",
color="blue")
How did I find the coordinate (1,3.3)? If you see closely the first row, you can see that the lower line (Q1) aka first quartile at 3.20 and y is between 2.9 and 4.2. You can also see the value of the outliers. The x value is between 0.625 and 1.375. You can follow the same principle and annotate the desired boxplot.
Hope that helps.