12

JD Long helped me with this: question about manual annotation.

But is it possible to do something similar on a facetted plot, such that the label style corresponds to the linestyle (aestetics) and in a way that I can annotate different facets individually?

Some data:

funny <- structure(list(Institution = structure(c(1L, 1L, 1L, 1L, 2L, 
2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 
3L, 3L, 3L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 1L, 
1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L), .Label = c("Q-branch", 
"Some-Ville", "Spectre"), class = "factor"), Type = structure(c(5L, 
6L, 1L, 3L, 5L, 6L, 2L, 4L, 5L, 6L, 2L, 4L, 5L, 6L, 2L, 4L, 5L, 
6L, 2L, 4L, 5L, 6L, 2L, 4L, 5L, 6L, 2L, 4L, 5L, 6L, 2L, 4L, 5L, 
6L, 2L, 4L, 5L, 6L, 2L, 4L, 5L, 6L, 2L, 4L, 5L, 6L, 2L, 4L), .Label = c("Korte videregående uddannelser", 
"Mammas beer", "Mellemlange videregående uddannelser", "Tastes good", 
"Unknown", "Your"), class = "factor"), År = c(2008L, 2008L, 
2008L, 2008L, 2008L, 2008L, 2008L, 2008L, 2008L, 2008L, 2008L, 
2008L, 2007L, 2007L, 2007L, 2007L, 2007L, 2007L, 2007L, 2007L, 
2007L, 2007L, 2007L, 2007L, 2007L, 2007L, 2007L, 2007L, 2007L, 
2007L, 2007L, 2007L, 2007L, 2007L, 2007L, 2007L, 2006L, 2006L, 
2006L, 2006L, 2006L, 2006L, 2006L, 2006L, 2006L, 2006L, 2006L, 
2006L), Mndr = c(3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
3L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 15L, 15L, 
15L, 15L, 15L, 15L, 15L, 15L, 15L, 15L, 15L, 15L, 27L, 27L, 27L, 
27L, 27L, 27L, 27L, 27L, 27L, 27L, 27L, 27L), Data = c(159L, 
NA, NA, 23L, 204L, NA, NA, 12L, 256L, NA, NA, 24L, 166L, 6L, 
NA, 43L, 228L, NA, NA, 20L, 196L, 11L, NA, 37L, 99L, 14L, 9L, 
96L, 147L, 7L, 5L, 91L, 100L, 10L, 7L, 126L, 60L, 17L, 6L, 106L, 
78L, 18L, 13L, 140L, 48L, 23L, 5L, 136L)), .Names = c("Institution", 
"Type", "År", "Mndr", "Data"), class = "data.frame", row.names = c(NA, 
-48L))

And a facetted plot:

ggplot(funny, aes(Mndr, y=Data, group=Type, col=Type)) +
  geom_line() +
  facet_grid(.~Institution)

Thanks in advance for your help!

3
  • @Shane: I don't think this is a duplicate - Andreas does not want to use the data in the plot to calculate the value, but rather format his labels to match the style of the lines. Commented Mar 10, 2010 at 14:49
  • @Shane Thank you for the attention. Aniko's answer is exactly what I was hoping for. So I guess there is no need to futher clarify? Unless for posterity of cause. Thanks again. Commented Mar 10, 2010 at 15:07
  • Related question: stackoverflow.com/questions/2050610/… Commented Mar 11, 2010 at 14:17

1 Answer 1

16

The idea is that for each manual annotation you have to define not only the label, but all the variables that define the panel, color, etc. The following code adds two labels in different panels.

pl <- ggplot(funny, aes(Mndr, y=Data, group=Type, col=Type))+geom_line()
      +facet_grid(.~Institution)   #your plot
nd <- data.frame(Institution=c("Q-branch","Some-Ville"),  #panel
                 Type=c("Unknown", "Tastes good"),        #color
                 Mndr=c(7,12),                            #x-coordinate of label
                 Data= c(170,50),                         #y-coordinate of label
                 Text=c("Label 1", "Label 2"))            #label text
# add labels to plot:
pl <- pl + geom_text(aes(label=Text), data=nd, hjust=0, legend=FALSE)
pl

The legend=FALSE option will ensure that the small a's denoting the text are not added to the legend. You don't have to have a data frame for the labels, you could have a separate geom_text for each, but I find this way simpler.

Sign up to request clarification or add additional context in comments.

2 Comments

As far as I can tell from the manual, the new ggplot syntax has changed a little, legend=FALSE is now show_guide = FALSE, I think.
I'm trying to reproduce the plots from code in the question as well as code in the answer, but both plots are identical. No text is written on the plots in the latter case. Has something changed in my version of ggplot2 (2.1.0)?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.