6
votes

I have data in the following format:

  # repetition, packet, route, energy level
  1, 1, 1, 10.0
  1, 1, 2, 12.3
  1, 1, 3, 13.8
  1, 2, 1, 9.2
  1, 2, 2, 10.1
  1, 2, 3, 11.2
  ...
  50,99,3, 0.01

Now, I want to create a plot showing box plots per route per packet over all repetitions. So, for example the x-axis would depict the packets and the y-axis the energy level. The first tick on the x-axis would show three box plots which contain data of three subsets

  subset(data, data$packet == 1 & data$route == 1)
  subset(data, data$packet == 1 & data$route == 2)
  subset(data, data$packet == 1 & data$route == 3)

and so on. I'm using ggplot2 and I'm wondering if I have to create each time a boxplot and try to add them into one or if there is a smart way to do this?

Thanks in advance! M.

1
boxplot(y ~ x) will make side by side boxplots for a variable y, grouped by the unique values for x but, since this is purely a programming question, this is off topic here, so I've voted close and migrate to stackoverflow.Macro
It is a bit more complicated. It is not only a variable y (energy) compared by x (route) but also packets (?).michael

1 Answers

8
votes

If you're using ggplot2, you'll be able to do this quite nicely with facet_wrap, which can create multiple boxplots next to each other. For example:

library(ggplot2)
mydata = data.frame(x=as.factor(rep(1:2, 5, each=5)), y=rnorm(50),
        division=rep(letters[1:5], each=10))

print(ggplot(mydata, aes(x, y)) + geom_boxplot() + facet_wrap(~division))

enter image description here

In the case of your code, you look like you might actually want to divide by two variables (it's a little unclear). If you want to divide it by route and then by packet (as your example seems to suggest) you can use facet_grid:

print(ggplot(data, aes(repetition, energy.level)) + geom_boxplot() + facet_grid(route ~ packet))

However, note that since you have 99 packets this would end up being 99 graphs wide, so you probably want to try a different approach.