Is there a way to plot multiple levels variables in R in separate graph with a simple code?

Question

I wonder if somebody knows how to execute this code, but that each level is separately plotted.

sites<-levels(OrchardSprays$treatment)
par(mfcol=c(4,2))
for (i in 1:length(sites)) {
  here_tmp<-sites[i]
  plot(droplevels(subset(OrchardSprays, Site = here_tmp, select = c(treatment,decrease))))
  }

This is the output that it gives me. But I want different graphs of the different levels. I don't know why it gives me the same graphs...

Answer 1

I agree with @SimonG that it's not clear why you'd want to have the each level plotted on a separate graph, but here's a way to do it with ggplot2 , which has a nice system for creating plots of each level of a variable without much extra code:

library(ggplot2)

ggplot(OrchardSprays, aes(treatment, decrease)) +
  geom_boxplot() +
  facet_wrap(~treatment, scales="free_x", ncol=2)

To put all the boxplots in a single graph, just remove the last line:

ggplot(OrchardSprays, aes(treatment, decrease)) +
  geom_boxplot()

Answer 2

Your problem seems to be with how you use subset . First, there is no variable named Site in the data set. Second, Site = here_tmp is not a logical expression.

Assuming you meant treatment instead of Site , try this:

sites <- levels(OrchardSprays$treatment)
par(mfcol=c(4,2))
for (i in 1:length(sites)) {
  here_tmp<-sites[i]
  plot(droplevels(subset(OrchardSprays, treatment == here_tmp, 
       select = c(treatment,decrease))))
}

I'm not really sure though if this is your desired outcome because, personally, I don't see a reason to put them in separate plots.

Let me know if this was helpful!