[R] Boxplot using Formula

[Sundar Dorai-Raj]

Matt Goff wrote:
> I am trying to use the formula interface for the boxplot.
> Currently running R 2.2.1 on Windows XP.
> 
> The problem is that boxplot is displaying groups that are empty in the  
> plot.
> 
> The following example demonstrates what it is happening (though my actual  
> situation is a little more complicated):
> 
> 
>>data<-data.frame(values=c(1:25), groups=rep(c("A","B","C","D","E"),  
>>each=5))
>>boxplot(data$values~data$groups, subset=data$groups!="C")
> 
> 
> The plot includes a space for level "C" with no boxplot.  I would like to  
> have only those levels with associated boxplots be shown in the figure.   
> So in this case, there should only be values "A", "B", "D", and "E" evenly  
> spaced along the horizontal axis.
> 
> I am a little confused because I am reusing old code and I don't remember  
> having this issue before, so perhaps there have been some changes in the  
> behavior of the boxplot function since the last time I use the code.
> 
> It seems like there's probably a simple solution, but I'm not seeing it.
> 
> Thanks,
> 
> Matt Goff
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html

Hi, Matt,

I'm afraid you cannot do what you want with boxplot.formula. You have at 
least two options:

1) Supply only data that has no empty levels.
2) Create a new function that does what you need:

## add `drop.unused.levels' argument to boxplot.formula
boxplot2 <-
function (formula, data = NULL, ...,
           subset, na.action = NULL,
           drop.unused.levels = TRUE)
{
     if (missing(formula) || (length(formula) != 3))
         stop("'formula' missing or incorrect")
     m <- match.call(expand.dots = FALSE)
     if (is.matrix(eval(m$data, parent.frame())))
         m$data <- as.data.frame(data)
     m$... <- NULL
     m$na.action <- na.action
     m$drop.unused.levels <- drop.unused.levels
     m[[1]] <- as.name("model.frame")
     mf <- eval(m, parent.frame())
     response <- attr(attr(mf, "terms"), "response")
     boxplot(split(mf[[response]], mf[-response]), ...)
}

values <- 1:25
groups <- factor(rep(c("A","B","C","D","E"), each=5))
boxplot2(values ~ groups, subset = groups != "C")

HTH,

--sundar