raggregate

aggregate means and keep N

I'm trying to aggregate a dataset, but I want to also keep the number of observations. So what I have is similar to this:

aggregate(iris$Sepal.Length, by=list(iris$Species), FUN=mean)

But that returns an object like so:

 Group.1     x
1     setosa 5.006
2 versicolor 5.936
3  virginica 6.588

when what I want is that AND the number of observations (rows) in each group (in a separate column)

Solution

  • Trial and error revealed that this works:

    FUN = function(x) c(m = mean(x), n = length(x))

    There are other ways to do this in packages like dplyr and data.table.