Breakdown of a numeric variable by grouping variable(s)
Calculates means, variances and ns for subgroups of numeric observations and displays the results.
brkdn(formula,data,maxlevels=10,num.desc=c("mean","var","valid.n"),
width=10,round.n=2)formula |
a formula with the variable to be broken down on the left and the names of one or more variables defining subgroups on the right |
data |
the data frame from which to select the variables |
maxlevels |
the maximum number of levels in any subgroup |
num.desc |
names of the summary functions to apply to the variable on the left side of the formula |
width |
The number of characters to allow for each column in the summary output. |
round.n |
The number of decimal places to round the output. |
brkdn will accept a formula referring to columns in a data frame. It calls describe.numeric for the calculations and displays a table of results.
The results of describe.numeric, or a multi-level list if more than one grouping variable is specified.
Jim Lemon
test.df<-data.frame(Age=rnorm(100)+3*10,Sex=sample(c("M","F"),100,TRUE),
Employ=sample(c("FT","PT","NO"),100,TRUE))
# add value labels for Employ in alphabetical order so they match
attr(test.df$Employ,"value.labels")<-c("Full time","None","Part time")
brkdn(Age~Sex+Employ,test.df)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.