Rcmdr (version 0.9-3)

stem.leaf: Stem-and-Leaf Display

Description

Creates a classical ("Tukey-style") stem-and-leaf display.

Usage

stem.leaf(data, unit, m, Min, Max, rule.line = c("Dixon", "Velleman", "Sturges"), 
    style = c("Tukey", "bare"), trim.outliers = TRUE, depths = TRUE,
    reverse.negative.leaves = TRUE, print=TRUE)

Arguments

data
a numeric vector.
unit
leaf unit, as a power of 10 (e.g., 100, .01); omit to let the function choose the unit.
m
number of parts (1, 2, or 5) into which each stem should be divided; omit to let the function choose the number of parts/stem.
Min
smallest non-outlying value; omit for automatic choice.
Max
largest non-outlying value; omit for automatic choice.
rule.line
the rule to use for choosing the desired number of lines in the display; "Dixon" = 10*log10(n); "Velleman" = 2*sqrt(n); "Sturges" = 1 + log2(n); the default is "Dixon".
style
"Tukey" (the default) for "Tukey-style" divided stems; "bare" for divided stems that simply repeat the stem digits.
trim.outliers
if TRUE (the default), outliers are placed on LO and HI stems.
depths
if TRUE (the default), print a column of "depths" to the left of the stems; the depth of the stem containing the median is the stem-count enclosed in parentheses.
reverse.negative.leaves
if TRUE (the default), reverse the leaves on negative stems (so, e.g., the leaf 9 comes before the leaf 8, etc.).
print
if TRUE (the default), prints the stem-and-leaf display.

Value

  • If print = TRUE, invisibly returns NULL; otherwise returns a list with information for constructing the stem-and-leaf display.

Details

Unlike the stem function in the base package, this function produces classic stem-and-leaf displays, as described in Tukey's Exploratory Data Analysis. Outliers are determined using the rule for boxplots (see boxplot.stats).

References

Tukey, J. Exploratory Data Analysis. Addison-Wesley, 1977.

See Also

stem

Examples

Run this code
data(Prestige)
stem.leaf(Prestige$income)

Run the code above in your browser using DataCamp Workspace