Freq: Frequency Table for a Single Variable

Description

Calculates absolute and relative frequencies of a vector x. Continuous (numeric) variables will be cut using the same logic as used by the function hist. Categorical variables will be aggregated by table. The result will contain single and cumulative frequencies for both, absolute values and percentages.

Usage

Freq(x, breaks = hist(x, plot = FALSE)$breaks, include.lowest = TRUE,
     ord = c("level", "desc", "asc", "name"),
     useNA = c("no", "ifany", "always"), ...)
# S3 method for Freq
print(x, digits = NULL, ...)

Value

an object of type "Freq", which is basically a data.frame with 5 columns (earning a specific print routine), containing the following components:

level: factor. The levels of the grouping variable.
freq: integer. The absolute frequencies.
perc: numeric. The relative frequencies (percent).
cumfreq: integer. The cumulative sum of the absolute frequencies.
cumperc: numeric. The cumulative sum of the relative frequencies.

Arguments

x: the variable to be described, can be any atomic type.
breaks: either a numeric vector of two or more cut points or a single number (greater than or equal to 2) giving the number of intervals into which x is to be cut. Default taken from the function hist(). This is ignored if x is not of numeric type.
include.lowest: logical, indicating if an x[i] equal to the lowest (or highest, for right = FALSE) "breaks" value should be included. Ignored if x is not of numeric type.
ord: how should the result be ordered? Default is "level", other choices are 'by frequency' ("descending" or "ascending") or 'by name of the levels' ("name"). The argument can be abbreviated. This is ignored if x is numeric.
useNA: one out of "no", "ifany", "always". Defines whether to include extra NA levels in the table. Defaults to "no" which is the table() default too.
digits: integer, determining the number of digits used to format the relative frequencies.
...: further arguments are passed to the function cut(). Use dig.lab to control the format of numeric group names. Use the argument right to define if the intervals should be closed on the right (and open on the left) or vice versa.
In print.Freq the dots are not used.

Author

Andri Signorell <andri@signorell.net>

Details

By default only the valid cases are considered for the frequencies, say NA values are excluded. (This is in accordance with the default behavior of the R function table, which seemed a reasonable reference.) If the NAs should be included you can set the useNA argument to either "ifany" or "always".

For numeric variables, if breaks is specified as a single number, the range of the data is divided into breaks pieces of equal length, and then the outer limits are moved away by 0.1% of the range to ensure that the extreme values both fall within the break intervals. (If x is a constant vector, equal-length intervals are created that cover the single value.) See cut.

Examples

Run this code

data(d.pizza)

# result is a data.frame
d.freq <- Freq(d.pizza$price)
d.freq

# it is printed by default with 3 digits for the percent values,
# but the number of digits can be defined in the print function
print(d.freq, digits=5)

# sorted by frequency
Freq(d.pizza$driver, ord="desc")

# sorted by name using all the observations, say including NAs
Freq(d.pizza$driver, ord="name", useNA="ifany")

# percentages and cumulative frequencies for a vector of count data
Freq(as.table(c(2,4,12,8)))

Run the code above in your browser using DataLab