DescTools (version 0.99.37)

PercTable: Percentage Table

Description

Prints a 2-way contingency table along with percentages, marginal, and conditional distributions. All the frequencies are nested into one single table.

Usage

# S3 method for default
PercTable(x, y = NULL, …)
# S3 method for table
PercTable(tab, row.vars = NULL, col.vars = NULL, justify = "right",
          freq = TRUE, rfrq = "100", expected = FALSE, residuals = FALSE,
          stdres = FALSE, margins = NULL, digits = NULL, …)
# S3 method for formula
PercTable(formula, data, subset, na.action, …)
# S3 method for PercTable
print(x, vsep = NULL, ...)
Margins(tab, ...)

Arguments

x, y

objects which can be interpreted as factors (including character strings). x and y will be tabulated via table(x, y). If x is a matrix, it will be coerced to a table via as.table(x).

tab

a r x c-contingency table

row.vars

a vector of row variables (see Details).

col.vars

a vector of column variables (see Details). If this is left to NULL the table structure will be preserved.

justify

either "left" or "right" for defining the alignment of the table cells.

freq

boolean. Should absolute frequencies be included? Defaults to TRUE.

rfrq

a string with 3 characters, each of them being 1 or 0. The first position means total percentages, the second means row percentages and the third column percentages. "011" produces a table output with row and column percentages.

expected

the expected counts under the null hypothesis.

residuals

the Pearson residuals, (observed - expected) / sqrt(expected).

stdres

standardized residuals, (observed - expected) / sqrt(V), where V is the residual cell variance (for the case where x is a matrix, n * p * (1 - p) otherwise).

margins

a vector, consisting out of 1 and/or 2. Defines the margin sums to be included. 1 stands for row margins, 2 for column margins, c(1,2) for both. Default is NULL (none).

digits

integer. With how many digits shoud the relative frequencies be formatted? Default can be set by DescToolsOptions(digits=x).

formula

a formula of the form lhs ~ rhs where lhs will be tabled versus rhs (table(lhs, rhs)).

data

an optional matrix or data frame (or similar: see model.frame) containing the variables in the formula formula. By default the variables are taken from environment(formula).

subset

an optional vector specifying a subset of observations to be used.

na.action

a function which indicates what should happen when the data contain NAs. Defaults to getOption("na.action").

vsep

logical, defining if an empty row should be introduced between the table rows. Default is FALSE, if only a table with one single description (either frequencies or percents) should be returned and TRUE in any other case.

…

the dots are passed from PercTable.default() to PercTable.table() and from Margins to the function Freq.

Value

Returns an object of class "ftable".

Details

PercTable prints a 2-dimensional table. The absolute and relative frequencies are nested into one flat table by means of ftable. row.vars, resp. col.vars can be used to define the structure of the table. row.vars can either be the names of the dimensions (included percentages are named "idx") or numbers (1:3, where 1 is the first dimension of the table, 2 the second and 3 the percentages). Use Sort() if you want to have your table sorted by rows. The style in which numbers are formatted is selected by Fmt() from the DescTools options. Absolute frequencies will use Fmt("abs") and Fmt("per") will do it for the percentages. The options can be changed with Fmt(abs=as.fmt(...)) which is basically a "fmt"-object containing any format information used in Format.

Margins() returns a list containing all the one dimensional margin tables of a n-dimensional table along the given dimensions. It uses margin.table() for all the dimensions and adds the appropriate percentages.

References

Agresti, Alan (2007) Introduction to categorical data analysis. NY: John Wiley and Sons, Section 2.4.5

See Also

Freq, table, ftable, prop.table, addmargins, DescToolsOptions, Fmt There are similar functions in package sfsmisc printTable2 and package vcd table2d_summary, both lacking some of the flexibility we needed here.

Examples

# NOT RUN {
tab <- table(driver=d.pizza$driver, area=d.pizza$area)

PercTable(tab=tab, col.vars=2)

PercTable(tab=tab, col.vars=2, margins=c(1,2))
PercTable(tab=tab, col.vars=2, margins=2)
PercTable(tab=tab, col.vars=2, margins=1)
PercTable(tab=tab, col.vars=2, margins=NULL)

PercTable(tab=tab, col.vars=2, rfrq="000")

# just the percentages without absolute values
PercTable(tab=tab, col.vars=2, rfrq="110", freq=FALSE)

# just the row percentages in percent format (pfmt = TRUE)
PercTable(tab, freq= FALSE, rfrq="010", pfmt=TRUE, digits=1)

# just the expected frequencies and the standard residuals
PercTable(tab=tab, rfrq="000", expected = TRUE, stdres = TRUE)


# rearrange output such that freq are inserted as columns instead of rows
PercTable(tab=tab, col.vars=c(3,2), rfrq="111")

# putting the areas in rows
PercTable(tab=tab, col.vars=c(3,1), rfrq="100", margins=c(1,2))

# formula interface with subset
PercTable(driver ~ area, data=d.pizza, subset=wine_delivered==0)

# sort the table by rows, order first column (Zurich), then third, then row.names (0)
PercTable(tab=Sort(tab, ord=c(1,3,0)))

# reverse the row variables, so that absolute frequencies and percents
# are not nested together
PercTable(tab, row.vars=c(3, 1))

# the vector interface
PercTable(x=d.pizza$driver, y=d.pizza$area)
PercTable(x=d.pizza$driver, y=d.pizza$area, margins=c(1,2), rfrq="000", useNA="ifany")

# one dimensional x falls back to the function Freq()
PercTable(x=d.pizza$driver)

# the margin tables
Margins(Titanic)

# }