janitor (version 0.3.0)

adorn_crosstab: Add presentation formatting to a crosstabulation table.

Description

Formats a data.frame containing counts of co-occurences of two variables (i.e., a contingency table or crosstab). Adds a mix of percentages, Ns, totals row/column, and custom rounding to a table of integer counts, in the style of a Microsoft Excel PivotTable. The result is no longer clean data, but is an audience-friendly way to report results.

Designed to run on the output of a call to janitor::crosstab, but can be called on any data.frame containing a contingency table, e.g., the result of dplyr::count() followed by tidyr::spread().

Usage

adorn_crosstab(dat, denom = "row", show_n = TRUE, digits = 1,
  show_totals = FALSE, rounding = "half to even")

Arguments

dat

a data.frame with row names in the first column and numeric values in all other columns. Usually the piped-in result of a call to crosstab that included the argument percent = "none".

denom

the denominator to use for calculating percentages. One of "row", "col", or "all".

show_n

should counts be displayed alongside the percentages?

digits

how many digits should be displayed after the decimal point?

show_totals

display a totals summary? Will be a row, column, or both depending on the value of denom.

rounding

method to use for truncating percentages - either "half to even", the base R default method, or "half up", where 14.5 rounds up to 15.

Value

Returns a data.frame.

Examples

# NOT RUN {
mtcars %>%
 crosstab(gear, cyl) %>%
 adorn_crosstab(denom = "all")
 
# showing with all parameters
mtcars %>%
 crosstab(gear, cyl) %>%
 adorn_crosstab(., denom = "col", rounding = "half up", show_n = FALSE, digits = 2)
mtcars %>%
 crosstab(cyl, am) %>%
 adorn_crosstab(., denom = "all", digits = 0, rounding = "half up") 
# }