df_props: Joint and conditional proportions

In mosaicModel: An Interface to Statistical Modeling Independent of Model Architecture

Description

Uses a formula interface to specify how the proportions are to be calculated.

Usage

df_props(formula, data, as.percent = FALSE, ..., wide = FALSE,
  margins = FALSE, format = c("proportion", "percent", "count"))

Arguments

formula	the formula describing the relationship
data	a data frame (or you can pipe this in)
as.percent	show proportions in percent (e.g. multiply by 100)
...	statistics functions to be applied to the data, e.g. mean, sd, confidence(0.95)
wide	reformat the output as a cross-tabulation. This makes sense only when there are just two variables
margins	show the marginal probabilities. Makes the most sense if wide = TRUE.
format	Use just for internal purposes.

Details

Using | in the formula specifies a conditional proportion

• ~ A : proportion of cases in each level of A

• ~ A + B: joint proportion: each level of A crossed with B

• ~ A | B: conditional proportion: for each level of B, what fraction are in each level of A

• A ~ B: another way of specifying the conditional proportion

Examples

df_props(mtcars, ~ cyl + gear) 
df_props(mtcars, ~ cyl | gear)
df_props(mtcars, ~ cyl + gear, wide = TRUE)
df_props(mtcars, ~ cyl + gear, margins = TRUE)
df_props(mtcars, ~ cyl | gear, margins = TRUE)

mosaicModel documentation built on May 2, 2019, 7:59 a.m.