inspectdf (version 0.0.2)

inspect_imb: Summarise and compare columnwise imbalance for non-numeric columns in one or two dataframes.

Description

Summarise and compare columnwise imbalance for non-numeric columns in one or two dataframes.

Usage

inspect_imb(df1, df2 = NULL, show_plot = FALSE)

Arguments

df1

A data frame

df2

An optional second data frame for comparing columnwise imbalance. Defaults to NULL.

show_plot

(Deprecated) Logical flag indicating whether a plot should be shown. Superseded by the function show_plot() and will be dropped in a future version.

Value

A tibble summarising and comparing the imbalance for each non-numeric column in one or a pair of data frames.

Details

When a single data frame is specified, a tibble is returned which contains columnwise imbalance, with columns

• col_name character vector containing column names of df1.

• value character vector containing the most common categorical level in each column of df1.

• pcnt the percentage of each column's entries occupied by the level in value column.

• cnt the number of occurrences of the most common categorical level in each column of df1.

When both df1 and df2 are specified, the most common levels in df1 are compared to columns in df2. If a categorical column appears in both dataframes, a simple test is performed to test the null hypothesis that the rate of occurrence of the common level in df1 is the same in both dataframes. The resulting tibble has columns

• col_name character vector containing column names of df1 and df2.

• value character vector containing the most common categorical level in each column of df1.

• pcnt_ the percentage of each column's entries occupied by the level in value column.

• cnt_ the number of occurrences of the most common categorical level in each column of df1 and df2.

Examples

# NOT RUN {
data("starwars", package = "dplyr")
# get tibble of most common levels
inspect_imb(starwars)
# compare imbalance 
inspect_imb(starwars, starwars[1:10, -3])
# }