Perform normality test for multiple variables by groups

This function performs a Shapiro-Wilk normality test on multiple variables in a dataframe by groups.

Usage

ntest_by(df, cols, group)

Arguments

df

A dataframe containing the data to be tested.

cols

A vector of column names in the dataframe to be tested for normality.

group

A column name in the dataframe that defines the groups.

Value

A data frame with the results of the Shapiro-Wilk test for each group and variable.

Details

The function first creates subsets of the data for each group. It then performs a Shapiro-Wilk normality test (`shapiro.test()`) on each group and returns the results in a tidy format. the function also checks whether the number of observations for each variable and group lies within the permissible range of the `shapiro.test()`. If not, the function will provide diagnostics on why a test could not be performed base on which groups are affected.

See also

[ntest()] to run normality tests for ungrouped data

Examples

if (FALSE) { # \dontrun{
# Perform a normality test for groups
ntest_by(df    = mtcars,
         cols  = mpg,
         group = cyl)

# Perform a normality test for multiple variables by groups
ntest_by(df    = starwars,
         cols  = c(mass, birth_year),
         group = species)
} # }