QualitativeStatistics: Compare qualitative variables between a grouping variable

In kwdoyle/kevtools: Kevin's Useful Functions for Stats, etc.

Compare qualitative variables between a grouping variable

Description

Make contingency table of counts per group and, if the table is 2x2, perform Fisher's exact test. If not, perform a chi square test. Can perform tests counts of each value within a variable as well if multilevel is set to TRUE.

Usage

QualitativeStatistics(
  data,
  id_var,
  group_var,
  tst_vars,
  multilevel = FALSE,
  test_use = "proportion",
  yates = TRUE,
  flip_dir = FALSE,
  correct_var = NULL
)

Arguments

data	the input data frame
id_var	the id variable to group obsetrvations by
group_var	the grouping variable, passed as a string
tst_vars	variables to perform the test on; can be passed as a vector of strings
multilevel	if TRUE, then function will perform analyses on all values with a variable between the group
test_use	specify either 'proportion' or 'logistic_regress' Note: logistic regression is only done if multilevel == TRUE
yates	whether to use yate's continuity correction or not. chisq.test defaults to TRUE, while gtsummary's add_p defaults to FALSE
flip_dir	short for 'flip directionality.' set to true if want to predict the variable *using* the outcome instead
correct_var	if performing a logistic regression, can specify an additional variable to correct for

Examples

out <- QualitativeStatistics(data, group_var = "CMD", tst_vars = c("Gender", "Etiology", "Race"), multilevel = TRUE)
# since multilevel is TRUE, will have results for male and female, all etiologies, and all races.
print(out$Etiology$SAH)
print(out$Etiology$ICH)
print(out$Gender$M)
print(out$Gender$F)

kwdoyle/kevtools documentation built on Jan. 4, 2023, 10:32 p.m.