crosstable (version 0.8.0)

crosstable_test_args: Default arguments for calculating and displaying tests in crosstable()

Description

This is the starting point for refining the testing algorithm used in crosstable. Users can provide their own functions for test.~.

Usage

crosstable_test_args(
  test_summarize = test_summarize_auto,
  test_tabular = test_tabular_auto,
  test_correlation = test_correlation_auto,
  test_survival = test_survival_logrank,
  test_display = display_test,
  plim = 4,
  show_method = TRUE
)

Value

A list with test parameters

Arguments

test_summarize

a function of two arguments (continuous variable and grouping variable), used to compare continuous variable. Must return a list of two components: p.value and method. See test_summarize_auto or test_summarize_linear_contrasts for some examples of such functions.

test_tabular

a function of two arguments (two categorical variables), used to test association between two categorical variables. Must return a list of two components: p.value and method. See test_tabular_auto for example.

test_correlation

a function of three arguments (two continuous variables plus the correlation method), used to test association between two continuous variables. Like cor.test, it must return a list of at least estimate, p.value, and method, with also conf.int optionally. See test_correlation_auto for example.

test_survival

a function of one argument (the formula surv~by), used to compare survival estimations. Must return a list of two components: p.value and method. See test_survival_logrank for example.

test_display

function used to display the test result. See display_test.

plim

number of digits for the p value.

show_method

whether to display the test name (logical).

Author

Dan Chaltiel

See Also

test_summarize_auto, test_tabular_auto, test_survival_logrank, test_summarize_linear_contrasts, display_test

Examples

library(dplyr)
my_test_args=crosstable_test_args()
my_test_args$test_summarize = test_summarize_linear_contrasts
iris %>%
  mutate(Petal.Width.qt = paste0("Q", ntile(Petal.Width, 5)) %>% ordered()) %>%
  crosstable(Petal.Length ~ Petal.Width.qt, test=TRUE, test_args = my_test_args)