tests/testthat/test-check_assumptions.R

In cedricbatailler/JSmediation: Mediation Analysis Using Joint Significance

skip_if_not_installed("performance")

test_that("check_assumptions run without error", {
  ho_et_al$condition_c <- build_contrast(ho_et_al$condition,
                                         "Low discrimination",
                                         "High discrimination")

  my_model <-
    mdt_simple(data = ho_et_al,
               IV = condition_c,
               DV = hypodescent,
               M = linkedfate)

  expect_error(check_assumptions(my_model), NA)
})

test_that("check_assumptions test assumptions for each models", {
  ho_et_al$condition_c <- build_contrast(ho_et_al$condition,
                                         "Low discrimination",
                                         "High discrimination")

  my_model <-
    mdt_simple(data = ho_et_al,
               IV = condition_c,
               DV = hypodescent,
               M = linkedfate)

  purrr::pluck(my_model, "js_models") %>%
    names() %>%
    stringr::str_replace("\\+", "\\\\+") %>%
    purrr::map(~ expect_output(check_assumptions(my_model), .x))
})

test_that("check_assumptions's `tests` argument ", {
  ho_et_al$condition_c <- build_contrast(ho_et_al$condition,
                                         "Low discrimination",
                                         "High discrimination")

  my_model <-
    mdt_simple(data = ho_et_al,
               IV = condition_c,
               DV = hypodescent,
               M = linkedfate)

  expect_error(check_assumptions(my_model, tests = NULL))
  expect_error(check_assumptions(my_model, tests = c(1L, 2L)))
  expect_warning(check_assumptions(my_model, tests = c("foo", "bar")))
  expect_warning(check_assumptions(my_model, tests = c("normality", "foo")))
  expect_output(
    check_assumptions(my_model,
                      tests = c("normality",
                                "heteroscedasticity",
                                "outliers"))
  )

})