This function uses car::Anova() to calculate global p-values for model covariates. Output from tbl_regression and tbl_uvregression objects supported.

add_global_p(x, ...)

# S3 method for tbl_regression
add_global_p(
  x,
  include = everything(),
  type = NULL,
  keep = FALSE,
  anova_fun = NULL,
  quiet = NULL,
  ...
)

# S3 method for tbl_uvregression
add_global_p(
  x,
  type = NULL,
  include = everything(),
  keep = FALSE,
  anova_fun = NULL,
  quiet = NULL,
  ...
)

Arguments

x

Object with class 'tbl_regression' or 'tbl_uvregression'

...

Additional arguments to be passed to car::Anova, aod::wald.test() or anova_fun (if specified)

include

Variables to calculate global p-value for. Input may be a vector of quoted or unquoted variable names. Default is everything()

type

Type argument passed to car::Anova(type=). Default is "III"

keep

Logical argument indicating whether to also retain the individual p-values in the table output for each level of the categorical variable. Default is FALSE.

anova_fun

Function that will be used in place of car::Anova() when specified to calculate the global p-values.

  • function must return a tibble matching the output of car::Anova() %>% broom::tidy() including a columns called "term" and "p.values"

  • function must accept arguments anova_fun(x, ...), where x is a model object

  • arguments passed in ... will be passed to anova_fun(...)

  • the add_global_p(type=) argument is ignored in anova_fun=

  • a common function used here is tidy_wald_test(), a wrapper for aod::wald.test()

quiet

Logical indicating whether to print messages in console. Default is FALSE

Example Output

Example 1

image of rendered example table

Example 2

image of rendered example table

Author

Daniel D. Sjoberg

Examples

# \donttest{
# Example 1 ----------------------------------
tbl_lm_global_ex1 <-
  lm(marker ~ age + grade, trial) %>%
  tbl_regression() %>%
  add_global_p()

# Example 2 ----------------------------------
tbl_uv_global_ex2 <-
  trial[c("response", "trt", "age", "grade")] %>%
  tbl_uvregression(
    method = glm,
    y = response,
    method.args = list(family = binomial),
    exponentiate = TRUE
  ) %>%
  add_global_p()
# }