Add an estimate value to references rows for categorical variables

For categorical variables with a treatment contrast (stats::contr.treatment()) or a SAS contrast (stats::contr.SAS()), will add an estimate equal to 0 (or 1 if exponentiate = TRUE) to the reference row.

Usage

tidy_add_estimate_to_reference_rows(
  x,
  exponentiate = attr(x, "exponentiate"),
  conf.level = attr(x, "conf.level"),
  model = tidy_get_model(x),
  quiet = FALSE
)

Arguments

x

(data.frame)
A tidy tibble as produced by tidy_*() functions.

exponentiate

(logical)
Whether or not to exponentiate the coefficient estimates. It should be consistent with the original call to broom::tidy()

conf.level

(numeric)
Confidence level, by default use the value indicated previously in tidy_and_attach(), used only for sum contrasts.

model

(a model object, e.g. glm)
The corresponding model, if not attached to x.

quiet

(logical)
Whether broom.helpers should not return a message when requested output cannot be generated. Default is FALSE.

Details

For categorical variables with a sum contrast (stats::contr.sum()), the estimate value of the reference row will be equal to the sum of all other coefficients multiplied by -1 (eventually exponentiated if exponentiate = TRUE), and obtained with emmeans::emmeans(). The emmeans package should therefore be installed. For sum contrasts, the model coefficient corresponds to the difference of each level with the grand mean. For sum contrasts, confidence intervals and p-values will also be computed and added to the reference rows.

For other variables, no change will be made.

If the reference_row column is not yet available in x, tidy_add_reference_rows() will be automatically applied.

See also

Other tidy_helpers: tidy_add_coefficients_type(), tidy_add_contrasts(), tidy_add_header_rows(), tidy_add_n(), tidy_add_pairwise_contrasts(), tidy_add_reference_rows(), tidy_add_term_labels(), tidy_add_variable_labels(), tidy_attach_model(), tidy_disambiguate_terms(), tidy_group_by(), tidy_identify_variables(), tidy_plus_plus(), tidy_remove_intercept(), tidy_select_variables()

Examples

# \donttest{
  df <- Titanic |>
    dplyr::as_tibble() |>
    dplyr::mutate(dplyr::across(where(is.character), factor))

  glm(
    Survived ~ Class + Age + Sex,
    data = df, weights = df$n, family = binomial,
    contrasts = list(Age = contr.sum, Class = "contr.SAS")
  ) |>
    tidy_and_attach(exponentiate = TRUE) |>
    tidy_add_reference_rows() |>
    tidy_add_estimate_to_reference_rows()
#> # A tibble: 9 × 14
#>   term        variable   var_class var_type var_nlevels contrasts contrasts_type
#>   <chr>       <chr>      <chr>     <chr>          <int> <chr>     <chr>         
#> 1 (Intercept) (Intercep… NA        interce…          NA NA        NA            
#> 2 Class1st    Class      factor    categor…           4 contr.SAS treatment     
#> 3 Class2nd    Class      factor    categor…           4 contr.SAS treatment     
#> 4 Class3rd    Class      factor    categor…           4 contr.SAS treatment     
#> 5 ClassCrew   Class      factor    categor…           4 contr.SAS treatment     
#> 6 Age1        Age        factor    dichoto…           2 contr.sum sum           
#> 7 Age2        Age        factor    dichoto…           2 contr.sum sum           
#> 8 SexFemale   Sex        factor    dichoto…           2 contr.tr… treatment     
#> 9 SexMale     Sex        factor    dichoto…           2 contr.tr… treatment     
#> # ℹ 7 more variables: reference_row <lgl>, estimate <dbl>, std.error <dbl>,
#> #   statistic <dbl>, p.value <dbl>, conf.low <dbl>, conf.high <dbl>

  glm(
    response ~ stage + grade * trt,
    gtsummary::trial,
    family = binomial,
    contrasts = list(
      stage = contr.treatment(4, base = 3),
      grade = contr.treatment(3, base = 2),
      trt = contr.treatment(2, base = 2)
    )
  ) |>
    tidy_and_attach() |>
    tidy_add_reference_rows() |>
    tidy_add_estimate_to_reference_rows()
#> # A tibble: 12 × 14
#>    term        variable  var_class var_type var_nlevels contrasts contrasts_type
#>    <chr>       <chr>     <chr>     <chr>          <int> <chr>     <chr>         
#>  1 (Intercept) (Interce… NA        interce…          NA NA        NA            
#>  2 stage1      stage     factor    categor…           4 contr.tr… treatment     
#>  3 stage2      stage     factor    categor…           4 contr.tr… treatment     
#>  4 stage3      stage     factor    categor…           4 contr.tr… treatment     
#>  5 stage4      stage     factor    categor…           4 contr.tr… treatment     
#>  6 grade1      grade     factor    categor…           3 contr.tr… treatment     
#>  7 grade2      grade     factor    categor…           3 contr.tr… treatment     
#>  8 grade3      grade     factor    categor…           3 contr.tr… treatment     
#>  9 trt1        trt       character dichoto…           2 contr.SAS treatment     
#> 10 trt2        trt       character dichoto…           2 contr.SAS treatment     
#> 11 grade1:trt1 grade:trt NA        interac…          NA NA        NA            
#> 12 grade3:trt1 grade:trt NA        interac…          NA NA        NA            
#> # ℹ 7 more variables: reference_row <lgl>, estimate <dbl>, std.error <dbl>,
#> #   statistic <dbl>, p.value <dbl>, conf.low <dbl>, conf.high <dbl>
# }