Should variable be populated for intercept terms? or stay NA as current? about broom.helpers HOT 6 CLOSED

OK decided. PR merged.

from broom.helpers.

I do not think it would be a problem to populate variable column with "(Intercept)".

Let me check

from broom.helpers.

Dear @ddsjoberg ,

in fact it is already possible. Please note that your label option in gtsummary::tbl_regression() corresponds to var_labels option in tidy_plus_plus()

library(broom.helpers)
mod <- glm(response ~ age + grade, gtsummary::trial, family = binomial)
res <- mod %>% tidy_plus_plus(
  intercept = T, 
  variable_labels = list("(Intercept)" = "custom")
)
res %>% dplyr::select(term, variable, var_label, label)
#> # A tibble: 5 x 4
#>   term        variable var_label label 
#>   <chr>       <chr>    <chr>     <chr> 
#> 1 (Intercept) <NA>     custom    custom
#> 2 age         age      Age       Age   
#> 3 gradeI      grade    Grade     I     
#> 4 gradeII     grade    Grade     II    
#> 5 gradeIII    grade    Grade     III

^{Created on 2020-10-21 by the reprex package (v0.3.0)}

from broom.helpers.

However, it is TRUE that it is not clearly stated in the documentation.

from broom.helpers.

Now, I have also explored the possibility that, for intercepts, variable column will be populated with term content (i.e. for most models it will be "(Intercept"). Few adjustments are required (cf. https://github.com/larmarange/broom.helpers/pull/68/files )

So, it will be a minor break change, but it could be considered.

from broom.helpers.

I like the idea of having the variable column populated. It'll make selecting that variable/term/intercept much easier to do by the name the users are seeing. My vote is to populate it!

Also, I should note that since it's populated in gtsummary, if we choose not to populate, I'll populate it manually after running tidy_plus_plus() so that all of the other tbl_regression() helper functions can select it as they need.

from broom.helpers.