table for multiple variables with multiple labels

Question

I am trying to create a expss like cross table for multiple variables . i have a database which already labelled.

df <- data.frame("TB1"=c("OPS", "OPS",  "HR",   "ADMIN",    "HR",   "ADMIN",    "ADMIN",    "HR",   "HR",   "HR",   "Sales",    "Sales",    "Sales",    "HR",   "HR",   "HR",   "HR",   "Sales",    "Sales"),
                 "TB2"=c("Sales",   "ADMIN",    "ADMIN",    "Sales",    "ADMIN",    "ADMIN",    "ADMIN",    "HR",   "HR",   "HR",   "HR",   "HR",   "HR",   "OPS",  "OPS",  "OPS",  "OPS",  "HR",   "HR"),
                 "TB3"=c("ADMIN",   "Sales",    "OPS",  "Sales",    "HR",   "ADMIN",    "HR",   "HR",   "ADMIN",    "ADMIN",    "HR",   "HR",   "HR",   "OPS",  "HR",   "OPS",  "HR",   "HR",   "Sales"),
                 "TB4"=c("Global",  "Regional", "Regional", "Global",   "Global",   "Regional", "Regional", "Global",   "Global",   "Regional", "Regional", "Global",   "Global",   "Regional", "Global",   "Regional", "Global",   "Regional", "Global"))



banner <- with(df, list(total(),TB4))

t1 <- df %>% tab_cells(mdset(TB1 %to% TB3)) %>%
  tab_cols(total(), banner) %>%
  cross_cpct() %>%
  tab_pivot()

the table should be look like below

Answer 1

Here is a suggestion! I am not quite sure about the percent. But using gtsummary with tbl_strata function and tbl_cross function could be a good start for you: https://www.danieldsjoberg.com/gtsummary/reference/tbl_strata.html https://www.danieldsjoberg.com/gtsummary/reference/tbl_cross.html

library(dplyr)
library(tidyr)
library(gtsummary)
df %>% 
  pivot_longer(
    -TB4
  ) %>% 
  tbl_strata(
    strata = TB4,
    .tbl_fun = 
      ~ .x %>% 
      tbl_cross(
        row = value, 
        col = name,  
        percent = "row",
        margin="row")
  )

Answer 2

Something like this:

library(expss)

df <- data.frame("TB1"=c("OPS", "OPS",  "HR",   "ADMIN",    "HR",   "ADMIN",    "ADMIN",    "HR",   "HR",   "HR",   "Sales",    "Sales",    "Sales",    "HR",   "HR",   "HR",   "HR",   "Sales",    "Sales"),
                 "TB2"=c("Sales",   "ADMIN",    "ADMIN",    "Sales",    "ADMIN",    "ADMIN",    "ADMIN",    "HR",   "HR",   "HR",   "HR",   "HR",   "HR",   "OPS",  "OPS",  "OPS",  "OPS",  "HR",   "HR"),
                 "TB3"=c("ADMIN",   "Sales",    "OPS",  "Sales",    "HR",   "ADMIN",    "HR",   "HR",   "ADMIN",    "ADMIN",    "HR",   "HR",   "HR",   "OPS",  "HR",   "OPS",  "HR",   "HR",   "Sales"),
                 "TB4"=c("Global",  "Regional", "Regional", "Global",   "Global",   "Regional", "Regional", "Global",   "Global",   "Regional", "Regional", "Global",   "Global",   "Regional", "Global",   "Regional", "Global",   "Regional", "Global"))



df %>% 
    to_long(keep = TB4) %>% 
    tab_cols(list(total(), TB4) %nest% variable) %>%
    tab_cells("|" = value) %>% 
    tab_stat_cpct() %>%
    tab_pivot()