Table with more than 2 dependent variables in finalfit

[andresimi]

Is it possible to create a table with more than 2 dependent variables in finalfit?

I as looking for something like this:

[ewenharrison]

Thanks for this.
This is a pivoted stratified table. So stratify in the normal manner and then pivot.
You will need library(gt) or similar to sort your final formatting, but finalfit will give you all the numbers you need.
Something like this.
If this was useful to people it could be wrapped into a neat function.
All you should need to do is change the explanatory, dependent and split variable names.
Let me know if you have issues.

library(finalfit)
library(tidyverse)

# Table rows
explanatory = c("obstruct.factor", "perfor.factor")

# Table columns
dependent = "rx.factor"
split = c("sex.factor", "age.factor")

# Piped function to generate stratified crosstabs table
colon_s %>%
  group_by(!!! syms(split)) %>% 
  group_modify(~ summary_factorlist(.x, dependent, explanatory)) %>%
  mutate(label = na_if(label, "")) %>% 
  fill(label) %>% 
  pivot_wider(names_from =  split, 
              values_from = colon_s %>% pull(dependent) %>% levels()) %>% 
  mutate(label = rm_duplicates(label)) %>% 
  as.data.frame()

        label levels Obs_Female_<40 years Obs_Female_40-59 years Obs_Female_60+ years Obs_Male_<40 years Obs_Male_40-59 years
1 Obstruction     No             8 (66.7)              50 (80.6)            57 (79.2)           7 (58.3)            51 (83.6)
2                Yes             4 (33.3)              12 (19.4)            15 (20.8)           5 (41.7)            10 (16.4)
3 Perforation     No           12 (100.0)              60 (95.2)            73 (98.6)          12 (92.3)            60 (98.4)
4                Yes              0 (0.0)                3 (4.8)              1 (1.4)            1 (7.7)              1 (1.6)
  Obs_Male_60+ years Lev_Female_<40 years Lev_Female_40-59 years Lev_Female_60+ years Lev_Male_<40 years Lev_Male_40-59 years
1          75 (83.3)             9 (90.0)              35 (71.4)            59 (81.9)           7 (77.8)            45 (70.3)
2          15 (16.7)             1 (10.0)              14 (28.6)            13 (18.1)           2 (22.2)            19 (29.7)
3          89 (96.7)           10 (100.0)              48 (96.0)            70 (95.9)           8 (88.9)            63 (96.9)
4            3 (3.3)              0 (0.0)                2 (4.0)              3 (4.1)           1 (11.1)              2 (3.1)
  Lev_Male_60+ years Lev+5FU_Female_<40 years Lev+5FU_Female_40-59 years Lev+5FU_Female_60+ years Lev+5FU_Male_<40 years
1          88 (87.1)                 9 (56.2)                  43 (81.1)                76 (83.5)               9 (90.0)
2          13 (12.9)                 7 (43.8)                  10 (18.9)                15 (16.5)               1 (10.0)
3         101 (98.1)               16 (100.0)                  52 (98.1)                91 (96.8)             10 (100.0)
4            2 (1.9)                  0 (0.0)                    1 (1.9)                  3 (3.2)                0 (0.0)
  Lev+5FU_Male_40-59 years Lev+5FU_Male_60+ years
1                41 (85.4)              63 (82.9)
2                 7 (14.6)              13 (17.1)
3                51 (98.1)              76 (96.2)
4                  1 (1.9)                3 (3.8)

[andresimi]

thanx! that solved the problem

[andresimi]

Actually, I have another question: can I get weighted frequencies for this table?

[ewenharrison]

See this discussion
#13