Search code examples
rgtsummarygt

gtsummary help repeated variable

enter image description here

df %>% 
    tbl_summary(by=Field)

It seems when there are repetitions in some variables, this package gives strange result. It has given me the n(%) of each variable instead of median(IQR) and p values. Is there a way to get around this?

I tried everything possible.

Solution

  • The issue is that by default tbl_summary uses the categorical summary type for numerics with less than 10 unique levels. From the docs ?tbl_summary:

    ... numeric variables with fewer than 10 unique levels default to type categorical.

    To fix that you have to explicitly set the type to be continuous:

    To change a numeric variable to continuous that defaulted to categorical, use type = list(varname ~ "continuous")

    Using some fake example data:

    library(gtsummary)

set.seed(123)

df <- data.frame(
  Alphonso = sample(1:6, 100, replace = TRUE),
  Field = sample(c("field1", "field2"), 100, replace = TRUE)
)

df %>% 
  tbl_summary(by=Field)

    enter image description here

    df %>% 
  tbl_summary(by=Field, type = list(Alphonso ~ "continuous"))

    enter image description here