Your help is most appreciated to include p-values for gtsummary::tbl_regression
function in scientific notation. I am using big data and it would be really useful to show 3 decimal points for p-values > 0.001 and scientific notation for p-values <0.001.
A representative example
# load packages
library(gtsummary)
theme_gtsummary_compact()
# linear model
mod1 <- lm(mpg ~ cyl + gear, data = mtcars %>% dplyr::mutate(cyl = factor(cyl)))
# regression table
tbl_regression(mod1)
I wonder if it is possible to get the 2.06e-4, 5.76e-8, and 0.422 p-value notations for cyl 6, cyl 8, and gear, respectively.
summary(mod1)
Thanks very much!
You can define your own function for the p-value format:
tbl_regression(
mod1,
pvalue_fun = function(x) {
if_else(
is.na(x),
NA_character_,
if_else(x < 0.001, format(x, digits = 3, scientific = TRUE), format(round(x, 3), scientific = F))
)
}
)