I have run 3 different Kappa2 analysis using the irr package.
library(irr)
data("diagnoses", package = "irr")
# Analysis 1
kappa2(diagnoses[, c("rater1", "rater2")], weight = "unweighted")
# Analysis 2
kappa2(diagnoses[, c("rater1", "rater3")], weight = "unweighted")
# Analysis 3
kappa2(diagnoses[, c("rater3", "rater2")], weight = "unweighted")
Is there a good way to generate summary of the output as presentation table (flextable if possible) for r markdown?. I`m looking for a table with: Analysis 1, Analysis 2, and Analysis 3 as rows. Subjects, Raters, Kappa, z and p-value as columns.
Would something like this work for your case ?
We gather all the analysis in a list and extract the relevant values from it using lapply
.
library(irr)
analysis1 <- kappa2(diagnoses[, c("rater1", "rater2")], weight = "unweighted")
analysis2 <- kappa2(diagnoses[, c("rater1", "rater3")], weight = "unweighted")
analysis3 <- kappa2(diagnoses[, c("rater3", "rater2")], weight = "unweighted")
list_model <- dplyr::lst(analysis1, analysis2, analysis3)
purrr::map_df(list_model, function(x) {
data.frame(Subjects = x$subjects,
Raters = x$raters,
kappa = x$value,
p_value = x$p.value)
}) -> result
flextable::flextable(cbind(model = names(list_model), result))