watchme_ira: Calculates interrater agreement using the 'irr' package. The...

Description Usage Arguments Value Examples

Description

Calculates interrater agreement using the irr package. The unit of comparison is one picture.

Usage

1
2
watchme_ira(results_list, names_list = NULL, one_to_one = TRUE,
  by_code = FALSE)

Arguments

results_list

a list of tibble created by watchme_prepare_data.

names_list

(optional) a vector of names for the coders. It must be the same length as results_list and contains different names.

one_to_one

a boolean indicating whether Cohen's kappa should be calculated for each possible pair of coders in case of more than 2 coders, instead of Fleiss's Kappa for all coders at the same time.

by_code

boolean indicating whether the IRR should be calculated for each code separately, or for annotations as a whole.

Value

A tbl_df presenting the results of a call to the irr function. If there are only two raters the called function is kappa2, unweighted. If there are more than two raters and one_to_one is FALSE, the called function is kappam.fleiss.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
data('coding1')
data('coding2')
# With two coders
results_list <- list(coding1, coding2)
names_list <- c('Cain', 'Abel')
watchme_ira(results_list, names_list = names_list)
watchme_ira(results_list, names_list = names_list, by_code = TRUE)
# With three coders
results_list2 <- list(coding1, coding1, coding2)
names_list2 <- c('Riri', 'Fifi', 'Loulou')
watchme_ira(results_list2, names_list = names_list2, one_to_one = FALSE)
watchme_ira(results_list2, names_list = names_list2, one_to_one = TRUE)
watchme_ira(results_list2, names_list = names_list2, one_to_one = FALSE, by_code = TRUE)
watchme_ira(results_list2, names_list = names_list2, one_to_one = TRUE, by_code = TRUE)

masalmon/watchme documentation built on May 21, 2019, 12:41 p.m.