meanKappa: Cohens kappa among several raters

In eatModel: Educational Assessment Tools for IRT Modeling using Conquest or TAM

Description

This is a wrapper for the kappa2 function from the irr package. Function computes Cohens kappa among several raters (at least 2) for one item and several persons.

Usage

meanKappa( dat , weight = "unweighted" , weight.mean = TRUE )

Arguments

dat	Data frame with at least two columns, with examiness in the rows and raters in the columns.
weight	either a character string specifying one predifined set of weights or a numeric vector with own weights (see details). (see help file of the kappa2 function).
weight.mean	Logical: TRUE, if agreement is weighted by number of rater subjects. FALSE, if it is averaged among all rater pairs.

Value

A list. First element is a data frame with kappa values between raters pairs. Second element is a scalar with mean kappa among all raters.

Author(s)

Alexander Robitzsch

Examples

data(rater)
v01 <- subset(rater, variable == "V01")
dat <- reshape2::dcast( v01, id~rater, value.var = "value")
kap <- meanKappa(dat[,-1])

eatModel documentation built on May 2, 2019, 6:49 p.m.