CohenK: Cohen's Kappa

In BavoDC/AGREL: Package for calculating agreement/reliability indices

Description

Slightly adjusted function from kappa2 to calculate Cohen's Kappa (Cohen, 1960).

Usage

CohenK(ratings, weight = c("unweighted", "equal", "squared"),
  sort.levels = FALSE)

Arguments

ratings	A dataframe or matrix of N x 2 with N the number of observations. The columns contain the ratings of the 2 raters.
weight	The weight matrix that has to be used to calculate the reliability. Default is unweighted, 'squared' can be used to calculate Cohen's Weighted Kappa (Cohen, 1968).
sort.levels	Sort if the levels are numeric.

Value

method	Method that was used to calculate reliability and weights used
subjects	Number of subjects in the dataframe
nraters	Number of raters
irr.name	Type of reliability measure
value	Value for Cohen's Kappa
StdErr	The standard error of the estimated Kappa value
stat.name	The corresponding test statistic
statistic	The value of the test statistic
p.value	The p-value for the test
Po	Overall proportion of agreement
Pe	Proportion of agreement expected by chance
ratings	The original dataframe with the ratings

References

Cohen, J. (1960). A Coefficient of Agreement for Nominal Scales. Educational and Psychological Measurement, Vol.20(1), pp.37-46

Cohen, J. (1968).Weighted kappa: Nominal scale agreement provision for scaled disagreement or partial credit. Psychological Bulletin, Vol.70(4), pp.213-220

Examples

# Load data
data(PsychMorbid)
Df = PsychMorbid[,1:2]

# Unweighted kappa
CohenK(Df)

# Weighted kappa
data(Agreement_deVet)
CohenK(Agreement_deVet[,2:3], weight = "squared")

BavoDC/AGREL documentation built on May 6, 2019, 7:22 a.m.