boot.pair: Compare dependent pairwise kappas (bootstrap method)

In svanbelle/multiagree: Comparaison of dependent kappa coefficients

Compare dependent pairwise kappas (bootstrap method)

Description

This function performs Hotelling's T square test using a variance-covariance matrix based on the bootstrap method to compare dependent pairwise kappa coefficients

Usage

boot.pair(cluster_id, data, weight = "equal", a.level = 0.05, ITN = 1000,
  summary_k = T)

Arguments

cluster_id	a vector of lenght N with the identification number of the clusters
data	a N x R matrix representing the classification of the N items by the R observers. The kappa coefficients are computed between column (1,2), (3,4), etc....
weight	the weighting scheme to be used for kappa coefficients. 'unweighted' for Cohen's kappa, 'equal' for linear weights and 'squared' for quadratic weights
a.level	significance level
ITN	the number of bootstrap iterations
summary_k,	if true, Hotteling's T square test is performed, if false, only the bootstraped kappa coefficients are returned

Details

This function compares several dependent pairwise kappa coefficients using Hotelling's T square with the variance-covariance matrix obtained by the bootstrap method. If only one kappa is computed, it returns the estimate and confidence interval.

Value

$kappa a G x 2 matrix with the G kappa coefficients to be compared in the first column and their corresponding standard error in the second column

$T_test a vector of length 2 with the value of Hotelling's T square test as first element and the corresponding p-value as second element

$confidence confidence intervals for the pairwise comparisons of the kappa coefficients

$cor the G x G correlation matrix of the kappa coefficients

$K when summary_k is false, the ITN x G matrix with the bootstrapped kappa coefficients

Author(s)

Sophie Vanbelle [e-mail]

References

Vanbelle S. and Albert A. (2008). A bootstrap method for comparing correlated kappa coefficients. Journal of Statistical Computation and Simulation, 1009-1015

Vanbelle S. Comparing dependent agreement coefficients obtained on multilevel data. submitted

Vanbelle S. (2014) A New Interpretation of the Weighted Kappa Coefficients. Psychometrika. Advance online publication. doi: 10.1007/s11336-014-9439-4

Examples

 
#dataset (not multilevel) (Vanbelle and Albert, 2008)

set.seed(103) #to get the same results as in the paper
data(depression)
attach(depression)
a<-boot.pair(data=cbind(diag,BDI,diag,GHQ),cluster_id=ID,weight='unweighted')


#dataset (multilevel) (Vanbelle, xxx)

data(FEES)
attach(FEES)
dat<-cbind(val_CO,val_COR,val_MH,val_MHR,val_TB,val_TBR) #formating the data matrix
boot.pair(data=dat,cluster_id=subject,weight='equal',summary_k=FALSE)

svanbelle/multiagree documentation built on Feb. 9, 2023, 2:37 p.m.