kripp.boot: Bootstrap Krippendorff's Alpha

In MikeGruz/kripp.boot: Bootstrap Krippendorff's Alpha Intercoder Reliability Statistic

Description

This function implements Prof. Klaus Krippendorff's algorithm for bootstrapping the Krippendorff's alpha coefficient. It computes confidence values (reliability estimates) for the given probabilities.

Usage

kripp.boot(x, iter = 2000, probs = c(.025, .975), 
           method = c("nominal", "ordinal", "interval", "ratio"))

Arguments

x	is a matrix with rows (in R: observations) corresponding to judges and columns (in R: variables) corresponding to rated objects. Should be numeric, with NAs for missing data.
iter	the number of iterations for bootstrapping.
probs	a vector of probabilities for which confidence values are computed.
method	the metric used to calculate the difference function. "nominal", "ordinal", "interval", and "ratio" are currently implemented.

Value

A list containing the following components:

$mean.alpha	the mean value of all bootstrapped alpha replicates
$alpha	a vector of bootstrapped alphas
$upper	upper alpha value for given probabilities
$lower	lower alpha value for given probabilities
$raters	number of raters used in calculating alpha
$iter	number of bootstrap replications
$probs	vector of probabilities used
$size	number of items used in calculating alpha

Author(s)

Polina Proutskova ([e-mail])

Mike Gruszczynski ([e-mail])

References

Krippendorff, K. (2011). Computing Krippendorff's Alpha-Reliability. Retrieved from http://repository.upenn.edu/asc_papers/43

See Also

Algorithm for bootstrapping a distribution of alpha (http://www.afhayes.com/public/alphaboot.pdf)

Andrew F. Hayes's SPSS code (http://www.afhayes.com/public/kalpha.sps)

Krippendorff, K. (2012). Content analysis: An introduction to its methodology. Sage.

Examples

# Krippendorff's "C" data (2011, 2)
nmm<-matrix(c(1,1,NA,1,2,2,3,2,3,3,3,3,3,3,3,3,2,2,2,2,1,2,3,4,4,4,4,4,
1,1,2,1,2,2,2,2,NA,5,5,5,NA,NA,1,1,NA,NA,3,NA),nrow=4)

# assume default nominal classification with 2000 replicates
kripp.boot(nmm)

# nominal classification with 5000 replicates
kripp.boot(nmm, iter=5000)

# ordinal classification with 5000 replicates
kripp.boot(nmm, iter=5000, method="ordinal")

MikeGruz/kripp.boot documentation built on May 6, 2019, 3:27 p.m.