krippen.alpha.raw: Krippendorff's alpha coefficient
In raredd/ragree: Rater agreement

Description Usage Arguments Details Author(s) References

Computes Krippendorff's alpha coefficient and standard error for multiple raters when data is an n x r matrix of alphanumeric ratings from n subjects and r raters, excluding all subjects that are not rated by any rater.

1 2	krippen.alpha.raw(ratings, weights = "unweighted", conflev = 0.95, N = Inf, print = TRUE)

`ratings`	an `n x r` matrix of raw alphanumeric ratings from `n` subjects and `r` raters
`weights`	optional weighting; see details
`conflev`	confidence level
`N`	used as denominator in finite population correction
`print`	logical; if `TRUE`, prints a summary of the agreement

weight is an option matrix of weights or one of "quadratic", "linear", "ordinal", "radical", "ratio", "circular", "bipolar", or "unweighted" for an identity matrix.

The algorithm used to compute Krippendorff's alpha is very different from anything that was published on this topic. Instead, it follows the equations presented by K. Gwet (2010).

Kilem L. Gwet

Gwet, K. (2012). Handbook of Inter-Rater Reliability: the Definitive Guide to Measuring the Extent of Agreement among Multiple Raters, 3rd Edition. Advanced Analytics, LLC; 3rd edition (March 2, 2012).

Krippendorff (1970). "Bivariate agreement coefficients for reliability of data." Sociological Methodology, 2, 139-150.

Krippendorff (1980). Content analysis: An introduction to its methodology (2nd ed.), New-bury Park, CA: Sage.

raredd/ragree documentation built on March 25, 2021, 1:42 p.m.