ac_weights: Weight-generating functions

In emilelatour/lagree: Calculate various interrater agreement coefficients

Weight-generating functions

Description

To do the weighted analysis, you may create your own weight matrix, or use one of the many existing weight-generating functions in the weights.ge.r script. Each weight function takes single mandatory parameter, which is a vector containing all categories used in the study. The weight functions always sort all numeric-type category vectors in ascending order. Consequently, the weighted coefficients are computed properly only if columns and rows in the input dataset are ordered the same way. For alphanumeric-type category vectors, they are assumed to already be ranked following an order that is meaningful to the researcher.

Usage

ac_weights(categ, weight_type = "unweighted")

identity_weights(categ)

quadratic_weights(categ)

linear_weights(categ)

ordinal_weights(categ)

radical_weights(categ)

ratio_weights(categ)

circular_weights(categ)

bipolar_weights(categ)

Arguments

categ	A vector containing all categories used. Be careful that the order matches the the columns and rows of the input data set.
weight_type	A character vector only available for the ac_weights function. Argument for the generic function to select type of weighting.

Value

A q × q matrix of weights, where q is the number of categories. The default argument is "unweighted". With this option, the function will create a diagonal weight matrix with all diagonal numbers equal to 1, and all off-diagonal numbers equal to 0. This special weight matrix leads to the unweighted analysis.

References

1. Handbook of Inter-Rater Reliability: The Definitive Guide to Measuring the Extent of Agreement Among Raters. 4th ed. Gaithersburg, MD: Advanced Analytics.

Examples

ac_weights(c(1:5), "unweighted")
ac_weights(c(1:5), "quadratic")
ac_weights(c("A", "B", "C", "D", "E"), "ordinal")
ac_weights(c("A", "B", "C", "D", "E"), "radical")
ac_weights(c("A", "B", "C", "D", "E"), "ratio")
ac_weights(letters[c(1:5)], "circular")
ac_weights(letters[c(1:5)], "bipolar")
identity_weights(1:5)
quadratic_weights(1:5)
linear_weights(1:5)
ordinal_weights(1:5)
radical_weights(1:5)
ratio_weights(1:5)
circular_weights(1:5)
bipolar_weights(1:5)

emilelatour/lagree documentation built on Sept. 18, 2024, 5:19 p.m.