pcX {eba}R Documentation

Paired-Comparison Design Matrix

Description

Computes a paired-comparison design matrix.

Usage

pcX(nstimuli)

Arguments

nstimuli number of stimuli in the paired-comparison design

Details

The design matrix can be used when fitting a Bradley-Terry-Luce (BTL) model or a Thurstone-Mosteller (TM) model by means of glm.

Value

A matrix having (nstimuli-1)*nstimuli/2 rows and nstimuli columns.

See Also

eba, glm.

Examples

data(celebrities)  # absolute choice frequencies
btl <- eba(celebrities)  # fit BTL model using eba
summary(btl)

y1 <- t(celebrities)[lower.tri(celebrities)]
y0 <- celebrities[lower.tri(celebrities)]

## Fit BTL model using glm
btl.glm <- glm(cbind(y1, y0) ~ pcX(9) - 1, binomial)
summary(btl.glm)  # BTL model does not fit these data!

## Fit Thurstone Case V model using glm
tm.glm <- glm(cbind(y1, y0) ~ pcX(9) - 1, binomial(probit))
summary(tm.glm)  # neither fits the TM model

[Package eba version 1.5-0 Index]