S3 Method for Computing Observed and Expected Frequencies of Univariate and Bivariate Marginals

This S3 method computes observed and expected frequencies for univariate and bivariate distributions.

IRT.frequencies(object, ...)

IRT_frequencies_default(data, post, probs, weights=NULL)

IRT_frequencies_wrapper(object, ...)

# S3 method for din
IRT.frequencies(object, ...)

# S3 method for gdina
IRT.frequencies(object, ...)

# S3 method for mcdina
IRT.frequencies(object, ...)

# S3 method for gdm
IRT.frequencies(object, ...)

# S3 method for slca
IRT.frequencies(object, ...)

Arguments

object: Object of classes din, gdina, mcdina, gdm or slca.
...: More arguments to be passed.
data: Item response data as extracted by IRT.data
post: Individual posterior distribution as extracted by IRT.posterior
probs: Individual posterior distribution as extracted by IRT.irfprob
weights: Optional vector of weights as included as the attribute weights in IRT.data

Value

List with following entries

uni_obs: Univariate observed distribution
uni_exp: Univariate expected distribution
M_obs: Univariate observed means
M_exp: Univariate expected means
SD_obs: Univariate observed standard deviations
SD_exp: Univariate expected standard deviations
biv_obs: Bivariate observed frequencies
biv_exp: Bivariate expected frequencies
biv_N: Bivariate sample size
cov_obs: Observed covariances
cov_cor: Expected covariances
cor_obs: Observed correlations
cor_exp: Expected correlations
chisq: Chi square statistic of local independence

Examples

if (FALSE) {
#############################################################################
# EXAMPLE 1: Usage IRT.frequencies
#############################################################################

data(sim.dina, package="CDM")
data(sim.qmatrix, package="CDM")

# estimate GDINA model
mod1 <- CDM::gdina( data=sim.dina,  q.matrix=sim.qmatrix)
summary(mod1)

# direct usage of IRT.frequencies
fres1 <- CDM::IRT.frequencies(mod1)

# use of the default function with input data
data <- CDM::IRT.data(object)
post <- CDM::IRT.posterior(object)
probs <- CDM::IRT.irfprob(object)
fres2 <- CDM::IRT_frequencies_default(data=data, post=post, probs=probs)
}