Learn R Programming

CDM (version 8.2-6)

IRT.modelfit: S3 Methods for Assessing Model Fit

Description

This S3 method assesses global (absolute) model fit using the methods described in modelfit.cor.din.

Usage

IRT.modelfit(object, ...)

# S3 method for din IRT.modelfit(object, ...) # S3 method for gdina IRT.modelfit(object, ...)

# S3 method for IRT.modelfit.din summary(object, ...) # S3 method for IRT.modelfit.gdina summary(object, ...)

Value

See output of modelfit.cor.din.

Arguments

object

Object of classes din or gdina.

...

More arguments to be passed.

See Also

For extracting the individual likelihood or posterior see IRT.likelihood or IRT.posterior.

The model fit of objects of class gdm can be obtained by using the TAM::tam.modelfit.IRT function in the TAM package.

Examples

Run this code
if (FALSE) {
#############################################################################
# EXAMPLE 1: Absolute model fit
#############################################################################

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

#*** Model 1: DINA model for DINA simulated data
mod1 <- CDM::din( sim.dina, q.matrix=sim.qmatrix, rule="DINA" )
fmod1 <- CDM::IRT.modelfit( mod1 )
summary(fmod1)
  ##  Test of Global Model Fit
  ##         type value     p
  ##  1   max(X2) 8.728 0.113
  ##  2 abs(fcor) 0.143 0.080
  ##
  ##  Fit Statistics
  ##                    est
  ##  MADcor          0.030
  ##  SRMSR           0.040
  ##  100*MADRESIDCOV 0.671
  ##  MADQ3           0.062
  ##  MADaQ3          0.059

#*** Model 2: GDINA model
mod2 <- CDM::gdina( sim.dina, q.matrix=sim.qmatrix, rule="GDINA" )
fmod2 <- CDM::IRT.modelfit( mod2 )
summary(fmod2)
  ##  Test of Global Model Fit
  ##         type value p
  ##  1   max(X2) 2.397 1
  ##  2 abs(fcor) 0.078 1
  ##
  ##  Fit Statistics
  ##                    est
  ##  MADcor          0.023
  ##  SRMSR           0.030
  ##  100*MADRESIDCOV 0.515
  ##  MADQ3           0.075
  ##  MADaQ3          0.071
}

Run the code above in your browser using DataLab