irtpp: Estimate a test item parameters.

In IRTpp: Estimating IRT Parameters using the IRT Methodology

Description

Estimate a test item parameters according to Item Response Theory.

Usage

irtpp(dataset = NULL, model, dims = 1, initialvalues = NULL,
  filename = NULL, output = NULL, loglikflag = F,
  convergenceEpsilon = 1e-04)

Arguments

dataset	The matrix with the responses from the individuals.
model	The model used to calibrate the parameters.
dims	The dimensions to use on the estimation, remember to use the initial parameters if you want highquality estimation.
initialvalues	The matrix with the initial values for the optimization process.
filename	Optional argument specifying a CSV file to read instead of a dataset in memory.
output	Optional. Additonal arguments that need.
loglikflag	Optional. Show the loglikelihood at the end of estimation procedure. Also shows AIC and BIC statistic.
convergenceEpsilon	Optional. Convergence value, default value of 1E-4

Value

A list containing the estimates of the model parameters, the number of iterations, the loglikelihood final, the final values of the estimation procedure EM.

Examples

## Simulation data for the model "1pl"
# data <- simulateTest(model = "1PL", items = 10, individuals = 500)
## Estimation of the parameters
# irtpp(dataset = data$test, model = "1PL")

## Simulation data for the model "2pl"
# data <- simulateTest(model = "2PL", items = 20, individuals = 800)
## Estimation of the parameters
# irtpp(dataset = data$test, model = "2PL")

## Simulation data for the model "3pl"
# data <- simulateTest(model = "3PL", items = 100, individuals = 1000)
## Estimation of the parameters
# irtpp(dataset = data$test, model = "3PL") 

Example output

IRTpp documentation built on May 29, 2017, 9:58 a.m.