itemfit: Parameter estimation of a test

In SICSresearch/LatentREGpp: Item Response Theory Implemented in R and Cpp

Description

Estimates the test parameters according to the Multidimensional Item Response Theory

Usage

itemfit(data, dim, model = "2PL", EMepsilon = 1e-04, clusters = NULL,
  quad_tech = NULL, quad_points = NULL,
  individual_weights = as.numeric(c()), initial_values = NULL, SD = FALSE,
  verbose = TRUE, save_time = TRUE)

Arguments

data	The matrix containing the answers of tested individuals
dim	The dimensionality of the test
model	"1PL", "2PL" or "3PL"
EMepsilon	Convergence value to determine the accuracy of the test
clusters	A vector with cluster per dimension
quad_tech	A string with technique. "Gaussian" for Gaussian quadrature or "QMCEM" for Quasi-Monte Carlo quadrature
quad_points	Amount of quadrature points. If quadratura_technique is "Gaussian". It can be NULL
individual_weights	A vector with Weights of the quadrature points.
initial_values	A matrix with initial values for estimation process. Be sure about dimension, model and consistency with data.
SD	calculate for standar desviation for items
verbose	True for get information about estimation process in runtime. False in otherwise.
save_time	True for save estimation time. False otherwise.

Models

LatentREGpp has different models to fit likelihood value according parameters to estimate.

3PL

General. Probability is given by

P_{ij} = c_j + \frac{1 - c_j}{1 + exp(-η_{ij})}

Where i references individual and j references the item; c is a value for guessing parameter between 0 and 1. i index is referenced by number of examinees or individuals and j index is referenced by items in test. η is

η_{ij} = \strong{\emph{a}}^{t}_jθ_i+d_j

In unidimensional an a is scalar, in multidimensional an a is vector. For 1PL model a has value 1

2PL

c = 0

1PL

c = 0 a vector has a value 1 for each element

Examples

## Not run: 
#Example 1

dir = normalizePath(system.file(package="LatentREGpp"),winslash = "/")
folder = "/dataset/1D/dicho/"
file = "1000x50-1.csv"
data_dir = paste(c(dir, folder, file), collapse = "")
data = read.table(file = data_dir, sep = ";")
est <- itemfit(data = data, dim = 1)

#Example 2

#Dichotomous and multidimensional data
dir = normalizePath(system.file(package="LatentREGpp"),winslash = "/")
folder = "/dataset/3D/dicho/"
file = "1000x55-1.csv"
data_dir = paste(c(dir, folder, file), collapse = "")
data = read.table(file = data_dir, sep = ";")
clust <- c(20,20,15)
st <- itemfit(data = data, model = "2PL",dim = 3, 
       EMepsilon = 1e-03, clusters = clust, quad_tech = "Gaussian")

## End(Not run)

SICSresearch/LatentREGpp documentation built on May 9, 2019, 11:13 a.m.