R: Generation of Simulated Item Response Data under gIRT Model

gendata_gIRT {lazy.girt}

R Documentation

Generation of Simulated Item Response Data under gIRT Model

Description

Generation of Simulated Item Response Data under gIRT Model

Usage

gendata_gIRT(Ntot, param, DinP = 1, Nmat = NULL, thetaphi = NULL,
  npoint = NULL, theta = NULL, npointth = 7, thmin = -3, thmax = 3,
  phi = NULL, npointph = 5, phmin = 0, phmax = 5, thd = NULL,
  thdist = "NORMAL", thmean = 0, thstd = 1, phd = NULL,
  phdist = "GBETA", paramab = c(1, 3), nomiss = 0, compress = 0,
  sort = 1, thetaname = NULL)

Arguments

`Ntot`	# of total observations to be generated If Ntot == 1, one observation per each theta point will be generated, otherwize, the # of observations per each theta point is proportional to round(Ntot*thd) where thd is the npoint x 1 vector of theta distribution. Therefore, Ntot must be large enough.
`param`	Item Parameter Data Frame
`DinP`	= 1 to include D=1.7 in logistic function
`Nmat`	npoints x nitems matrix consisting of # of trials, or NULL. This will be used only when Ntot=1 and compress=0.
`theta`	Discrete theta values
`thmin`	Minimum value of discrete thata value
`thmax`	Maximum value of discrete thata value
`thd`	theta distribution probability vector
`thdist`	= "NORMAL" or "UNIFORM" or "rnorm" or "runif" When "NORMAL" or "rnorm", thmean and thstd will be used to generate thd. When "UNIFORM" or "runif", thmin and thmax will be used to generate thd. When "rnorm" or "runif", theta will be generated using npoints random numbers and thd and Ntot is set equal to 1.
`thmean`	mean of theta distribution
`thstd`	std of theta distribution
`nomiss`	= 1 to exclude those theta points with N = 0.
`compress`	= 1 to compress the data when Ntot = 1.
`sort`	= 1 to sort the result according to theta.
`thetaname`	= Name of the variable which will contain theta value or NULL.
`npoints`	# of discrete points for theta, or # of random variables

Value

A list of
U, N, npoints, theta, thd, Ntot, fromP, toP, type, ncat, thmean, thstd , compress, sort

where

If compress = 0 and Ntot is large,
U is npoints x sum of ncat[j]
U[,fromP[j]:toP[j]] is npoints x ncat[j]
U[,fromP[j]:toP[j]][i,k] contains # of responses to
the k-th category (k=0,1,...,ncat[j]) at theta[i]
to which N[i]=round(Ntot*thd[i]) observations belong.

If compress = 1 and Ntot = 1,
U is npoints x nitems
U[i,j] contains the response to the j-th item at theta[i]
where 0 <= U[i,j] <= ncat[j]-1.
In this case, fromP and toP are not compressed.

(theta, thd, N) are npoints x 1
N[i] is the # of observations at theta[i]

Examples

set.seed(1701)
# 30 binary items
param=paramBr
# one obs per random theta-phi
resg <- gendata_gIRT( 1, param, thdist="NORMAL", phdist="GBETA", compress=1
                     , npoint=500, phmin=0, phmax=2, paramab=c(1,4) )
Uc1=as.data.frame( resg$U )
thetaphi_ORG1=resg$thetaphi

set.seed(1701)
param=paramBG
# one obs per theta-phi grid
resg <- gendata_gIRT( 1, param, thdist="NORMAL", phdist="GBETA", compress=1
                     , npointth=501, thmin=-4, thmax=4
                     , npointph=3, phmin=0, phmax=2 )
Uc2=as.data.frame( resg$U )
thetaphi_ORG2=resg$thetaphi

[Package lazy.girt version 0.1.3 Index]