| responsesm4pl {irtProb} | R Documentation |
Simulation of response patterns and computation of the probability of the patterns according to the multidimensional one, two, three and four person parameters logistic item response models.
grm4pl(N = 10, theta = 0, S = 0, C = 0, D = 0, s = 1/1.702, b = 0, c = 0, d = 1)
ggrm4pl(n=5,rep=1,theta=0,S=rep(0,length(theta)),C=rep(0,length(theta)),
D=rep(0,length(theta)),s=rep(1/1.702,n),b=rep(0,n),c=rep(0,n),
d=rep(1, n))
pggrm4pl(x=ggrm4pl(rep=1),rep=1,n=dim(x)[2],N=dim(x)[1],theta=rep(0,N),
S=0,C=0,D=0,s=rep(1/1.702,n),b=rep(0,n),c=rep(0,n),d=rep(1,n),
log.p=FALSE, TCC=FALSE)
x |
integer matrix; response patterns (0 or 1). |
rep |
numeric; number of replications of the simulation of the response patterns. |
n |
numeric; number of items. |
N |
numeric; number of response patterns |
theta |
numeric; vector of proficiency levels (z sscores). |
S |
numeric; person fluctuation parameter. |
C |
numeric; person pseud0-guessing parameter. |
D |
numeric; person inattention parameter. |
s |
numeric; item fluctuation parameters. |
b |
numeric; item difficulty parameters. |
c |
numeric; item pseudo-guessing parameters. |
d |
numeric; item inattention parameters. |
log.p |
logical; if TRUE, probabilities p are given as log(p). |
TCC |
logical; if TRUE generate the TCC figures for each response patterns. Default FALSE. |
The function grm4pl generates N responses to an item according to the person parameters and the items parameters.
The funcfion ggrm4pl will be used to generate rep respose patterns at n items. To compute
the probability of the response patterns, according to known person and item parameters, the function pggrm4pl will be applied.
grm4pl |
integer; vector of item responses (0 or 1). |
ggrm4pl |
integer data.frame; responses for n items. |
pggrm4pl |
graphic; if (TCC ==TRUE) return(list(prob=prob, tcc=tcc)). If (TCC==FALSE) return(prob). |
Gilles Raiche, Universite du Quebec a Montreal (UQAM),
Departement d'education et pedagogie
Raiche.Gilles@uqam.ca, http://www.er.uqam.ca/nobel/r17165/
Ferrando, P. J. (2004). Person reliability in personality measurement: an item response theory analysis. Applied Psychological Measurement, 28(2), 126-140.
Hulin, C. L., Drasgow, F., and Parsons, C. K. (1983). Item response theory. Homewood, IL: Irwin.
Levine, M. V., and Drasgow, F. (1983). Appropriateness measurement: validating studies and variable ability models. In D. J. Weiss (Ed.): New horizons in testing. New York, NJ: Academic Press.
Magis, D. (2007). Enhanced estimation methods in IRT. In D. Magis (Ed.): Influence, information and item response theory in discrete data analysis. Doctoral dissertation, Liege, Belgium: University de Liege.
Trabin, T. E., and Weiss, D. J. (1983). The person response curve: fit of individuals to item response theory models. In D. J. Weiss (Ed.): New horizons in testing. New York, NJ: Academic Press.
gr4pl, ggr4pl, pggr4pl,
ctt2irt, irt2ctt
## ....................................................................
# Generation of response patterns (0,1) from r4pl() for N subjects (default value
# of N = 10)
# Generation of a response (0,1) from rm4pl for N subjects
grm4pl(theta=0)
grm4pl(N=5, theta=c(-4,4), c=0)
# Generation of n m4pl response patterns (0,1) for [rep * length(theta)] subjects
# The subject number ia equal to [rep * length(theta)]
# a,b,c et d are item parameters vectors
nitems <- n <- 7; N <- 1
s <- rep(0,nitems); b <- seq(-4,4,length=nitems); c <- rep(0,nitems)
d <- rep(1,nitems)
theta <- seq(-4,4,length=5)
x <- ggrm4pl(n=nitems, rep=N, theta=theta,s=s,b=b,c=c,d=d)
x
# TO BE REWORKED - Probability of a response pattern and test caracteristic curve
# (TCC)
nItems <- n <- 7; N <- 1
s <- rep(0,nItems); b <- seq(-4,4,length=nItems)
c <- rep(0,nItems); d <- rep(1,nItems)
theta <- seq(-4,4,length=5); S <- rep(1/1.702,length(theta));
C <- rep(0.3,length(theta)); D <- rep(0,length(theta))
x <- ggrm4pl(n=nItems, rep=N, theta=theta, S=S, C=C, D=D, s=s, b=b, c=c, d=d)
x
res <- pggrm4pl(x=x, rep=N, theta=theta, S=1/1.702, C=0.3, D=0, s=s, c=c, d=d,
TCC=TRUE)
res
res <- pggrm4pl(x=x, rep=N, theta=rep(2,length(theta)), S=1/1.702, C=0, D=0,
s=s, c=c, d=d, TCC=FALSE)
res
pggrm4pl(theta=3)
pggrm4pl(n=10, theta=seq(-4,4,length=5), x=ggrm4pl(rep=1), TCC=TRUE)
## ....................................................................