sirt (version 0.36-30)

data.read: Dataset Reading

Description

This dataset contains $N=328$ students and $I=12$ items measuring reading competence. All 12 items are arranged into 3 testlets (items with common text stimulus) labeled as A, B and C. The allocation of items to testlets is indicated by their variable names.

Usage

data(data.read)

format

A data frame with 328 persons on the following 12 variables. Rows correspond to persons and columns to items. The following items are included in data.read: Testlet A: A1, A2, A3, A4 Testlet B: B1, B2, B3, B4 Testlet C: C1, C2, C3, C4

Examples

data(data.read)
dat <- data.read
I <- ncol(dat)

library(eRm); library(ltm); library(TAM); library(mRm)
library(CDM); library(mirt)

#****************
# Model 1: Rasch model
#****************

# M1a: rasch.mml2 (in sirt)
mod1a <- rasch.mml2(dat)
summary(mod1a)

# M1b: smirt (in sirt)
Qmatrix <- matrix(1,nrow=I , ncol=1)
mod1b <- smirt(dat,Qmatrix=Qmatrix)
summary(mod1b)

# M1c: gdm (in CDM)
theta.k <- seq(-6,6,len=21)
mod1c <- gdm(dat,theta.k=theta.k,irtmodel="1PL", skillspace="normal")
summary(mod1c)

# M1d: tam.mml (in TAM)
mod1d <- tam.mml( resp=dat )
summary(mod1d)

# M1e: RM (in eRm)
mod1e <- RM( dat )
summary(mod1e)

# M1f: mrm (in mRm)
mod1f <- mrm( dat , cl=1)
mod1f$beta  # item parameters

# M1g: mirt (in mirt)
mod1g <- mirt( dat , model=1 , itemtype="1PL" )
summary(mod1g)
coef(mod1g)

# M1h: ltm (in ltm)
mod1h <- ltm( dat ~ z1 , control=list(verbose=TRUE ) )
summary(mod1h)
coef(mod1h)

#****************
# Model 2: 2PL model
#****************

# M2a: rasch.mml2 (in sirt)
mod2a <- rasch.mml2(dat , est.a=1:I)
summary(mod2a)

# M2b: smirt (in sirt)
mod2b <- smirt(dat,Qmatrix=Qmatrix,est.a="2PL")
summary(mod2b)

# M2c: gdm (in CDM)
mod2c <- gdm(dat,theta.k=theta.k,irtmodel="2PL", skillspace="normal")
summary(mod2c)

# M2d: tam.mml (in TAM)
mod2d <- tam.mml.2pl( resp=dat )
summary(mod2d)

# M2e: mirt (in mirt)
mod2e <- mirt( dat , model=1 , itemtype="2PL" )
summary(mod2e)
coef(mod2e)

# M2f: ltm (in ltm)
mod2f <- ltm( dat ~ z1 , control=list(verbose=TRUE ) )
summary(mod2f)
coef(mod2f)
plot(mod2f)

#****************
# Model 3: 3PL model
#****************

# M3a: rasch.mml2 (in sirt)
mod3a <- rasch.mml2(dat , est.a=1:I, est.c=1:I)
summary(mod3a)

# M3b: smirt (in sirt)
mod3b <- smirt(dat,Qmatrix=Qmatrix,est.a="2PL" , est.c=1:I)
summary(mod3b)

# M3c: mirt (in mirt)
mod3c <- mirt( dat , model=1 , itemtype="3PL" )
summary(mod3c)
coef(mod3c)

# M3d: ltm (in ltm)
mod3d <- tpm( dat , control=list(verbose=TRUE ) , max.guessing=.3)
summary(mod3d)
coef(mod3d) # => numerical instabilities

#****************
# Model 4: 3-dimensional Rasch model
#****************

# define Q-matrix
Q <- matrix( 0 , nrow=12 , ncol=3 )
Q[ cbind(1:12 , rep(1:3,each=4) ) ] <- 1
rownames(Q) <- colnames(dat)
colnames(Q) <- c("A","B" , "C")

# define nodes
theta.k <- seq(-6,6,len=13 )

# M4a: smirt (in sirt)
mod4a <- smirt(dat,Qmatrix=Q,irtmodel="comp" , theta.k=theta.k , maxiter=30)
summary(mod4a)

# M4b: rasch.mml2 (in sirt)
mod4b <- rasch.mml2(dat,Q=Q,theta.k=theta.k , mmliter=30)
summary(mod4b)

# M4c: gdm (in CDM)
mod4c <- gdm( dat , irtmodel="1PL" , theta.k=theta.k , skillspace="normal" , 
            Qmatrix=Q , maxiter=30 , centered.latent=TRUE )
summary(mod4c)

# M4d: tam.mml (in TAM)
mod4d <- tam.mml( resp=dat , Q=Q , control=list(nodes=theta.k , maxiter=30) )
summary(mod4d)

# M4e: R2noharm (in NOHARM, running from within R using sirt package)
# define noharm.path where 'NoharmCL.exe' is located
noharm.path <- "c:/NOHARM"
# covariance matrix
P.pattern <- matrix( 1 , ncol=3 , nrow=3 )
P.init <- 0.8+0*P.pattern
diag(P.init) <- 1
# loading matrix
F.pattern <- 0*Q
F.init <- Q
# estimate model
mod4e <- R2noharm( dat = dat , model.type="CFA" , F.pattern = F.pattern , 
    F.init = F.init , P.pattern = P.pattern , P.init = P.init , 
    writename = "ex4e" , noharm.path = noharm.path  , dec ="," )
summary(mod4e)

#****************
# Model 5: 3-dimensional 2PL model
#****************

# M5a: smirt (in sirt)
mod5a <- smirt(dat,Qmatrix=Q,irtmodel="comp" , est.a="2PL" , theta.k=theta.k , 
           maxiter=30)
summary(mod5a)

# M5b: rasch.mml2 (in sirt)
mod5b <- rasch.mml2(dat,Q=Q,theta.k=theta.k ,est.a=1:12, mmliter=30)
summary(mod5b)

# M5c: gdm (in CDM)
mod5c <- gdm( dat , irtmodel="2PL" , theta.k=theta.k , skillspace="loglinear" , 
            Qmatrix=Q , maxiter=30 , centered.latent=TRUE ,
            standardized.latent=TRUE)
summary(mod5c)

# M5d: tam.mml (in TAM)
mod5d <- tam.mml.2pl( resp=dat , Q=Q , control=list(nodes=theta.k , maxiter=30) )
summary(mod5d)

# M5e: R2noharm (in NOHARM, running from within R using sirt package)
noharm.path <- "c:/NOHARM"
# covariance matrix
P.pattern <- matrix( 1 , ncol=3 , nrow=3 )
diag(P.pattern) <- 0
P.init <- 0.8+0*P.pattern
diag(P.init) <- 1
# loading matrix
F.pattern <- Q
F.init <- Q
# estimate model
mod5e <- R2noharm( dat = dat , model.type="CFA" , F.pattern = F.pattern , 
    F.init = F.init , P.pattern = P.pattern , P.init = P.init , 
    writename = "ex5e" , noharm.path = noharm.path  , dec ="," )
summary(mod5e)