npar.CDM {ACTCD} | R Documentation |
This function is used to classify examinees into labeled classes given responses and the Q-matrix.
npar.CDM(Y, Q, cluster.method = c("HACA", "Kmeans"), Kmeans.centers = NULL,
Kmeans.itermax = 10, Kmeans.nstart = 1, HACA.link = c("complete", "ward", "single",
"average", "mcquitty", "median", "centroid"), HACA.cut = NULL, label.method =
c("2b", "2a", "1", "3"),perm=NULL)
Y |
A required |
Q |
A required |
cluster.method |
The cluster algorithm used to classify data. Two options are available, including |
Kmeans.centers |
The number of clusters when |
Kmeans.itermax |
The maximum number of iterations allowed when |
Kmeans.nstart |
The number of random sets to be chosen when |
HACA.link |
The link to be used with HACA. It must be one of |
HACA.cut |
The number of clusters when |
label.method |
The algorithm used for labeling. It should be one of "1","2a", "2b" and "3" corresponding to different labeling methods in Chiu and Ma (2013). The default is "2b". See |
perm |
The data matrix of the partial orders of the attribute patterns. |
att.pattern |
A |
att.dist |
A |
cluster.size |
A set of integers, indicating the sizes of latent clusters. |
cluster.class |
A vector of estimated memberships for examinees. |
print.npar.CDM
, cd.cluster
,labeling
# Classification based on the simulated data and Q matrix
data(sim.dat)
data(sim.Q)
# Information about the dataset
N <- nrow(sim.dat) #number of examinees
J <- nrow(sim.Q) #number of items
K <- ncol(sim.Q) #number of attributes
# Compare the difference in results among different labeling methods
# Note that the default cluster method is HACA
labeled.obj.2a <- npar.CDM(sim.dat, sim.Q, label.method="2a")
labeled.obj.2b <- npar.CDM(sim.dat, sim.Q, label.method="2b")
labeled.obj.3 <- npar.CDM(sim.dat, sim.Q, label.method="3")
data(perm3)
labeled.obj.1 <- npar.CDM(sim.dat, sim.Q, label.method="1",perm=perm3)
remove(perm3)
#User-specified number of latent clusters
M <- 5
labeled.obj.2b <- npar.CDM(sim.dat, sim.Q, cluster.method="HACA",
HACA.cut=M, label.method="2b")
labeled.obj.2a <- npar.CDM(sim.dat, sim.Q, cluster.method="HACA",
HACA.cut=M, label.method="2a")
#The attribute pattern for each examinee
attpatt <- labeled.obj.2b$att.pattern