rpf (version 0.28)

rpf.sample: Randomly sample response patterns given a list of items

Description

Returns a random sample of response patterns given a list of item models and parameters.

Usage

rpf.sample(theta, items, params, design, prefix = "i", mean = NULL,
  cov = NULL)

Arguments

theta

either a vector (for 1 dimension) or a matrix (for >1 dimension) of person abilities or the number of response patterns to generate randomly

items

a list of item models

params

a list or matrix of item parameters. If omitted, random item parameters are generated for each item model.

design

a matrix assigning person abilities to item factors

prefix

Column names are taken from param or items. If no column names are available, some will be generated using the given prefix.

mean

mean vector of latent distribution (optional)

cov

covariance matrix of latent distribution (optional)

Value

• Returns a data frame of response patterns

Details

The design matrix can accomodate more person abilities than item dimension. Refer to Cai (2010) for design matrix examples.

TODO: Add restrictions to design matrix to match restrictions imposed by Cai (2010).

References

Cai, L. (2010). A two-tier full-information item factor analysis model with applications. Psychometrika, 75, 581-612.

See Also

sample

Examples

# 1 dimensional items
i1 <- rpf.drm()
i1.p <- rpf.rparam(i1)
i2 <- rpf.nrm(outcomes=3)
i2.p <- rpf.rparam(i2)
rpf.sample(5, list(i1,i2), list(i1.p, i2.p))

# multidimensional items
numItems <- 4
items <- vector("list", numItems)
correct <- vector("list", numItems)

i1 <- rpf.drm(factors=2)
i2 <- rpf.drm(factors=1, multidimensional=TRUE)

for (ix in 1:(numItems-1)) {
  items[[ix]] <- i1
  correct[[ix]] <- rpf.rparam(i1)
}
items[[4]] <- i2
correct[[4]] <- rpf.rparam(i2)

design <- matrix(c(1, 1, 1, 1,
                   2, 2, 3, NA), nrow=2, byrow=TRUE)
rpf.sample(10, items, correct, design)