irt.prob-class: Class "irt.prob"

In plink: IRT Separate Calibration Linking Methods

Description

The formal S4 class for irt.prob. This class contains the expected probabilities of responding in a given category for a set of items and theta values under the specified IRT models. The class also includes characteristics of the items.

Objects from the Class

Objects can be created by calls of the form new("irt.prob", ...), but this is not encouraged. Use one of the functions drm, gpcm, grm, mcm, nrm, or mixed instead.

Slots

prob:

data.frame of item probabilities with n rows and j+m columns for n theta values and j items (the first m column contains theta values for m dimensions)

info:

data.frame of item information

p.cat:

vector identifying the number of categories for each item for which probabilities were computed

mod.lab:

character vector of labels for the model(s).

dimensions:

numeric value identifying the number of modeled dimensions

D:

numeric vector identifying scaling constants for drm, grm, and gpcm

pars:

list of the item parameters used to compute the probabilities

model:

character vector identifying all the models used to compute the probabilities in prob. The only acceptable models are drm, gpcm, grm, mcm, and nrm (see class poly.mod for more information).

items:

list with the same length as model, where each element identifies the items associated with the model(s) specified in model.

Extends

Class poly.mod, directly.
Class list.poly, by class poly.mod, distance 2.

Author(s)

Jonathan P. Weeks [e-mail]

See Also

These models provide information on both unidimensional and multidimensional formulations
drm: for dichotomous response models (1PL, 2PL, and 3PL)
gpcm: for the partial credit/generalized partial credit model
grm: for the graded response model
mcm: for the multiple-choice model
nrm: for the nominal response model
mixed: for mixed-format items
irt.pars

plink documentation built on May 1, 2019, 8:07 p.m.