rpf.drm

numChoices

dimensions

defaults to 1 or pass in
  <code><a href="/link/rpf.ogive?package=rpf&version=0.3" rd-options="" data-mini-rdoc="rpf::rpf.ogive">rpf.ogive</a></code>

whether to use a multidimensional
  model. Defaults to <code>TRUE</code> when <code>dimensions>1</code>
  and <code>FALSE</code> when <code>dimensions==1</code>.

multidimensional

This function instantiates a dichotomous response model.
  The discrimination prior defaults to the lognormal
  distribution with <code>meanlog=0</code> and <code>sdlog=.5</code>.
  The guessing prior is the beta distribution. See the
  source code for details.  For discussion on the choice of
  these Bayesian priors see Baker & Kim (2004, pp.
  187-188).

The purpose of this package is to factor out logic and
        math common to Item Response Theory fitting, diagnostics, and
        analysis. It is envisioned as core support code suitable for
        more specialized IRT packages to build upon. A version 1.0 will
        offer optimized, low-level functions to map parameters to
        response probabilities for a wide variety of dichotomous and
        polytomous item models.

rpf.drm: Create a dichotomous response model and associated hyperparameters.

Description

Usage

Arguments

Value

Details

References