create_prob_array

polytomous IRT model. Options: <code>"GRM"</code> for Graded Response
Model and <code>"NRM"</code> for Nominal Response Model

model

matrix of the item bank. Rows represent items, and columns
represent parameters. If the model is <code>"GRM"</code>, the first column
represents the <code>alpha</code> parameters and the next columns represent the
<code>beta</code> parameters. If the model is <code>"NRM"</code>, odd columns represent
the <code>alpha</code> parameters and even columns represent <code>beta</code>
parameters

bank

vector of number of possible responses for every item

nres

For every item (dim 1) in an item bank and every evaluated ability level
(dim 2), computes the probability of picking every possible response (dim 3)
given the ability level

Implements the Merged Tree-CAT method (Javier Rodriguez-Cuadrado et al., 2020, <doi:10.1016/j.eswa.2019.113066>) to generate Computerized Adaptive Tests (CATs) based on a decision tree. The tree growth is controlled by merging branches with similar trait distributions and estimations. This package has the necessary tools for creating CATs and estimate the subject's ability level.

create_prob_array: Multidimensional array of response probabilities

Description

Usage

Arguments

Value