Statistics, Department of
The R Journal
Date of this Version
12-2021
Document Type
Article
Citation
The R Journal (December 2021) 13(2); Editor: Dianne Cook
Abstract
This article introduces the cat.dt package for the creation of Computerized Adaptive Tests (CATs). Unlike existing packages, the cat.dt package represents the CAT in a Decision Tree (DT) structure. This allows building the test before its administration, ensuring that the creation time of the test is independent of the number of participants. Moreover, to accelerate the construction of the tree, the package controls its growth by joining nodes with similar estimations or distributions of the ability level and uses techniques such as message passing and pre-calculations. The constructed tree, as well as the estimation procedure, can be visualized using the graphical tools included in the package. An experiment designed to evaluate its performance shows that the cat.dt package drastically reduces computational time in the creation of CATs without compromising accuracy
Included in
Numerical Analysis and Scientific Computing Commons, Programming Languages and Compilers Commons
Comments
Copyright 2021, The R Foundation. Open access material. License: CC BY 4.0 International