Create Constraints for Small Test Assembly Problems.

eatATA

Overview

eatATA provides a small R interface to mathematical optimization solvers specialized on solving simple automated test assembly problems (ATA). Internally, sparse matrices are used via the Matrix package. Currently supported solvers are GLPK, lpSolve, Symphony, and Gurobi. See below for a list of implemented features and feature to come.

Installation

# Install eatATA from CRAN via
install.packages("eatATA")

# Install development version from GitHub via
remotes::install_github("beckerbenj/eatATA", build_vignettes = TRUE, dependencies = TRUE)

Documentation

An extensive tutorial paper including a variety of use cases has been published here: https://doi.org/10.3390/psych3020010. Additionally, a set of vignettes describing the current functionality can be found on CRAN.

Implemented Features

eatATA is suitable for the automated test assembly of fixed linear test forms or multi-stage testing modules. For example, the following types of constraints can be set via eatATA:

• no item overlap between test forms

• complete item pool depletion

• categorical and numerical constraints across test forms

• excluding items from being together in the same booklet (item exclusions)

• force items to be in the same test form (item inclusions)

• force a set of items to be included in the test forms

• various optimization constraints

Outlook

Features that might be implemented in the future:

• taking into account hierarchical stimulus-item structures

• adopting the ROI framework to access a larger number of solvers

If you wish to contribute to the package, please send an email to [email protected].