Create Tests According to QTI 2.1 Standard

Create tests and tasks compliant with the Question & Test Interoperability (QTI) information model version 2.1. Input sources are Rmd/md description files or S4-class objects. Output formats include standalone zip or xml files. Supports the generation of basic task types (single and multiple choice, order, pair association, matching tables, filling gaps and essay) and provides a comprehensive set of attributes for customizing tests.

rqti

The objective of rqti is to establish a robust and stand-alone R library tailored for crafting exercises and exams in adherence to the QTI v2.1 standard directly from R. Users have the flexibility to render the exercises either locally (using qtijs) or integrate them seamlessly into the OPAL learning management system. Our primary audience comprises instructors specializing in research methods and statistics who seek to harness the full capabilities of R while aligning with the QTI standard. In particular, instructors utilizing OPAL will find our package especially advantageous, as it capitalizes on OPAL’s robust API.

Documentation

Comprehensive documentation, encompassing installation instructions, a quick start guide, example usage, and detailed functionality documentation, can be accessed at the following link: https://shevandrin.github.io/rqti/

Testing

Our commitment to robustness is reflected in our pursuit of high test coverage through automated tests, located in the “tests” folder. Note that the qtijs-server and the Opal API cannot be tested with github actions (continuous integration); local test coverage is around 87%.

Support and Bug Reports

Should you find any missing features or encounter issues, please do not hesitate to inform us via email or open an issue on GitHub. We will offer support until the project’s funding concludes in September 2024. Following that, we will maintain a stable, usable version, with support for new features provided as time permits.

Contribute

To contribute to this package, clone the repository, implement your feature, and submit a pull request. Kindly adhere to the tidyverse style and the general package guidelines outlined at https://r-pkgs.org. Your contributions are greatly appreciated.

The need for rqti / comparison with exams

Why do we need another package for creating exams when there is already the exams-package? The philosophy of exams is one for all, whereas rqti follows the Unix-philosophy: do one thing and do it well. tldr:

• rqti renders QTI files directly in the browser (or the RStudio Viewer pane) from within R, allowing you to see and interact with the exercise before uploading it to the LMS.
• rqti supports more QTI-specific features (e.g., exercise types) than exams.
• Built with S4 object-oriented programming, rqti is easy to extend for LMS-specific needs.
• The Rmd interface is cleaner, focusing exclusively on QTI.
• Low-level functions can be used flexibly to create exercises — something not possible in exams.
• Includes functions for uploading files to an LMS via REST API (implemented for OPAL, but you can adapt it to your own LMS).

If you work exclusively with QTI exercises, you should try our package. If you need additional formats besides QTI — for example, print output or LMS platforms without QTI support — exams remains the better choice.

OpenOlat Docker

For testing purposes, we have set up a Docker container for OpenOlat. Instructions for running it can be found in the openolat-docker folder.