The webapp was last updated November 2025. Additionally, the webapp uses a slightly different and more up-to-date version of SAQUET, if you would like access, just shoot me an email!
SAQUET provides an automatic method to apply the 19-criteria Item Writing Flaws for evaluating the quality of multiple-choice questions (MCQs)!
This toolkit aids in assessing MCQ quality effectively.
01/21/2025: Updated around half the criteria to follow a ratings-based approach, that now asks for a score 1-10 when the LLM verification happens and takes the average of three scores. Also uses gpt-4o currently.
For more details, please refer to these two papers:
"An Automatic Question Usability Evaluation Toolkit"
"Assessing the Quality of Multiple-Choice Questions Using GPT-4 and Rule-Based Methods"
- Steven James Moore
- Gilles Chen