Automatic Item Generation

description13 papers

group3 followers

lightbulbAbout this topic

Automatic Item Generation (AIG) is a process in educational assessment that utilizes algorithms and computational techniques to create test items or questions automatically. This approach aims to enhance the efficiency, scalability, and diversity of assessment materials while maintaining alignment with learning objectives and content standards.

lightbulbAbout this topic

by Tahani Alsubait

—Several attempts have already been made to automate the generation of assessment questions. These attempts were mainly technical and lacked a theoretical backing. We explore psychological and educational theories to support the... more

descriptionView Paper arrow_downwardDownload

Ontology-Based Multiple Choice Question Generation

by Tahani Alsubait

Multiple choice questions (MCQs) are considered highly useful (being easy to take or mark) but quite difficult to create and large numbers are needed to form valid exams and associated practice materials. The idea of re-using an existing... more

Fig. 1 Quality of questions according to reviewers’ evaluations A question is too difficult for a particular group of stu- dents if it is answered correctly by less tha n 30 % of the students and is too easy if answered by more than 90 % of the students [9]. In both cases, the question needs to be reviewed and improved. Accordingly, we consider a question to be difficult if it is answered 30-60 % and easy if it is answered correctl correctly by y by 60-90 % of the students. With regard to the six questions delivered on paper, two questions were reasonably di were reasonably easy for the students. These were in line with difficulty estimations by t fficult and two four questions he MCQ Gen- erator tool. One out of the six questions was too difficult for the students. Remarkably, the tool and the t hree reviewers

Fig. 2. Tool-reviewers’ agreements on difficulty predictions

descriptionView Paper arrow_downwardDownload