Objective: This study aims to evaluate the suitability of the automatic
item generation (AIG) for producing Turkish case-based multiplechoice
questions (MCQs) in psychiatry.
Method: The study was planned as a descriptive study. In the first stage,
topics were determined and a cognitive model was created by subject
matter experts. In the second stage, a question template was created,
variables were determined, the format of answer options was organized,
and two equivalent templates of question content with different
combinations were created. In the final stage, questions were generated
using Python-based software based on these models. Following the
question generation, random samples were selected and evaluated by
experienced educators using a structured form.
Results: A total of 1189 questions were generated, with 11 questions
sampled for each diagnosis. In the evaluation conducted by experts, six
of the questions were deemed appropriate for each parameter, while
minor corrections were suggested for five questions. It was stated that
all the questions assess clinical reasoning skills rather than factual recall.
Conclusion: The template-based AIG method allows for the rapid
and effective production of high-quality questions needed in medical
education. The study demonstrated that AIG in the Turkish language for
generating MCQs that assess clinical reasoning is applicable in the field
of psychiatry. This method enables the production of a large number of
questions in a short time, enriched with various combinations.
Keyword: Automated item generation, clinical reasoning, medical
education, multiple choice question, psychiatry education