Date
Publisher
arXiv
Controllable text generation (CTG) by large language models has a huge
potential to transform education for teachers and students alike. Specifically,
high quality and diverse question generation can dramatically reduce the load
on teachers and improve the quality of their educational content. Recent work
in this domain has made progress with generation, but fails to show that real
teachers judge the generated questions as sufficiently useful for the classroom
setting; or if instead the questions have errors and/or pedagogically unhelpful
content. We conduct a human evaluation with teachers to assess the quality and
usefulness of outputs from combining CTG and question taxonomies (Bloom's and a
difficulty taxonomy). The results demonstrate that the questions generated are
high quality and sufficiently useful, showing their promise for widespread use
in the classroom setting.
Study at a glance
- Question: Do teachers find classroom questions generated by artificial intelligence useful and high quality?
- Setting and data: Teachers evaluating AI-generated educational questions.
- Method: Teacher evaluation of the quality and usefulness of generated questions (human evaluation).
Why it matters
Large language models can generate questions to reduce the load on teachers. However, recent work fails to show whether real teachers judge these questions as sufficiently useful for classrooms, or if they contain errors and pedagogically unhelpful content.
Key findings
- In evaluations with teachers, large language models produced educational questions that educators judged to be high quality and sufficiently useful for classrooms.
- Combining automated text generation with Bloom's taxonomy and difficulty levels produced questions that met educators' practical instructional needs.
- The generated questions demonstrated sufficient pedagogical quality to show strong promise for widespread use in classroom settings.
Summary drafted with AI from the paper’s abstract and may contain errors. See the full paper for details.
What is the application?
Who is the user?
Who age?
Why use AI?
Study design
