Day 4 – Using generative AI to create multiple choice question activities
Welcome to Day 4! Today we’ll explore how generative AI tools can support educators in creating high-quality multiple-choice questions (MCQs). We’ll focus on practical strategies for crafting effective prompts to maximize AI’s potential in asynchronous learning design.
Today’s activity was written by Giacomo Occhipinti, Digital Learning Producer at London College Fashion.
Generative AI platforms can quickly create engaging and varied MCQs and can enable educators to design asynchronous activities efficiently while maintaining quality and alignment with learning outcomes.
Imagine this scenario: You’d like to include a short multiple-choice quiz to help students consolidate their understanding of key concepts. However, creating well-designed MCQs can feel daunting, especially if you’re pressed for time or unsure how to write effective descriptions. As a result, you might skip the quiz altogether or create unclear questions which results in missing an opportunity to check learning and provide timely feedback.
Let’s go back: You’ve set up the tools on your VLE, but now you need to write the MCQs to check students’ understanding of a topic. The way these questions and answers are structured—your choice of words, structure, and tone—can make all the difference.
In asynchronous and online teaching scenarios it’s useful to design communications and learning materials with clarity to ensure the expectation, pacing and direction of teaching materials are accessible for the viewer.
A clear, engaging question can inspire participation, while a vague or overly complex one might leave participants uncertain of what should be learning and paying attention to.
Example of MCQs
General Example:
Question: What is AI?
- A. A computer
- B. A robot
- C. A programming language
- D. Artificial intelligence
The question is too broad and lacks context. The correct answer is obvious, and the distractors (incorrect answers) are not plausible.
Concise Example:
Question: Which of the following best describes artificial intelligence (AI)?
- A. A set of techniques that enable machines to mimic human behaviour and decision-making.
- B. A physical robot that interacts with humans.
- C. A software application used only for data entry tasks.
- D. A programming language developed in the 1990s.
The question is specific, and the answers are plausible but clearly incorrect. This tests understanding rather than guessing.
Activity
In this activity, you’ll learn how to use AI tools to create MCQs that are clear, engaging and effective, saving you time while improving the quality of your learning materials.
Here’s what you’ll do:
- Create Multiple-Choice Questions (MCQs)
Choose a topic related to your subject area or interests that you’d like to explore.
- Use the AI to generate MCQs with this prompt: Create 5 multiple-choice questions on [topic] suitable for [beginner/intermediate/advanced] learners. Ensure each question includes 1 correct answer and 3 plausible, but incorrect, answers.
- Evaluate the results: Do the questions align with your learning objectives? Are the distractors plausible and well-written?
- Back up your questions design with some research: You can ask the GenAI bot to benchmark your questions against some guidelines like: Writing Good Multiple Choice Test Questions | Center for Teaching | Vanderbilt University
- Experiment and Improve
Adjust the prompts to see how different phrasing affects the multiple choice questions and answers. For example, if you chose ‘beginner’, change to ‘intermediate’ or advanced’.
If you have time, use multiple AI platforms to compare the quality of their responses.
- Edit and Refine
Customise the AI-generated texts and questions to suit your learners’ needs and the context of your course.
Suggested platforms:
We recommend using Claude (Anthropic), ChatGPT (OpenAI) or Gemini (Google).
These platforms are accessible and offer user-friendly interfaces. They excel in natural language generation, making them ideal for creating educational content.
Alternative:
You can also explore Bing Chat.
Discussion
Join us in the Teams space to share your responses and the MCQs you’ve generated. If you don’t have access to the space, email us at teachingexchange@arts.ac.uk for the attention of Hannah.
- What challenges did you meet when crafting prompts?
- How would you refine the generated content to better suit your objectives?
- Do you see generative AI as a tool to complement, replace or support traditional learning design practices? Why?
- How do you see this process supporting accessibility and inclusivity in asynchronous learning?