Creating 'Workshop Wizard'

I have previously written about the GPTs I have created to give feedback on lectures and that I use for feedback and editing of these blog posts. In this post I will discuss my GPT for designing workshop sessions.

The term ‘GPT’ is here used for the specific AI applications that can be created by ChatGPT4. These allow you to effectively give a structured prompt that outs the main model to work on a specific problem. This can be combined with other background data that you uniquely own to create something useful. For now these are accessible only to those with a paid GPT4 account (though this may be about to change with the release of the new ChatGPT model), but I will give you the full prompt at the end which could be used in other AI models.

The aim

The ‘Workshop Wizard’ is designed to help with the process of designing an academic workshop. It will produce a broad plan of activities for the session, this can include ideas on pre and post workshop activities, ideas for specific software or tools that can be used, and ideas on what might go wrong in the session and how to overcome these potential problems. 

The GPT will begin by asking various details about your plans for the session, such as length, topic, learning outcomes and any assessment to be used, as well as the level of the students and length of the session 

The reason for including the topic, session length and level of the students are fairly obvious. Including specific learning outcomes focuses the plan for the session as part of the wider module and programme of study. Similarly I am a firm believer that sessions should be planned with the assessment being used in the module in mind, so knowing this will help in the planning. A session for students who soon have to give an assessed presentation should be different to one where students soon have to take an exam. 

However you could leave these details out if you were either unsure or wanted the widest possible variety of ideas, without being limited. You could bring in your ideas later in the conversation or could ask for multiple plans and choose amongst them. 

An important aspect is that the GPT will specify any problems you may face and how you might deal with these. This is something that I find myself rarely doing, but this pre-planning for potential failure or difficulties can help you adapt as the session develops. This will be particularly useful for longer sessions such as a three-hour long workshop. 

I have designed the GPT to offer a variety of tasks, this is particularly important at De Montfort University (DMU) where I work, as our main teaching mode is a three-hour long workshop. To maintain interest and focus of the students for this amount of time requires a variety of tasks, just doing economics problems sets for three hours will feel pretty tortuous, pretty quickly!

The GPT will also specify relevant work that could contribute to the session, either as pre-work or post-work. Most student work is done outside of the classroom, and a clear link to other activities will be very helpful for wider planning.

The final element of the GPT is that I have attached two examples of workshop plans that I have developed in the first to guide colleagues working with me on my courses. These plans set out step-by-step instructions for how the session would run, the aims of each activity, and a time guide for roughly how long each session would take. My colleagues found these very helpful, and I consider them a good model for planning such sessions, so I have asked the GPT to use these as a templated for its responses.

I developed this GPT from my own reflection and from conversations with colleagues at DMU early in 2024 as myself and Dr Swati Virmani gave our own workshop on how to use AI in teaching. One of our activities for the session was to compare the workshop development process with and without AI, and this helped provide ideas for exactly what it was AI could do to help in the process.

In particular the idea of how you might deal with problems is a novel innovation in my experience. We may often want to try something new or interesting but be put off by possible technological problems, or lack of student engagement or other potential issues. Having a clear idea of what these problems might be and how we might deal with them can lead to less risk-aversion, and AI can speed up this process greatly.

A Demonstration

With all that in mind, let’s walk through a conversation to see the results and assess the good and bad, and see if we need to make any improvements. I will open with a fairly terse prompt inspired by what I am actually teaching this week:

I wish to plan a workshop introducing key concepts about the economics of international trade, help me design it.

To which the reply is:

To design a workshop focused on the economics of international trade, it’s important to consider how best to introduce and explore its key concepts effectively. Let’s start by defining some details for the workshop:

1. Subject & Topics: Introduction to the Economics of International Trade

2. Student Level: (Specify whether these are undergraduate, graduate, or other.)

3. Session Length: (How long will the session be?)

4. Learning Outcomes: (What do you want students to learn from this session? Examples might include understanding comparative advantage, the impact of tariffs and trade barriers, and the role of international trade organizations.)

5. Assessment Types: (How will you assess learning? This could be through quizzes, group presentations, etc.)

Based on general details, here's a draft plan that could be adapted to your specifics:

Workshop Plan: Economics of International Trade

Introduction (10 minutes)

• Briefly introduce the workshop, outline the session's objectives, and emphasize the importance of the pre-assigned readings or videos on international trade basics.

Ice-Breaker Activity: Global Trade Insights (10 minutes)

• Have students introduce themselves and mention an interesting fact or news they recently read about international trade. This can be structured as a quick roundtable discussion to set the stage for interaction and relevance.

Main Content Delivery: Interactive Lecture (30 minutes)

• Use a combination of short lectures and interactive polls or questions to cover the basic concepts:

  • Comparative Advantage

  • Trade Barriers and Tariffs

  • Trade Agreements and International Trade Organizations

Case Study Breakouts (30 minutes)

• Divide students into small groups. Each group analyzes a different real-world trade scenario (e.g., NAFTA, EU, US-China Trade War) discussing the involved economic principles, winners, and losers.

• Each group will prepare a quick presentation of their findings to encourage information sharing and peer learning.

Simulation Game: Trading Game (20 minutes)

• Conduct a trading game where each student or group represents a country with specific resources and needs. This activity helps illustrate the benefits and complexities of trade negotiations.

• Discuss the outcomes, focusing on negotiation strategies and economic impacts.

Q&A and Group Discussion (20 minutes)

• Open the floor for questions and a guided discussion on the implications of international trade on global economies and individual countries.

• Encourage students to relate the discussion to current events and policy debates.

Wrap-Up and Feedback (10 minutes)

• Summarize key learning points of the workshop.

• Gather feedback on the session to improve future workshops.

Follow-Up Work

• Assign a follow-up essay or project where students analyze a specific trade agreement and its impact on the member countries.

Assessment

• Prepare a quiz based on the session content or evaluate the presentations/group discussions for understanding and analytical thinking.

Would this structure meet your needs? If you have specific topics, duration, or outcomes in mind, let me know, and we can refine this further!

The response has clearly set out the thing I want to focus on at the start, the level of the students, the length of the session, the learning outcomes and the module assessment should all be accounted for. The specific suggestions do provide variety, and without having lots of detail, there are some good starting ideas here.

The output is also generated in the style I want, simple bullet points with clear defined sections and an estimated time for each part.

I want to add some detail though:

The workshop will actually be 3 hours long, and will be for first year undergraduate students who are being introduced to these ideas for the very first time. A major learning objective will be for them to have a sound grasp of the concept of comparative advantage, so this should be a key focus. Please modify the plans accordingly.

To which the response is:

Great, thanks for the additional details! Focusing on first-year undergraduates and emphasizing the concept of comparative advantage, we can adjust the workshop plan to make it more accessible and engaging for beginners. Here’s a revised version of the plan for a three-hour session:

Revised Workshop Plan: Introduction to the Economics of International Trade

Introduction (15 minutes)

• Welcome and introduction to the workshop.

• Briefly explain the importance and relevance of international trade in the global economy.

• Outline the goals for the session, emphasizing the key concept of comparative advantage.

Ice-Breaker Activity: Global Connection (15 minutes)

• Have students form small groups and discuss one item they commonly use (e.g., smartphones, clothing) and guess which countries these items might have come from. This will set a tangible context for discussing international trade.

Interactive Lecture Part 1: Basics of International Trade (30 minutes)

• Start with an engaging multimedia presentation to introduce basic terms and concepts like imports, exports, tariffs, and trade agreements.

• Use visuals and simple examples to explain how countries benefit from trade.

Deep Dive: Comparative Advantage (45 minutes)

• Explain the theory of comparative advantage with clear, step-by-step examples. Use a two-country, two-product model to illustrate how countries can benefit from specializing in goods where they have a comparative advantage.

• Interactive element: Use a classroom poll or quiz to check understanding as you progress through the examples.

Group Activity: Simulating Trade (45 minutes)

• Break the class into groups, assigning each a country with certain production costs for two different products.

• Task each group with determining whether to produce both goods or specialize based on their comparative advantages.

• Groups will then "trade" with each other to maximize their benefits. After the activity, discuss as a class what decisions were made and why.

Case Study and Discussion (30 minutes)

• Present a case study involving a real-world trade scenario (e.g., agricultural products between two countries).

• In groups, students analyze the trade dynamics and identify which country has a comparative advantage in what product.

• Reconvene to discuss the groups' findings and how the theory applied or didn’t apply in real-world terms.

Wrap-Up and Reflective Discussion (20 minutes)

• Summarize the key points learned about comparative advantage and international trade.

• Open the floor for students to ask questions or express any confusion about the concepts discussed.

• Provide a brief overview of what will be covered in the next session and assign a simple reading on international trade effects on local economies for deeper engagement.

Assessment and Feedback (10 minutes)

• Conduct a quick quiz to assess understanding of comparative advantage.

• Request feedback on the workshop’s format and content to improve future sessions.

This revised plan should cater well to the needs of first-year undergraduate students by providing interactive and practical learning experiences focused on comparative advantage within the broader context of international trade. Does this approach align with what you had in mind?

Did this better account for time? It is longer, but notice it has the maths wrong, the time adds to more than three hours! This is something that can be easily adapted to, but is worth keeping in mind (current LLMs are know to get simple maths wrong as it is not what they are designed for).

Was this better suited for the level of the students? Somewhat yes, the first few activities are designed to clearly explain the subject, whilst keeping it relevant. What is interesting is that several of these mirror what I am already planning to do. I do have a video and an activity explaining the basics of comparative advantage, I do have an activity looking at every day items and where they are produced, and we will be playing a game where teams compete according to their comparative advantage.

Overall I am happy that the GPT is providing a good variety of ideas for a workshop, that it sets out these plans in a way that is helpful, and responds well to extra information to modify the plans (this is based on other interactions as well as the one shown here).

One thing it is missing is the element of possible problems and how to be ready for them. I will use the editing feature to emphasise this.

Editing

The editor asks me what the GPT could be doing better, I tell it:

I would like to emphasise more strongly what can go wrong in a specific plan. At the end of every outline please include a short passage on possible problems the user may encounter and contingencies for these problems.

Trying the original prompt again, the results were not much improved, there were some extra suggestions at the end on how to prepare for the session, but not the troubleshooting that I wanted. Therefore to make it even more explicit I prepared a short example of what I would want the problem-solving to look like, and asked the editor to use this as the basis. Now trying again I get at the end of the plan this element:

Potential Issues and Solutions

• Problem: Students may struggle with the abstraction of comparative advantage.

• Solution: Use concrete examples in the trading game and case studies to illustrate the concepts clearly.

• Problem: Difficulty in applying the theory to current events due to complexity of real-world trade dynamics.

• Solution: Provide guiding questions that help frame their research and analysis, focusing on comparative advantage.

I would like to see some more detail here perhaps, but the key to a good result is to have a conversation with the AI, so having this here as a starter can prompt further conversations.

In conclusion, like a lot of output from current LLMs, I think you are unlikely to be able to generate an entire workshop set of resources from this process, or at least not to the standard you should be aiming for. Our experience as academics and the knowledge of our own students are valuable and cannot be exactly replicated by an AI (for now).

However I think this is a great tool for generating initial plans, making comparison to existing plans or refining ideas you already have. The principles of variety in tasks, of a focus on learning outcomes and keeping assessment methods in mind whilst considering the level of the students and the topic at hand are fundamental, and having some idea of possible problems and how to respond to them will help a lot.

The Prompt

This is the background outline of the GPT that you could use as the basis for a prompt in other LLMs to achieve similar workshop planning results, or use the GPT directly here. Please give it a go and let me know how you get on. There are many other things that could be included in plans and that you may consider important in order to design a good session, what do you suggest?

Role and Goal: This GPT, named Workshop Wizard, is a learning design assistant tailored for academics teaching in universities. It specializes in helping design workshop activity plans for any subject, with a focus on innovative and effective learning strategies to engage students. At the beginning of conversations, it always inquires about the workshop's subject, student level, specific topics, session length, targeted learning outcomes, and types of assessments to tailor its advice accurately.

Constraints: It avoids recommending activities that require extensive resources or are unfeasible in an academic setting and steers clear of suggesting unproven practices.

Guidelines: The GPT encourages the incorporation of a variety of activities such as quizzes, group work, research activities, games, and problem sets into the workshop plans. It suggests including pre-work and follow-up work to reinforce learning, outlines potential issues that might arise during the workshop, and provides solutions.

Clarification: When faced with vague information, it uses its best judgment to advise that more detailed inputs lead to better recommendations. It suggests specific digital tools using internet searches when appropriate, noting that other options are available. It will confirm details with the user before finalizing instructions, ensuring they are comprehensive and aligned with academic standards.

Personalization: Maintains a supportive tone throughout, adapting its suggestions to the specific context of the workshop. It focuses on a collaborative and creative planning process, providing examples based on uploaded workshop plans to ensure the output is practical and adheres to academic standards.

Each workshop plan is concluded with a section detailing potential issues that could arise and how to develop contingency plans, using examples from relevant documents to illustrate effective problem-solving strategies.

Addendum

I am writing this as they hype about the new ChatGPT model is raging. More to follow on this soon!