Curious about how AI can help plan, design and evaluate learning experiences? This AI-Enhanced Learning Design series will look at how AI tools can augment key tasks in the learning design process, starting with Analysis.
Enhancing Analysis with AI
The ADDIE model is a useful framework outlining the key stages in instructional design: Analysis, Design, Development, Implementation, and Evaluation. The first stage is Analysis, where you investigate what needs to be taught, to whom and under what circumstances. AI can help with the key tasks in this stage including research, learner needs analysis and understanding the organisational environment — we'll explore these below.
Profiling learner needs
During the analysis stage, it’s important to gauge the learners' existing knowledge and skills, and identify the gaps between where they are and where they need to be. This ensures that learning interventions are relevant and address specific areas where learners require support.
AI can evaluate responses from an array of activities, comparing them against benchmarks to generate comprehensive reports. This AI-powered analysis allows for a more comprehensive understanding of each learner's capabilities. Armed with this knowledge, learning designers can craft personalised learning paths and experiences that cater to the needs of all learners.
Some e-learning platforms, like 360Learning, analyse learner data to offer insights, predict future learning and personalise recommendations. Other specialised AI tools, such as iMocha, offer assessments tailored to job requirements, measuring learners' current knowledge levels and pinpointing skill gaps.
If you don't have access to such platforms or tools, you can upload data, such as pre-assessment results, to ChatGPT-4 and ask it to perform various analyses. This is particularly useful for qualitative or subjective responses, which may be slow and difficult to analyse without AI assistance. Below is a simple example of using ChatGPT-4 to analyse a hypothetical pre-assessment dataset — select any image to make it full screen.
ChatGPT's response was:
ChatGPT can also support learner personas and empathy mapping. It can help you write interview or survey questions and highlight potential gaps in your personas. When given a particular framework or approach, it can guide you through an empathy mapping process, asking follow-up questions and encouraging deeper analysis. Although it can’t inherently empathise, it can play the role of a persona to brainstorm or explore ideas. This can then be built upon during the design stage, where relevant examples and authentic scenarios are needed.
Evaluating organisational environments
Understanding the organisational learning environment is pivotal for tailoring interventions that align with company culture, learner habits and other learning and development (L & D) initiatives. This helps ensure learning programs are well-received and have a real, long-term impact on the learner and organisation. AI tools, like ChatGPT, can simplify this investigation.
After collecting relevant internal documents, employee feedback and data, you can upload it to ChatGPT-4 to analyse trends, sentiments and possible barriers that exist in the organisation. Here is an example scenario of using ChatGPT to analyse organisational context — select the arrows to reveal (or hide) each step.
Step 1: Collect information
Gather internal documents, such as previous training evaluations, relevant processes and employee satisfaction surveys.
Step 2: Upload and analyse with ChatGPT
Step 3: Findings summary
Step 4: Recommendations
Making research simpler with AI
You’ve probably been researching with AI for years when using search engines like Google, but the traditional search experience is changing. Instead of navigating numerous webpages or only being able to ask basic queries, AI tools like ChatGPT and Perplexity can integrate search into your workflow and dig deeper for richer or more precise results.
ChatGPT-4 offers a quick and easy way to research, especially if you’re already using it in your learning design. You can ask it for strategies, research or case studies relevant to your project. It also has access to the internet, which could help you identify particular details about an organisation. Furthermore, the GPT Store has thousands of GPTs designed for various purposes (or you can make your own) and there’s an entire section dedicated to research and analysis — learn more about GPTs here.
Perplexity has an intuitive search experience and is making waves as a competitor to Google. It can make research easier by providing straightforward, relevant responses without the clutter of traditional search results. For example, you can change the search ‘focus’ and even attach images and pdfs.
There is a Copilot feature, which asks follow-up questions, summarises the most relevant findings and pulls from a diverse range of sources for more personalised, comprehensive results.
You can use the basic features of Perplexity for free. The premium version costs $20 per month or $200 per year and includes benefits such as unlimited Copilot usage, more file uploads, access to GPT-4 and dedicated support. Visit Perplexity’s getting started page for more information, tips and examples.
Considering AI’s limitations in learning analysis
AI can quickly analyse an array of data, giving you valuable insights for learning design. Yet AI alone may not paint a complete picture and could overlook subtleties in organisational culture and employee sentiments. For example, AI analyses of surveys, documents or assessments may not identify informal learning channels like social media groups or coffee break conversations. AI tools may also lack criticality, so your professional expertise is required. For example, when asked for ideas, ChatGPT often suggests more common and traditional interventions, such as training courses.
When researching, AI tools can save time and give comprehensive and targeted results. Yet, they may not uncover insights that aren’t documented or that they don't have access to. So, although AI can be a great assistant and collaborator, it doesn’t replace the need for further investigation and discussions with key stakeholders and learners.
You should also be aware of how data is used by AI tools and consider potential privacy risks. These risks, as well as other ethical considerations, are discussed in more detail in my blog post: Responsible Learning Design with AI.
Although AI tools are advancing quickly, the challenges above highlight the importance of integrating AI's capabilities with your learning design expertise. Nevertheless, AI offers clear benefits when analysing data, information and research to inform your learning design. So, it helps to understand how AI tools work, including their limitations, and how they can enhance your learning design.
ChatGPT was used to write some sections of this article and to generate examples — all examples are hypothetical and not based on real data.
For the next installments of the AI-enhanced Learning Design series, subscribe to this blog by selecting the button in the footer below.
Follow me on LinkedIn for more tips and insights on using AI in learning design.
Found this post helpful? Please share it with your network.
Comments