It’s easy to become fixated by the hype of AI and technical jargon and forget about the day-to-day benefits it can bring.
My reminder came recently, speaking to a primary school teacher who had successfully reduced her workload by three hours a week. She had found preparing for reasoning lessons the least enjoyable aspect of her job. Hours were spent having to source a number of different problem sets for the wide range of abilities within her class. Technology enabled her to, instead, outsource all reasoning lesson and homework preparation to an AIpowered online learning platform. Three hours may not seem like a lot, but it was significant for her.
AI may not have swept through education at quite the same speed as in other industries, but AI-powered adaptive learning is having a meaningful impact on primary school teachers’ workload and students’ academic progress.
Adaptive learning
Traditionally, adaptive learning engines relied upon teachers to grade the difficulty of content and then feed easier or harder content depending on how students performed. However, this method was proven not to be effective; humans are unable to grade difficulty levels as well as computers, missing out on nuances such as poor question wording.
Modern adaptive learning systems use AI to grade questions by analysing how previous students have answered them. Such approaches are all the more sophisticated because they can determine relationships between content from seemingly different topics.
For example, the AI can create a highly adaptive lesson in percentages based on a student’s completion of work in operations. For teachers, it helps to introduce a new topic at the right level for each student, ensuring the student remains engaged and does not become demotivated before they’ve spent time becoming acquainted with the new subject matter.
Instilling confidence
Very sophisticated adaptive learning machines go one step further, not only basing content on the likelihood of a student answering a question correctly, but adapting the probability based on the user’s engagement profile.
For example, a student who has been revising for five minutes is going to have a different tolerance to challenging content to a student who has already been revising for 40 minutes. In practice this means a teacher starts a lesson with students answering 75 per cent of questions correctly and this percentage increases over time so that after 25 minutes, the students have an 85 per cent chance of answering questions correctly.
Why is this important? It gives students a sense of progress, which is fundamental to instilling confidence, and maximising their long term engagement.
Very sophisticated systems can also determine what topic a student should sit next in order to best improve their grade, ultimately providing the student with their own personalised learning path.
Rather than using a fixed learning path in maths for all students in the class, teachers could create flexible learning paths personalised to each student according to their pre-existing subject knowledge, strengths and weaknesses.
The benefits of AI-powered adaptive learning are easy to measure with significant time savings for primary school teachers. It won’t be long before they create a lesson personalised to each student in their class in a matter of minutes, which will be a hugely improved learning experience for students.
