Beyond Multiple Choice: Embracing a New Era in Assessment

As the education sector embraces AI, personalisation has taken centre stage. The idea of a one-size-fits-all approach is becoming increasingly outdated, not only in the school sector but also in higher education and beyond, as we recognise that individuals learn differently, at their own pace, leveraging their own strengths. Therefore it’s a no-brainer that personalised learning journeys are essential for maximising students' potential and ensuring their academic success.

In fact, there is enough research evidence to demonstrate the power of such personalised learning - tutoring. In 1984, Benjamin Bloom's ground-breaking research unveiled the power of one-on-one tutoring. His study demonstrated that students receiving individual attention outperformed peers a staggering two grade levels. These students also demonstrated greater engagement and time spent in learning. Bloom's "two sigma problem" challenged educators to replicate these results in classroom settings, a quest that continues till today.

And, at sAInaptic, we are always thinking about what the foundation of this personalisation should be.

Our opinion is that personalisation based on performance insights obtained from the marking of descriptive student work is more valuable than performance on multiple-choice quizzes.

For example, using AI to personalise learning based on application & consolidation of knowledge, rather than simple recall of facts, will truly personalise how a learner progresses through their learning. These insights will also provide assessors with a much deeper understanding of their learners’ subject knowledge.

Personalisation using Multiple Choice Quizzes has several limitations

For decades, since digital assessments became a thing, personalisation of learning has relied on automation in assessments, which could only be achieved with multiple-choice quizzes. While they have served a purpose, they do have significant limitations when it comes to creating a truly personalised learning experience.

1. Lack of Depth: Multiple choice quizzes only test the recall of facts and do not capture the depth of a student's understanding or their critical thinking skills.
2. Limited Insights: Performance on multiple-choice quizzes primarily reflects a student's ability to recall facts. It doesn't provide insight into their analytical, problem-solving abilities, writing skills or consolidation of knowledge.
3. Redundant Approach: These quizzes often use a common set of questions, which become repetitive and may not align with the unique learning needs and goals of individual students.

Most importantly, short answer quizzes remove the possibility of students getting answers correct, purely by chance! After all, one has a 25% chance of selecting the correct answer, from 4 choices in any multiple-choice question.

AI-Assisted Marking of Descriptive Student Work: Breaking the Norm

In contrast, AI-assisted marking of descriptive student work offers a more comprehensive and insightful approach to assessing students' performance and tailoring their learning pathways.

1. Deep Understanding: Open-ended questions look for deeper skills such as the application of knowledge. With AI marking these responses, educators can not only automate the marking of such questions but also quickly identify the quality of their explanations and the coherence of their arguments.
2. Identifying Gaps in Learning: AI can not only pinpoint specific gaps in a student's knowledge by recognising concepts that are missing or inaccurately addressed in their responses but also help identify deeper issues such as problematic writing or mathematical skills. This level of targeted feedback is invaluable for personalising learning.
3. Better Consolidation of Knowledge: Personalisation based on AI-assisted marking can recommend topics that are directly relevant to a student's current level of understanding. This ensures that they build a solid foundation before moving on to more complex concepts.
4. Promotes Critical Thinking: Assessing descriptive student work challenges students to think critically, apply their knowledge, and communicate their ideas effectively, all of which are essential skills for success in the real world.

The Educational Impact of AI-Assisted Marking of Short Answer Assessments

1. Effective Personalisation: Using AI to assess descriptive work allows educators to tailor their teaching and assessments to an individual's strengths and weaknesses, not just factual knowledge, ensuring they receive content at the right level of challenge.
2. Increased Engagement: Engaging with descriptive tasks encourages learners to actively think and communicate their ideas, leading to deeper comprehension and engagement with the subject matter.
3. Targeted Remediation: Identifying gaps in knowledge allows educators to provide targeted support or additional resources to help learners overcome specific challenges, right from the start.
4. Holistic Skill Development: AI-assisted marking fosters the development of not only subject-specific knowledge but also critical thinking, problem-solving, and effective communication skills, because of detailed, timely and consistent feedback.

While multiple choice quizzes have their place in assessments, they fall short when it comes to personalisation. To truly tailor teaching & learning, AI-assisted marking of descriptive student work is the way forward. The goal of education is not just to pass exams but to prepare learners for success in the workplace. AI-assisted assessment is a significant step towards achieving that goal, creating a more personalised and effective learning experience for all.

‍