ExAIm vs traditional revision: what the data shows

The study

During the 2025 spring term, we compared outcomes between two groups of Year 11 students at a British curriculum school in the UAE:

Both groups took the same end-of-term mock exam.

Results

Group B students scored an average of **18% higher** on the mock exam compared to Group A. The improvement was most significant on extended-response questions (6+ marks), where AI grading feedback had the strongest impact.

Why the difference?

Three factors stood out:

1. More practice attempts

Group B students completed an average of 4.2 practice exams per week, compared to 1.8 for Group A. The instant feedback loop encouraged more frequent practice.

2. Targeted revision

ExAIm's analytics showed students exactly which topics they were weakest on. Instead of revising everything equally, they focused their time where it mattered most.

3. Immediate feedback

Group A students waited an average of 5 days for marked work to be returned. Group B students received detailed feedback within seconds of submitting.

Limitations

This was a single-school study with a small sample size. We're running larger studies across multiple schools in 2026 to validate these findings.

What this means for schools

The data suggests that AI-powered exam prep doesn't replace good teaching — it amplifies it. When students get more practice, better feedback, and targeted revision, outcomes improve significantly.