AQA Mark Schemes and AI Marking

AQA mark schemes are used widely in England. Understanding how AI marking tools interact with them — and where teacher oversight is needed — is essential for any department considering AI-assisted assessment.

Types of AQA marking and how AI handles each

AQA uses two main marking approaches depending on the question type. Point-based marking awards a set mark for each specific correct point — used for shorter recall and application questions. Level-of-response marking uses band descriptors to holistically judge extended answers — used for essays, evaluations, and longer analytical responses.

Point-based questions can give an AI clearer criteria to work against, while level-of-response questions need closer teacher review. The AI can use band descriptors to suggest a placement and provide evidence for its judgement; teachers remain responsible for the decision.

Indicative content and marking flexibility

A feature of AQA mark schemes that trips up many teachers — and that AI systems must handle carefully — is indicative content. These are suggested points that could receive credit, but they are not exhaustive. A student who makes a valid point not listed in the indicative content should still receive credit for it.

Well-designed AI marking systems are trained to recognise valid responses that go beyond the indicative content list, rather than only awarding marks for exact matches. When Easy Examiner encounters a response that appears valid but doesn't match listed content, it flags it for teacher review rather than automatically penalising the student.

Point-based questions: high AI confidence, minimal teacher review needed
Level-of-response questions: AI provides band placement with evidence, teacher confirms
Indicative content: AI flags plausible unlisted responses for teacher decision
Quality of Written Communication: AI assesses spelling, grammar, specialist vocabulary per AQA guidance
Alternative mark schemes: upload custom or department-adapted schemes alongside the official one

Practical tips for using AI with AQA papers

The quality of AI marking output is directly related to the quality of the mark scheme you upload. For best results with AQA papers, upload the full published mark scheme — including any additional guidance notes and specific examples — rather than a simplified version.

For papers with multiple components (e.g., AQA English Literature Paper 1 includes both a Shakespeare extract question and a 19th-century novel question), upload each component's mark scheme separately. This gives the AI the clearest possible context for each section.