Marking AQA GCSE English Literature Paper 2 Faster With AI

AQA GCSE English Literature Paper 2 is one of the heaviest marking loads of the summer term. Three sections, four texts, two and a quarter hours of student writing, and a class of thirty papers landing on your desk at once. The marking does not respect the rest of the timetable. This guide is about how UK English teachers are using GradeOrbit to mark Paper 2 faster — without giving up the parts of marking that only a teacher can do.

Paper 2 is the right place to test an AI marking workflow because the marking criteria are well-defined, the volume is high, and the time pressure is real. The aim is not to replace the teacher's judgment on individual responses. The aim is to compress the mechanical parts of marking — applying AOs consistently, drafting initial comments, identifying SPaG patterns — so the teacher's time goes into the moments where judgment actually matters.

Why AQA Paper 2 Marking Takes So Long

Paper 2 has three sections and they each demand a different style of marking. Section A is the modern text response — usually a play or novel like An Inspector Calls, Lord of the Flies, or Animal Farm. Section B is the poetry anthology, asking for a comparison between a named anthology poem and a chosen second. Section C is the unseen poetry — one response on a previously unseen poem followed by a comparison with a second unseen poem.

Each section assesses different combinations of AO1 through AO4 and rewards different things. Section A wants critical reading of a whole text under timed conditions. Section B wants comparative analysis of methods. Section C wants the same comparative skill applied to texts the student has never seen before, plus the confidence to read closely under pressure. Marking each section against its mark scheme is the right thing to do — but it is also why a class set of Paper 2 papers can swallow an entire weekend.

Add the realities of physical scripts — handwriting, page numbering, the occasional crossed-out paragraph — and the time per paper climbs further. The bottleneck is not the act of judging good writing. It is everything else surrounding it.

Scanning Physical Paper 2 Scripts Into GradeOrbit

Paper 2 lands on paper. GradeOrbit handles this with a phone camera workflow that pairs your phone to your browser via a QR code, then lets you photograph each script page-by-page directly into the marking workspace. There is no separate scanner, no flat-bed setup, no need to leave the desk.

The flow that works for Paper 2 is to photograph all three sections of a single script in order — Section A, Section B, Section C — into a single student entry. GradeOrbit's OCR reads the handwriting and presents the transcribed text alongside the original page images, so you can spot-check accuracy at the points where it matters.

Before the script reaches the AI, you redact student names and any other personal information by drawing black boxes on the images. The redaction is burnt into the pixels, not metadata — the AI never sees what was under the box. The student appears in the system as "Student 1", "Student 2", and so on. Uploaded student work is never stored after the marking run completes.

Setting the AQA AO Criteria Once and Reusing Across the Class

GradeOrbit's marking criteria step is where you tell the AI what mark scheme to apply. For Paper 2 the right move is to upload the AQA mark scheme for the specific paper sitting — the AO weightings, the level descriptors, and the section-specific guidance for Section A versus Section B versus Section C.

You set this once at the start of the batch. The same criteria are then applied to every script in the class, which is the consistency win that hand-marking thirty papers in a weekend rarely achieves. A student marked at 8:00am on Saturday and a student marked at 8:00pm on Sunday now get the same rubric applied with the same precision. The variance that creeps in as a marker tires across a long session does not creep in here.

How GradeOrbit Handles the Three Paper 2 Sections Differently

Once the criteria are in place, GradeOrbit applies them to the student's writing and returns a mark for each section along with positive feedback and constructive feedback, both tied to bounding boxes on the original script image. The teacher reviews the marks, edits any they disagree with, and accepts or rewrites the feedback.

Section A responses tend to produce the most stable AI marks because the AO1/AO2/AO3 framework maps cleanly onto the writing. Section B comparisons are where the AI sometimes flags a comparison that is implicit rather than explicit — useful, but worth a teacher review. Section C unseen poetry is where AI marking is most useful as a first pass: it catches the obvious AO1 evidence and points you to passages worth a closer look, while the final mark and the higher-order judgment stays with the teacher.

For background on the handwriting side of the workflow, see our guide on whether AI can read messy student handwriting. For the physical-exam-paper side, our piece on marking physical exam papers faster with AI covers the broader workflow this fits into.

Reviewing AI Suggestions: Where Teacher Judgment Still Leads

The point of an AI marking workflow is not to accept what the model suggests. It is to start from a worked draft and refine it. For Paper 2 specifically, the moments where teacher judgment is irreplaceable are these.

The first is the borderline mark. The mark scheme bands have soft edges between Level 4 and Level 5, between Level 5 and Level 6. The AI gives you a starting point. You make the call. The second is the personal voice in a response — a Year 11 who writes about Lady Macbeth's complicity in a way you have not seen before, or a poetry comparison that finds a connection the mark scheme did not anticipate. AI marking will often score these competently; only a teacher will mark them generously where generosity is earned.

The third is the feedback tone. AI-generated feedback is competent and clear. It is not always written in the voice of a teacher who has taught this student all year. You edit the feedback into your voice before it goes back to the student. The marking workflow saves time precisely so this final step can happen properly.

Start Marking Your Paper 2 Stack With GradeOrbit

GradeOrbit gives UK English teachers a marking workflow built around the realities of summer term Paper 2 marking — physical scripts, AQA criteria, three sections per paper, thirty papers per class. Student work is never stored. New teacher accounts include free credits so you can mark a class set of mocks or a real Paper 2 batch before committing.

Visit the GradeOrbit homepage to see how AI marking and AI detection fit together, and create a free teacher account to try the workflow on your next set of Paper 2 scripts.