Two students write on the same problem question, and both reach the conclusion that the defendant is liable. One states the rule from Donoghue v Stevenson, cites it, applies it precisely to the facts in the scenario, and weighs the counter-argument before landing. The other asserts liability, name-drops a case without explaining it, and never touches the facts. If your marking gives those two answers grades that are one band apart when they should be two, the problem isn’t the students — it’s that Cambridge International A Level Law (9084) is marked by levels of response, and levels-of-response marking is where a tired examiner’s judgement quietly slides. This guide is about holding that judgement steady across a full class set, and being honest about which parts of legal marking a machine can help with and which stay firmly on your desk.

Why 9084 marking isn’t a tally

There are no method marks in Law. You are not counting correct steps toward a right answer the way a maths marker does. Instead, a 9084 script is read holistically against levels of response — mark bands that describe the quality of a candidate’s knowledge, their application to the problem, and their evaluation of the law. Cambridge builds those bands around its assessment objectives, which broadly separate into:

• Knowledge and understanding — does the candidate know the relevant rules, statutes, cases and legal principles, and can they state them accurately?
• Application and analysis — can they take that knowledge and apply it to the specific facts of a scenario, or analyse an issue rather than describe it?
• Evaluation — can they weigh the law critically: its strengths, weaknesses, competing arguments, and the case for reform?

A high-band answer isn’t the one with the most cases crammed in. It’s the one that selects the right authority, applies it to these facts, and reasons toward a judgement. That is a matter of professional reading, not addition — which is exactly why it drifts.

Where levels-of-response marking drifts — and why it isn’t carelessness

Be honest about the twenty-eighth script. On the first few essays you read every line, hold the band descriptors in your head, and place each answer carefully — a strong Level 3 here, a mid Level 2 there. By the time the pile is two-thirds gone, the descriptors blur. You start marking by gut against the last script you read rather than against the fixed standard, so an essay marked at 9pm and a near-identical one marked at 11pm land in different bands. The “halo effect” creeps in too: a beautifully written but legally thin answer floats up, while a scruffy but doctrinally sharp one sinks.

None of this is a competence problem. It’s the predictable result of applying qualitative band descriptors to a stack of extended answers in one sitting — the limit is human attention, not effort. You can mitigate it: mark question-by-question across all scripts, keep the band descriptors and any exemplar answers open beside you, and re-read borderlines against the standard rather than against each other. But you can’t fully eliminate the drift by will alone. This is the same challenge the generic parent guide covers for every subject, marking to the Cambridge mark scheme with class-wide consistency — 9084 just makes it acute, because so much of the paper is extended legal reasoning.

What software can genuinely steady in 9084 — and what it can’t

Here is the honest scope, and it matters more in Law than in almost any subject on the platform. A machine does not mark an extended legal essay or a full problem-question answer well. Judging whether a candidate has correctly applied the ratio of a case to a novel set of facts, or whether their evaluation of a reform proposal is genuinely critical rather than merely descriptive, is a matter of legal judgement. That stays with you, full stop. Any tool that claims to auto-grade a 9084 evaluation essay is overselling.

Where consistency support is real is on the more structured, lower-tariff parts of a Law course:

• Recall and definition checks — the elements of a tort, the requirements of a valid contract, the definition of a criminal offence, the meaning of a source of law. These are close-ended enough to mark reliably.
• Case-and-rule matching — does a candidate correctly pair an authority with the principle it establishes? A structured item can hold that steady.
• Structured knowledge questions used as low-stakes homework or starters, where the answer is a defined set of points rather than an argument.

For those, a consistent first pass frees your attention for the essays. For the extended writing, the right model is what you already do at your best: a fixed standard, the band descriptors open, and every borderline re-read against the same line — supported by exemplars and a clear rubric rather than by a grade a machine assigned.

A 9084-specific marking workflow

1. Mark the structured knowledge items first and fast. Definitions, elements-of-an-offence recall, source-of-law identification — get these done consistently so the essays have your fresh attention.
2. Mark the essays and problem questions question-by-question, not script-by-script. Read every candidate’s answer to the same tort problem before moving on. Holding one scenario’s band descriptors in mind across the whole class beats context-switching between contract, tort and criminal law on every script.
3. Anchor to exemplars, not to the last script. Keep a top-band and a mid-band model answer beside you and place each essay against them. This is the single biggest defence against drift.
4. Separate the AOs when a borderline is stubborn. Ask explicitly: is the knowledge there? Is it applied to these facts, or just recited? Is there genuine evaluation? Splitting the judgement often resolves which band an answer belongs in.
5. Re-read every borderline before you commit. A band boundary in Law can be a grade. Mark the clear ones, flag the borderlines, and return to them against the standard.

Why consistent 9084 marking matters beyond the time saved

The bigger payoff isn’t speed — it’s that your data becomes trustworthy. When you mark evaluation consistently, a pattern in your analytics (“this class states the law well but never evaluates it”) is a real teaching signal, not an artefact of you marking the evaluation-heavy questions last and hardest. Inconsistent marking buries that signal in noise.

It also makes your marks defensible. When a student asks why their contract essay sat a band below a friend’s, “yours stated the law but didn’t apply it to the facts; theirs did, and that’s the difference between the bands” is an answer grounded in the assessment objectives — not in how tired you were on script twenty-eight.

How this looks on the platform

Let me be straight about where Tutopiya is with 9084: there is no live Cambridge International A Level Law resource on the platform today, and the tool does not auto-mark extended legal essays or full problem-question answers — that judgement stays with the teacher, as it should. What the platform offers a Law department right now is the underlying methodology: a consistent way to mark the structured, close-ended items (definitions, elements of an offence, case-and-rule recall) to a fixed standard across a class, and a rubric-and-exemplar approach that keeps your own essay marking anchored. The same levels-of-response marking support that works on the platform’s live subjects will apply to 9084 once its resources are built. You can see how that methodology is put to work across the teacher platform behind these guides.

This is one of four 9084 guides for teachers. The others cover the 9084 past-paper question bank, building a 9084 mock exam from past papers, and 9084 lesson resources mapped to the syllabus.

FAQ

Can a machine mark a 9084 evaluation essay for me? No — and you should distrust any tool that says it can. Judging whether a candidate has genuinely applied a case to the facts, or evaluated the law critically rather than described it, is legal judgement that stays with the teacher. What software can steady is the structured recall and definition items, freeing your attention for the essays.

How is marking Law different from marking a numeric subject? There are no method or accuracy marks in 9084. A script is read holistically against levels-of-response bands describing the quality of knowledge, application and evaluation — not tallied step by step. That makes anchoring to band descriptors and exemplars, rather than counting points, the core marking skill.

What actually causes marking to drift across a class set? Fatigue applied to qualitative descriptors: the bands blur, you start marking against the last script instead of the fixed standard, and the halo effect lifts well-written but thin answers. Marking question-by-question and anchoring to exemplars are the main defences.

Should I mark by script or by question? By question. Read every candidate’s answer to the same problem before moving on, so you hold one scenario’s standard in mind across the whole class rather than switching between contract, tort and criminal law on every script.

How do I keep my essay marks defensible? Anchor to the assessment objectives and to exemplar answers. When you can say “this answer stated the law but didn’t apply it to the facts, which is the difference between the bands,” your marks are grounded in the standard rather than in how far down the pile you were.

The bottom line

Marking 9084 well means judging knowledge, application and evaluation against fixed levels-of-response bands the same way on every script — which is exactly what a tired reader can’t sustain across a class set of essays. Let consistency support handle the structured recall, keep the extended legal reasoning firmly in your own judgement, and anchor every borderline to the band descriptors and an exemplar. That’s how your Law marks become both fairer to students and trustworthy as data.

See the marking methodology behind these guides →

Ready to Excel in Your Studies?

Get personalised help from Tutopiya's expert tutors. Whether it's IGCSE, IB, A-Levels, or any other curriculum — we match you with the perfect tutor and your first session is free.

Book Your Free Trial