University Exam Mark Schemes: What Markers Are Actually Looking For

Nobody shows you the mark scheme. That is the whole problem with university exams.

Your lecture notes explain the content. The past papers show you the format. But the document that tells you exactly what earns the marks — the mark scheme — stays in the exam office and is never handed to students.

Here is the structure of that document, and what it means for how you should write.

What university mark schemes contain

Written exam mark schemes typically include three components:

A model answer or key points list: the facts, arguments, or solutions the examiner considers correct or credible. These are the building blocks. If you include them, you have started building toward the passing mark.

A grade descriptor: a description of what an answer at each grade boundary looks like. Something like: "A first-class answer demonstrates original analytical argument, engages with counterpositions, and accurately applies relevant theory. A 2:1 answer demonstrates accurate factual knowledge and competent application but lacks analytical depth." This descriptor is what the marker holds in their head as they read your script.

Guidance on discretion: how to handle unexpected but credible answers, partial answers, or correct points not on the key list. A well-written mark scheme tells markers that a point not listed can still receive credit if it is relevant and well-argued.

The structure most students misread

Most students think exam marking works like this: list the key points, get the marks. Get all of them, get a First.

What the grade descriptors reveal is that this is roughly correct for a 2:2. It is not how you get a First.

A mark above 70% in a humanities, social science, or law exam almost always requires something beyond accurate knowledge. It requires an argument. The marker is not just checking whether you know the material — they are checking whether you have done something analytically interesting with it.

In quantitative subjects — maths, hard sciences, engineering — a First is more about execution accuracy and showing method clearly. But in discursive subjects, the grade descriptor for a first-class answer almost universally uses language like: "critical analysis", "evaluates competing perspectives", "develops a sustained argument", "goes beyond description to synthesis."

Describing accurately is a 2:1. Arguing originally is a First.

Four practical implications

Lead with your argument, not your introduction.

Many students write a long introductory paragraph that restates the question and outlines what they are about to cover. Markers have seen this in every script. The answers that stand out start with a claim: "The central tension in X is Y, and this essay argues Z." One sentence. The marker now knows where you are going.

Acknowledge the counterargument explicitly.

Grade descriptors for first-class answers consistently mention engagement with competing views. You do not need to demolish the opposing position — just show that you know it exists and that your argument holds up against it. "Critics of this view argue X; however, when applied to [specific case], this objection does not hold because Y." This signals analytical thinking rather than one-track knowledge reproduction.

Use the full mark budget consistently.

Most written exams have mark allocations per question. Getting 18/25 on every question in a five-question paper is a 72% — a First. Getting 25/25 on four questions and 0/25 on one is also mathematically 80%, but that zero drags you down in ways the consistent approach does not. Time management that produces even, solid answers across all questions outperforms perfectionism on some and neglect on others.

Show your working in quantitative answers.

When the mark scheme says "allow method marks", it means the examiner can award marks for correct procedure even if the final answer is wrong. A wrong numerical result with the right method is better than a blank. This is also why clear, annotated working matters: markers cannot award method marks for workings they cannot follow.

The variation between modules

One thing that rarely gets discussed: what it takes to reach the first-class band varies significantly between modules, even on the same course.

Some modules have mark schemes that are genuinely generous — the key points list is long, partial credit is awarded liberally, and a 70%+ mark is achievable with thorough coverage. Others are structurally harder — the grade descriptor requires original argument, and the distribution reflects it, with very few students breaking 70% in any year.

This module-level variation in grade thresholds is exactly what the underlying data captures. Historical mark distributions by module, sourced via FOI requests, show whether a module regularly produces Firsts or whether the distribution typically tops out in the high 60s — before you commit your credit weight.

For tactical guidance on translating that into module choices, see how to choose university modules.

Getting feedback

Mark schemes are not shown to students during exams. But feedback on assessed coursework — and, at some universities, on exam scripts on request — gives you a proxy for what the mark scheme rewards.

When you read back marker feedback on a piece of work you scored well below your expectation on, look for the language the marker used. "Could be more analytical." "Needs to engage with counterarguments." "Accurate but descriptive." These phrases are the grade descriptor for a 2:1 answer. They tell you exactly what would have pushed the mark into first-class territory.

Using feedback this way — as a guide to the mark scheme's logic — is more useful than treating it as commentary on a single piece of work.