In the UK, GCSE and A-level exam scripts are marked by external examiners employed by the exam board — not by the student's own teacher — and most marking today happens on-screen, using scanned images of the original paper scripts rather than the paper itself. Before live marking starts, examiners go through a standardisation process against a senior examiner's benchmark, and marking is monitored throughout the process. Other countries use quite different models, from centralised national marking centres to systems that lean more heavily on teacher assessment with external moderation — there is no single worldwide standard.

Marking is one of the least visible parts of the exam system to parents, because it happens entirely out of sight, between the exam hall and results day. It's also one of the more reassuring parts once you understand it, because the UK's model is specifically designed to keep marking independent of the relationship between a student and their own teacher.

Who actually marks a UK exam paper

Final GCSE and A-level written exams are marked by examiners recruited, trained and employed by the exam board — AQA, Pearson Edexcel, OCR, WJEC/Eduqas or CCEA — not by the school where the student sits the exam. Many examiners are practising or retired teachers themselves, but they are marking scripts from schools they have no connection to, under anonymised conditions (scripts are typically identified by candidate number, not name). This separation is one reason boards can be confident that marking isn't influenced by a teacher's existing relationship with, or expectations of, a particular student.

Where the picture is different is non-exam assessment (NEA) — coursework and controlled assessment components that exist in some subjects. These are typically marked initially by the student's own teacher, using the board's mark scheme, and then moderated by the exam board: the board reviews a sample of the school's marking and can adjust the whole school's marks up or down if its marking is found to be out of line with the standard. So teacher involvement in marking is real, but it's a specific, bounded part of the system, not how final written exams are marked. See exam vs coursework around the world for how the balance between exam and coursework varies by country.

On-screen marking: how it actually works today

Most UK exam marking now happens on-screen. After the exam, scripts are collected from schools and scanned into digital images — this is the physical handwriting itself, digitised, not a re-typed or re-formatted version. Examiners then log into the board's marking system and mark the scanned images on a computer screen, question by question, entering marks directly into the system rather than writing on paper.

On-screen marking has a few practical advantages boards have cited in their own guidance (see Ofqual's collection of GCSE, AS and A-level marking guidance):

  • A single script can be split between several examiners, each marking only the questions that match their specialism, rather than one examiner marking an entire paper cover to cover.
  • Marking consistency can be monitored electronically and in near-real time, because the board's systems can flag an examiner whose marks on shared or seeded scripts are drifting from the expected standard, allowing intervention during the marking window rather than only after results are issued.
  • Scripts are stored digitally, which matters for the review-of-marking and appeals process a school or student can request after results day.

Standardisation: keeping thousands of examiners consistent

A single GCSE subject can involve marking from hundreds or thousands of examiners nationally. To keep marking consistent across that many people, boards run a standardisation process before live marking begins: examiners mark a shared set of sample scripts, and their marks are compared against marks agreed by a senior examiner (sometimes called the principal examiner) for those same scripts. Examiners whose marking doesn't match closely enough are given further guidance or additional sample scripts before being approved to mark live scripts — and monitoring continues throughout the marking period, not just at the start. AQA and Pearson Edexcel both publish guidance for examiners describing this process (see AQA's guidance on managing marking and moderation and Pearson's examiner recruitment and training pages).

How other countries approach marking

Marking models vary considerably worldwide, and it would be inaccurate to describe any single alternative as "the" international standard. Broadly, three patterns are common:

  • Centralised marking centres or national examination bodies. Some countries route scripts through a single national examinations agency for marking, similar in spirit to the UK's external-examiner model but organised through one government body rather than several competing boards — see national curriculum vs board specifications for how that connects to curriculum design.
  • Teacher-marked systems with external moderation. In some systems, particularly those built around continuous or coursework-based assessment, a student's own teacher marks a significant share of the final grade, with a national or regional body sampling and moderating that marking rather than re-marking every script externally.
  • A mix of external written exams and school-based assessment, similar in principle to the UK's exam-plus-NEA structure but with different proportions and different moderation arrangements.

France's baccalauréat is a genuinely instructive middle case. Written papers are marked by teachers — but, critically, not by the candidate's own teacher, and usually not even by a teacher from the candidate's own school. Scripts are anonymised (identified by a candidate number, similar to the UK) and distributed to examining teachers drawn from across an académie (a regional education authority), so a Parisian student's philosophy paper might be marked by a teacher from an entirely different school in the same région. This sits somewhere between the UK's model and a teacher-marked system: the marker is a serving teacher rather than a board-employed specialist examiner, but the independence-from-the-candidate principle the UK relies on is preserved through geographic and institutional separation rather than through employing a wholly separate professional class of examiners.

International comparative frameworks like the OECD's PISA assessment framework note that assessment and marking practices differ significantly between education systems, which is part of why direct score comparisons between countries need care — a difference discussed further in PISA rankings and exam systems.

At a glance

UK (GCSE/A-level) France (baccalauréat) Teacher-marked + moderation model
Who marks Board-employed external examiners Teachers, but not the candidate's own, drawn from across the académie The candidate's own teacher
Independence mechanism Separate professional examiner workforce Geographic/institutional separation between marker and candidate External sampling and moderation of the teacher's marks
Format Scanned scripts marked on-screen Paper scripts, marked by hand Varies — often paper-based
Consistency check Standardisation against a senior examiner's benchmark before live marking Marking guidelines set centrally per subject; moderation meetings (commissions d'harmonisation) reconcile marks across examiners External moderator reviews a sample of the school's marking

Could AI ever mark GCSE papers?

This is a live question the UK's marking system hasn't settled, and it's worth raising rather than glossing over. On-screen marking already generates the kind of large, structured dataset — scanned scripts, marks awarded per question, examiner-level consistency data — that makes automated or AI-assisted marking technically conceivable, and Ofqual and awarding bodies have published research exploring comparative judgement and machine-assisted approaches for particular question types, mostly short or objectively-scored answers rather than extended essay responses. None of the major UK boards currently use AI to award live GCSE or A-level marks — human examiners, standardised and monitored as described above, remain the actual decision-makers — and extended, judgement-heavy answers (an English Literature essay, a History source evaluation) are exactly the kind of response current AI marking research treats as the hardest case, not the easiest. Whether that changes, and on what timescale, is genuinely unresolved rather than a settled "yes" or "no."

Why this matters to a parent

The core reassurance for a UK parent is structural: your child's final written exam is not marked by the person who taught them, is marked on a scanned digital copy monitored for consistency, and is produced by an examiner who has been checked against a shared standard before being allowed to mark live scripts. That doesn't mean marking is infallible — reviews of marking and appeals exist precisely because individual marking errors can and do happen — but the system is built with independence and consistency checks as deliberate design features, not afterthoughts.

The bottom line

UK GCSE and A-level marking is done by external, trained examiners working from scanned on-screen copies of a student's actual script, standardised against a shared benchmark before live marking begins and monitored throughout. Coursework is the exception, marked first by teachers and then moderated by the board. Elsewhere in the world, marking is organised differently — sometimes through a single national marking centre, sometimes leaning more on teacher assessment with external checks — and no single model is a universal standard against which the UK's approach should be measured as better or worse.