How GCSE grades are actually decided behind the number scale

Every August, students discover that the marks needed for a grade 7 in one subject differ from last year's, and differ again between exam boards. This is routinely reported as boundaries being "lowered" or "raised" as if by generosity or stinginess. In fact it is the system working exactly as designed, and the design is worth understanding before results day rather than after.

GCSE grades in England run from 9 at the top to 1, replacing the old A* to G scale. The crucial fact is that grade boundaries are set after all the papers are marked, not before students sit them. Senior examiners review the year's scripts, statistical evidence and archive examples of previous years' work, and place the boundaries so that the standard a grade represents stays constant even though the difficulty of any particular paper cannot be made identical year to year. A harder paper gets lower boundaries. An easier one gets higher boundaries. The grade, not the mark, is the stable unit.

Underpinning this is an approach the regulator Ofqual calls comparable outcomes. Each cohort's results are anchored to that cohort's prior attainment, measured from earlier assessments. If this year's sixteen-year-olds arrived with a similar profile to last year's, the proportions achieving each grade should look broadly similar nationally. The intent is to stop two distortions at once: grade inflation drifting standards upwards over time, and a cohort being punished because a paper turned out harder than intended.

What this means in practice

Several familiar mysteries dissolve under this lens. Boundaries differ between exam boards because their papers differ in difficulty, while the grades are aligned to the same standard, so a 6 is designed to mean the same thing everywhere. Practising on last year's paper with last year's boundaries only ever gives an estimate. And no teacher, however experienced, can promise that a given raw mark will secure a given grade next summer.

The system has real limitations, and they get debated every cycle. Anchoring outcomes to prior attainment means genuine national improvement in teaching shows up only slowly. Grades near a boundary carry unavoidable marking uncertainty, which is why the review and appeal system exists, and why a script one mark below a boundary is always worth a priced, considered decision about a review rather than an automatic one.

For students and parents, the usable summary is this: chase marks, not boundaries. The boundary will move to wherever the standard sits, and the only variable a candidate controls is how much of the paper they can do.