ESAT candidate numbers: the real scale of the 2026 cycle

Technical report Table 4 lists four more: Maths 1 with Maths 2, 303 candidates (2%); Maths 1 with Biology and Maths 2, 110 (1%); Maths 1 with Biology and Physics, 103 (1%); and 16 candidates (0%) grouped as Other. The table is steeply concentrated: the Maths 1, Maths 2 and Physics combination alone accounts for 74%, and everything below it is tail. Which modules you take follows from the entry requirements of the courses you apply to, not from this distribution.

Annual report Table 3 splits bookings by test and window: for the ESAT, 5423 UK and 5079 overseas in October, 2257 UK and 2022 overseas in January. The technical report separately records area of residence, where UK candidates make up 48% of Maths 2 up to 51% of Chemistry, with a small EU share and most of the rest outside the UK and the EU. Score differences by residence and first language are not on this page: see the international candidates page.

Annual report Table 4 covers all overseas candidates: Asia 8745, Europe 1789, North America 650, Africa 173, Oceania 135 and South America 36. It is not broken down by test, so none of it can be reported as an ESAT figure. The report adds that against 2025 entry the fastest growth was in African markets at 42%, followed by Asian markets at 39%, and names China, India, Singapore, Malaysia, Hong Kong, the United States, the UAE, Türkiye, Spain and France as key markets outside the UK. Figure 1 charts individual countries, but no country-level value is printed in the text, so none is estimated here.

Across the compulsory Mathematics 1 cohort, 69% identified as Man (9820) and 30% as Woman (4195). The modules differ sharply: Biology has the highest proportion of women at 60% and Physics the lowest at 24%. The share with English as a first language is fairly uniform, between 53% and 56%. Among UK candidates the most common school type is a further education or sixth form college, from 37% on Biology to 41% on Physics and Maths 2; grammar school candidates are 17% to 21%, against a national figure of around 5% of state-funded secondary pupils attending grammar schools; and 7% to 10% received free school meals, below the national figure of almost 25%.

The test costs 75 pounds for candidates in the UK and the Republic of Ireland and 130 pounds anywhere else, and the annual report sets bursaries alongside that: 1300 UK candidates were awarded one in October (11.2% of UK candidates) and 1079 in January (15.4%), 2379 in all (12.8%), with 94.1% of applications approved and a 4% year-on-year rise in awards across the UK cohort. On access arrangements, annual report Table 7 gives an ESAT-specific figure of 614 candidates, or 4.3%; the technical report describes the same 614 candidates as around 5% of the candidate population, of whom 607 asked for extra time, pause-the-clock or both. The count is identical in both; only the way the percentage is stated differs, so name the document you are quoting.

The annual report keeps them in separate columns: 14,781 ESAT bookings against 14,193 tests delivered. The geographic breakdown in Table 3, with 5423 UK and 5079 overseas in October, counts bookings, and the annual report never repeats that split for tests actually taken. The technical report does, on a different basis: for the compulsory Mathematics 1 module it records 6,909 candidates resident in the UK (49%), 777 in the EU (5%) and 6,507 outside both (46%). Where a booking was made and where a candidate lives are two different measures, so do not quote them interchangeably.

Table 4 (continents), Table 5 (bursaries) and Table 6 (access arrangements by window) are not broken down by test: Tables 4 and 6 describe the whole UAT-UK cohort and Table 5 the whole UK cohort, and not one row in them belongs to the ESAT. Only Table 7 separates access arrangements by test, giving 614 and 4.3% for the ESAT. Attaching the label ESAT to figures from the other tables overstates what they say.

The annual report gives overseas candidates only by continent, and the 8745 for Asia is a continental total that is never broken down by test. Figure 1 does plot bars for the UK, China, India, Singapore, Malaysia and Hong Kong, but not one country-level value is printed in the text, and reading heights off a chart is not reliable. So how many candidates in mainland China sat the ESAT is a question the published data does not answer, and any specific figure is not an official one.

Counts can be compared across cycles, and both documents do it: the annual report prints +11% for the ESAT October window and +47% for January, while the technical report records 11,919 ESAT Maths 1 candidates in the previous cycle, 77% of them in October and 23% in January. Scores cannot. The scaling is fixed independently for each admissions cycle, so this cycle's scaled scores are not directly comparable with those from 2024/25. That is why this page compares counts year on year and prints no score at all.