ESAT preparation guide

How ESAT scoring works, and how to read your result

There is no overall ESAT score. You receive one score for every module you sit, reported from 1 to 9 to one decimal place, and that number is not a count of the questions you answered correctly. This page covers how the number is produced, how to read it, when it arrives and who sees it, and which distribution figures UAT-UK has actually published for you to measure yourself against.

Practise 57 ESAT papers (2 free)Full ESAT guide

Papers: 57
Questions: 2392
Free to try: 2

In short

Every ESAT module is scored separately from 1 to 9 to one decimal place. There is no total, so modules cannot compensate for each other.
The reported score is not your raw mark. Equating maps performance onto a unified scale, which is why a 6.0 means the same thing across sittings and why no official conversion table exists.
UAT-UK sets no pass mark and publishes no grade boundaries, but its per-sitting Explanation of Results gives the distribution: typical candidates score around 4.5, roughly 10% score above 7.0, and scores are capped at 1.0 and 9.0.
Results take about four weeks: 16 November 2026 and 8 February 2027 for the 2027 cycle. They are matched to UCAS course by course and sent automatically, with no candidate choice.
There is no appeal and only one sitting per cycle, so everything you control happens before test day: module choice, sitting choice and timed practice.

Your result is several scores, not one

There is no overall ESAT mark. You get one score for each module you sit, reported on a scale from 1 (low) to 9 (high) to one decimal place. UAT-UK puts it plainly: "For the ESAT, you will receive a separate score for each module that you take."

That has a consequence most candidates underestimate. Because nothing is aggregated, modules cannot compensate for one another at the reporting stage. A strong Physics score does not rescue a weak Mathematics 1, and the reverse is equally true. An admissions tutor opens your file to three separate numbers, or two if you are applying to one of the few courses that requires only two modules, such as MEng Design Engineering at Imperial, which specifies Mathematics 1 and Mathematics 2. There is no single rounded impression to hide behind.

For most courses the structure is Mathematics 1 plus two further modules. Mathematics 1 is compulsory for everyone; the other two are usually set by the course, though some courses let you choose freely from Biology, Chemistry, Physics and Mathematics 2. The revision discipline follows directly. Pouring your effort into the module you already enjoy is the most common strategic mistake on this test. What actually decides the outcome is the weakest of the modules your chosen course specifies.

The number on your result is not the number you got right

Each module contains 27 multiple choice questions. Each correct answer scores one mark and nothing is deducted for a wrong one: "In all modules, each correct answer will score 1 mark. No marks are deducted for incorrect answers." So your raw mark for a module is a whole number between 0 and 27. That number never appears on your result.

What appears is a scaled score between 1 and 9, produced from the raw mark by the statistical process described below. Conflating the two causes two recurring errors. The first is self-estimating: walking out of the centre, counting roughly twenty right, and converting that into a scaled score by proportion. There is no basis for that conversion. The second is reading forum posts. When somebody says they "got 22", that is almost certainly a raw mark, and it is not measured in the same units as the figure on your result, so comparing the two tells you nothing. UAT-UK also forbids discussing the paper afterwards in as many words, "You must not share or discuss the test with anyone else", so those recollection threads should not exist in the first place and are worth very little.

The absence of negative marking also carries an operational instruction. Never leave a question blank. The official wording is that it is worth attempting all questions, and with seconds left on the clock a guess is strictly better than an omission.

One related misconception is worth correcting: leaving a question blank does not bank you any time. Each module is separately timed, if you finish early the remaining minutes are not carried over to the next module, and there are no breaks between modules. For most candidates the whole ESAT is 120 minutes run straight through. Filling in every uncertain answer within the module you are in costs you nothing at all.

Equating, and why a 6.0 this year matches a 6.0 last year

Raw marks become scaled scores through equating. On its results page UAT-UK says the response data is analysed using the Rasch item response theory model, and the outcome is placed on a common scale. The stated purpose is precise: "This allows us to share scores from various test versions and sittings on a unified scale."

In plain terms, no two versions of a paper are exactly equal in difficulty. If scores were reported simply as marks out of 27, candidates who happened to sit an easier version would be rewarded for their luck. Equating estimates the difficulty of each question, infers from the response pattern the level of ability that implies, and maps that ability onto the 1 to 9 scale. The result is that a 6.0 from this October and a 6.0 from last January describe the same standard of performance, even where the two candidates answered different numbers of questions correctly.

It also explains why no official raw-to-scaled conversion table exists. Such a table would differ between versions and between sittings, so publishing one would mislead more than it informed. Any table you find circulating online can be treated as unsourced by default.

One further implication is useful when you choose a sitting. If equating exists to neutralise differences in version difficulty, then the question of which sitting is easier does not really arise. UAT-UK states on its deadlines page that there is no advantage to sitting in either session, only that you may sit once within an admissions cycle.

Each ESAT module is equated on its own, unlike the TMUA

If you are preparing for more than one UAT-UK test, one sentence on the results page is worth memorising: "For the ESAT and the TARA each module is equated and scaled separately. For the TMUA, Papers 1 and 2 are equated and scaled together to produce a single overall score."

So your Chemistry score and your Mathematics 1 score sit on two independent scaling chains, each aligned against its own history of earlier versions. How your Chemistry module is scaled has nothing to do with how you performed on Mathematics 1 an hour earlier. The TMUA works the opposite way round: two papers, one combined score, so on that test compensation between the papers genuinely exists.

This is where advice borrowed from one test to another comes unstuck. A 6.0 on the TMUA and a 6.0 on an ESAT module are both written on a 1 to 9 scale, but they emerge from different equating processes applied to different candidate populations. They cannot be converted into one another and should not be compared.

In practice, most candidates sit three modules, which means three independent scaling outcomes rather than three points on one curve. Review them module by module, and resist the urge to average them into a single sense of how you did. That average exists in no university's system.

No pass mark, but there is an official score distribution

Two separate points here, and they should not be run together.

First, there is no pass mark: "There is no pass/fail score for the ESAT. Scores are typically used alongside other information provided as part of your university application." UAT-UK publishes no grade boundaries either. This is not information being withheld until later in the cycle. The test is designed not to have them. The second half of that sentence matters as much as the first: the score is read alongside everything else in your application rather than functioning as a gate on its own.

Second, no pass mark does not mean no reference points. UAT-UK publishes an Explanation of Results for every sitting, and the edition for October 2025 is explicit: "The scale has been designed so that typical candidates will score around 4.5. Approximately 10% of candidates will achieve scores higher than 7.0. Low scores are capped at 1.0 and high scores are capped at 9.0."

Those three sentences give you three usable coordinates. Typical candidates sit around 4.5; roughly one in ten scores above 7.0; and the scale is capped at both ends, so 9.0 does not mean twenty-seven correct answers, it means the top of the scale, with 1.0 as its mirror image. The same document carries a score distribution for each of the five modules, so what the Mathematics 1 cohort looks like and what the Chemistry cohort looks like are published separately. Read the edition for your own sitting once results are out: it is closer to what you actually want to know than anyone's recollection thread.

One website error is worth flagging. On the UAT-UK test results page the 90th percentile appears twice with two different numbers: an earlier passage says 7.0, while the later section headed How our Tests are Scored says 9.0. Judged against the Explanation of Results published each sitting, 7.0 is the correct figure and the 9.0 is a slip on the web page. Anyone working a target out of the 9.0 has almost certainly copied that slip.

One last thing has to be said plainly. Knowing that 10% score above 7.0 is not the same as knowing what a particular course wants. The first is the distribution across all candidates; the second is each university's own judgement, and nobody publishes it.

What a course wants: only one university publishes anything

The distribution in the previous section describes all candidates, which does not answer what your own course wants. UAT-UK publishes no thresholds, and hardly any university publishes outcome data at that level of detail. The one exception for the ESAT is Imperial College London, whose page on understanding your ESAT and TMUA scores hosts a dashboard of 2025-entry scores and outcomes by department.

Several things have to be understood before reading it.

The dashboard reports a raw mark out of 27, labelled Mean test score (maximum is 27), not the 1 to 9 scaled score that appears on your result. No mapping between the two is published, so nothing on that dashboard transfers straight across as a target.
The page does not state whether the figure covers a single module or is averaged across the modules a candidate sat. That ambiguity alone defeats any fine-grained reading.
Only the offer-holder column is reasonably solid. The dashboard also shows interviewed-but-unsuccessful and not-interviewed averages, and Imperial does not say whether the latter includes applicants who registered but never sat, so it cannot be read as the true standard of unsuccessful candidates.
The all-departments figure includes applicants who did not sit the test at all, so it is skewed by construction. Only the department-level rows carry any meaning.
Imperial says on the same page that the data is historic and "should not be taken as indicative of requirements or outcomes for the current or future admissions cycles", that a high score does not guarantee a place while a low score does not mean the application will fail, and that there are no grade boundaries or pass marks for either test.

There is one further reading note for candidates in mainland China. The dashboard splits Home from Overseas, and in every department the mean score of Overseas applicants who received offers is higher than the Home equivalent, while Overseas offer rates are generally lower. Candidates from mainland China count as Overseas, so if you are going to read the data at all, read the Overseas column rather than the combined one. The widest gap is Electrical and Electronic Engineering, where the 2025-entry mean for Overseas offer holders was 19.8 against 14.2 for Home non-WP applicants, both out of 27. Again, those are historic averages, not thresholds.

Carrying numbers across years is just as risky. Durham states on its own website that a TMUA 6.5 in 2023 is equivalent to a 5.0 from 2024 onwards, which is to say that the same figure on two different scales is not the same thing at all. So when somebody hands you a number and tells you anything above it is safe, ask three questions: which year, which scale, published by whom. If any of the three has no answer, the number is not usable.

The test is meant to be hard, so school marks are the wrong yardstick

The specification contains the sentence that explains why so many candidates leave the centre convinced they have failed: "The test is designed to be challenging in order to differentiate effectively between able applicants, including those who might have achieved the highest possible grades in school examinations."

Read that as a statement of intent. The target population is the group already achieving the highest possible grades in school examinations, and the paper has to spread that group out or it is useless for admissions. "I have never slipped in maths at school, so why did that feel impossible" is not an anomaly. It is the design working as specified. Set it beside the distribution above and it becomes easier to live with: typical candidates land around 4.5, so this was never a test on which a high percentage correct is the normal experience.

Two disciplines follow. Calibrate against full timed modules rather than against your school marks: 27 questions in 40 minutes, no calculator and no dictionary, and nothing to write on but the erasable materials the centre provides. UAT-UK describes those materials differently in different places, an erasable booklet in the specification, an erasable notebook and pen on the test day page, a whiteboard and pen in the FAQ, so expect whatever your centre issues and raise your hand for a fresh one when you fill it. None of that resembles a school test where a calculator sits on the desk and there is time to go back over the paper.

Second, treat how it felt in the room as a poor predictor of the score. Because nothing is deducted for wrong answers, and because equating flattens differences in version difficulty, the link between "that went badly" and the eventual scaled score is weaker than it feels. Losing heart on a later module because an earlier one was rough is the thing that genuinely does show up in the numbers, and since there are no breaks between modules there is no pause in which your mood can recover on its own. That has to be a habit you build in practice.

When results arrive, with the exact 2027 entry dates

The official line is about four weeks: "We will notify you when your test results are available via your UAT-UK account. This will be approximately four weeks after your test sitting."

For 2027 entry the dates are already fixed, so there is nothing to estimate. Results from the October sitting, which runs from 12 to 16 October 2026 and is restricted to 12 or 13 October in mainland China, Hong Kong and Macau, are released on 16 November 2026. Results from the January sitting, 4 to 8 January 2027 and 6 January in mainland China, Hong Kong and Macau, are released on 8 February 2027. Both are accessed through your UAT-UK account.

UAT-UK has explained why those three territories have fewer test days. After reports in October 2024 of candidates attempting to obtain and sell test content without authorisation, the ESAT and TMUA are now offered on only two days in October and one day in January in mainland China, Hong Kong and Macau. That constrains your booking directly, which is why candidates in those places need to settle a date earlier than everyone else.

Put the results dates in the calendar before you choose a sitting. Cambridge and Oxford applicants whose course requires the test normally have to sit in October, the exceptions named by UAT-UK being Cambridge mature colleges with a January admissions deadline and Oxford's Astrophoria Foundation Year. Applicants elsewhere can choose, and UAT-UK states there is no advantage to either session. Even so, a January result does not exist until 8 February, and whether that timing suits you depends entirely on which universities are on your form.

Scores go out automatically, course by course, and you cannot choose

You do not send your scores anywhere. They are matched against your UCAS application and delivered automatically: "We match to your UCAS application on a course-by-course basis. So, your score will only be visible to an institution if the course to which you are applying requires the test. This will happen whether the test is compulsory or optional. Candidates are not able to choose which institutions can see their score."

Three consequences. You may sit each UAT-UK test only once per admissions cycle, and the 2027 cycle contains both the October 2026 and the January 2027 sittings, so there is no trial run you can quietly withhold. Attempting to register or sit a second time within the cycle, including by booking under a second UAT-UK account, is treated as misconduct. Matching is by course rather than by institution, so at the same university a score is visible for a course that requires the test and invisible for one that does not. And optional does not mean opt-out: courses that list the test as optional receive the score just the same.

It is worth being precise about who the recipients are. The official UAT-UK page on the ESAT names four universities that use it: Imperial College London, the University of Cambridge, the University of Oxford and UCL. Seven UK universities use one or more UAT-UK tests, but for 2027 entry the other three appear in the course list only against other tests. Imperial's own page carries a list headed ESAT for other universities that reproduces all seven member universities, which reads easily as though all seven required the ESAT. This is exactly where those three questions earn their keep: when you meet a list, establish who published it and which test it is actually describing.

There is no appeal, so everything you control happens beforehand

The official wording is unambiguous: "The scoring process is rigorously tested, and quality assured at every stage before, during and after each testing window. For this reason, we do not accept appeals of test scores."

Put the chain together. One sitting per cycle, automatic delivery, no choice of recipients, no appeal afterwards. Every controllable step happens before you sit, and in practice there are three of them: getting the module combination right at the point of registration, since "these cannot be changed without cancelling and re-booking your test"; choosing the sitting; and doing enough properly timed practice at the real rhythm of 27 questions in 40 minutes per module. The fee is a reminder that this is a one-off act: £78 at test centres in the UK and the Republic of Ireland, £133 elsewhere, set by where the centre is rather than by your nationality or where you live.

Module choice carries one rule that is easy to miss. If you are applying to two or more courses that require the ESAT, and one lets you choose freely while another specifies particular modules, your selection must follow the course with compulsory modules. Register without them and your application to that course may be invalid. Conflicts like that have to be resolved before you book, because nothing can be salvaged after results day.

One last thing worth knowing. UAT-UK states that it is not connected to and does not endorse any commercial preparation service, and that all necessary preparatory materials are available free of charge to all candidates. Treat that as good news rather than a warning. Build the foundation on the official specification, the official preparation materials and the Explanation of Results for each sitting, and let any third-party resource, ours included, add timed practice and error analysis on top of that foundation rather than stand in for it.

FAQ

Frequently asked questions

Is 9 the maximum ESAT score, and what counts as a good score?

Each module is reported from 1 (low) to 9 (high) to one decimal place, and UAT-UK states that scores are capped at 1.0 and 9.0, so 9.0 is the top of the scale rather than twenty-seven correct answers. As for what counts as high, the Explanation of Results published for each sitting gives the coordinates: typical candidates score around 4.5 and roughly 10% score above 7.0. That is a distribution, not a threshold: UAT-UK sets no pass mark and publishes no grade boundaries. The only university that has published department-level ESAT figures is Imperial, and those are raw marks out of 27 rather than scaled scores, accompanied by its own statement that the data is historic, that a high score does not guarantee an offer and a low one does not rule you out. Whenever you meet a specific threshold figure, ask which year, which scale and published by whom.

When do ESAT results come out?

About four weeks after your sitting, through your UAT-UK account. For 2027 entry the dates are fixed: 16 November 2026 for the October sitting and 8 February 2027 for the January sitting.

Does the ESAT deduct marks for wrong answers?

No. Each correct answer scores one mark and nothing is deducted for a wrong one, and the official guidance says it is worth attempting every question. Never leave one blank, even when the clock has almost run out.

How many of the 27 questions do I need for a 6.0?

There is no published conversion, and any conversion would differ between versions and sittings. Raw marks pass through equating precisely so that differences in version difficulty are cancelled out. Conversion tables circulating online have no official source and should not be used to set targets. If you want the real picture, read the Explanation of Results for your sitting, which carries a score distribution for each of the five modules.

Are my ESAT scores sent to every university on my UCAS form, and can I opt out?

They are matched course by course and sent automatically, and you cannot opt out. An institution sees your score only where the course you applied to requires the test, and delivery happens whether that course lists it as compulsory or optional. Candidates cannot choose which institutions see the score.

Can I appeal or have my ESAT score remarked?

No. UAT-UK states that the scoring process is quality assured before, during and after every testing window and that it does not accept appeals of test scores. Combined with the one-sitting-per-cycle rule, that leaves nothing to fix after results day, which is why everything that matters happens before it.

Can I compare an ESAT score with a TMUA score?

Not directly. Both report on a 1 to 9 scale, but each ESAT module is equated and scaled separately while the TMUA combines Papers 1 and 2 into a single overall score. Different equating processes and different candidate populations mean the numbers are not interchangeable.

SOURCES

Official sources

MORE GUIDES

Other ESAT preparation guides

FrontierVUE is an independent practice platform. It is not affiliated with or endorsed by UAT-UK, Pearson, OCR, the University of Cambridge, Imperial College London, or any official admissions-test owner.

FrontierVUE 是独立的备考练习平台，与 UAT-UK、Pearson、OCR、剑桥大学、帝国理工学院或任何官方入学考试主办方均无隶属或背书关系。

Put it into practice

57 ESAT papers on FrontierVUE, including 26 Frontier Original mocks built to the current format, every question with a worked solution. The first 2 papers are free.

Practise ESAT papers See pricing