Oxford runs an agreed interview mark scale and bands candidates centrally — here is what its own framework says
Book a Free ConsultationOxford runs an agreed interview mark scale. Its Common Framework for Admissions says so in as many words, and adds that faculties band candidates centrally at the final offer stage, with colleges “guided but not bound” by that banding. None of the tutoring pages ranking for this question mention any of it. Around 10,000 of roughly 23,000 applicants are interviewed in December, and this is the machinery that sorts them.
If there is a mark scale, there is something to rehearse against. Our subject interview packs (£180 each) are written by specialist subject tutors as worked reasoning rather than model answers, so you can see what a strong response does step by step. Nine subjects have a free sample.
See the packs and free samples Or book a free consultationAt Oxford there is, and the evidence is public. Its Common Framework for Admissions lists, among the things faculties and colleges must have regard to, “the need for agreed interview criteria and an agreed interview mark scale, with guidelines to be provided by the Undergraduate Admissions Office.”
The framework exists to make outcomes comparable rather than idiosyncratic. Its first stated aim is to attract “the most academically able individuals, irrespective of socio-economic, ethnic or national origin”, and its second is to select applicants “on the basis that they are well qualified and have the most potential to excel in their chosen course of study”. A mark scale is what makes either aim auditable.
Two qualifications matter. The scale is agreed per subject, not imposed universally — the framework is explicit that “procedures may vary between subjects, depending on the number of applications, whether there is pre-interview testing, whether they are part of a Joint School and other factors”. And the scale itself is not published, so nobody outside can tell you what a 7 means in History.
What follows from its existence is still substantial. An interview is not a tutor forming an impression; it is a tutor producing a number against criteria their faculty has agreed. That is why the same question can be asked of different candidates by different tutors and still be comparable, and it is why preparation that targets the criteria beats preparation that targets a personality.
The interview is the second filter, not the first, and the framework is candid about how the first one works. Under “Shortlisting for interview” it notes “the potential for the more popular subjects to be more stringent in their shortlisting so as to allow adequate time for interviewing those who have been shortlisted.”
That is an unusually practical admission. Shortlisting stringency is partly a function of interviewing capacity: a subject with very many applicants cannot interview them all in a fortnight in December, so it cuts harder on paper. It is not a statement about the quality of the field, and it explains why shortlisting rates differ so much between courses that are equally hard to get into.
The framework also sets a standard for the tests used to do that cutting. Testing should be planned carefully “especially to ensure that as far as possible they test for aptitude and do not dissuade good candidates from applying” — a criterion about the test’s effect on who applies at all, not only about what it measures.
And the shortlist itself is banded centrally, on “information in application forms, pre-interview test results (where these exist), written work where requested and the assessment of ability”. For a course with no test and no written work, that leaves the UCAS form doing almost all of the work at this stage. Oxford Biochemistry is the clearest example: it requires neither, and it interviews 41% of its applicants against 24% for Biomedical Sciences, which requires the ESAT.
Most candidates assume the tutor who interviewed them decides. The framework describes something more layered.
| Stage | What the faculty does | What the college does |
|---|---|---|
| Shortlisting | Central banding of candidates, based on the application form, pre-interview test results where they exist, and written work where requested | Applies the agreed criteria |
| Interview | Sets agreed interview criteria and an agreed interview mark scale, with guidelines from the Undergraduate Admissions Office | Interviews, sometimes at two separate colleges |
| Final offers | Central banding again, now using the interview and any testing done during it | “Guided but not bound” by that banding |
| Afterwards | Reports annually to the Admissions Executive and to EPSC, which signs off each subject | Provides an annual certificate of assurance |
The load-bearing sentence is this one, on final offers: “The benefits to be had from central banding by the relevant faculty at the final offer stage, based on the information already used for shortlisting and the additional information available as a result of interviews and any testing during the interview process. Colleges to be guided but not bound by such banding.”
So the faculty ranks the whole field and the college makes the offer, with the ranking in front of it but not binding it. That division explains two things candidates find strange: why a strong interview at one college can still end in no offer there, and why the same candidate can be picked up by a different college. It is one field being sorted, not a set of separate contests.
Because criteria are set per subject, the useful ones are published by faculties rather than centrally. Oxford Law’s are among the most explicit, and they repay reading in full because they are unusually operational.
| Criterion | What Oxford Law says it means |
|---|---|
| Application | “Motivation and capacity for sustained and intense work” |
| Reasoning ability | “Ability to analyse and solve problems using logical and critical approaches, ability to draw fine distinctions, ability to separate the relevant from the irrelevant, capacity for accurate and critical observation, capacity for sustained and cogent argument, creativity and flexibility of thought and lateral thinking” |
| Communication | “Willingness and ability to express ideas clearly and effectively; ability to listen; ability to give considered responses” |
| Not a criterion | “Existing knowledge of the law is not a criterion of admission” |
Oxford Law states that the admissions process “as a whole is designed to identify which applicants possess them in the greatest measure”. Read that with the last row and a common preparation strategy collapses: “Existing knowledge of the law is not a criterion of admission.” A candidate who spends October memorising case names is optimising for something the faculty has said in writing it does not assess.
The same page explains what grades are for at this stage. Existing achievement in examinations, predicted results and school reports, “as well as performance in the written test and interview, is relied upon mainly as evidence of future potential.” Nothing there is a reward for past work; it is all treated as a predictor.
Criteria like “capacity for sustained and cogent argument” only mean something once you have been held to them out loud. Our specialist tutors run mock interviews that push on exactly those points and tell you afterwards which one gave way first.
Book a Free Consultation Message us on WhatsAppMy son was low in confidence after COVID and the gaps in learning were building up. Using this service helped plug those gaps, rebuild confidence, and improve his ability to learn properly again.
We arranged maths and English tuition ahead of GCSEs and both tutors were brilliant. My daughter's grades improved, and by results day she achieved two grades higher than in her mock exams.
I attended an English bootcamp with Leading Tuition and the experience was excellent. The teacher gave focused attention to the small group while keeping the lessons interactive and engaging.
A second interview at a college you did not apply to reads, to a nervous candidate, like a bad sign. The framework treats it as ordinary: candidates are interviewed “where appropriate in two separate colleges, and for these interviews, where possible, to be arranged in advance.”
Joint School applicants have their own rule: they “should be interviewed by at least one tutor in each of the subjects that make up the Joint School. This may be in a single panel, or by subject tutors separately, or in some other combination.” If you have applied for History and Politics or Physics and Philosophy, expect both halves to be examined by someone who owns that half.
All of this serves the framework’s third stated aim: “to ensure that the prospects of admission are not affected by the college an applicant has chosen or been assigned to through an open application.” The second interview is the mechanism, not the verdict.
The governance is public too, and it names the metric. Faculties “will be required to report annually on the procedures in place for their subjects, to the Admissions Executive”, which reports to the Education Policy and Standards Committee. Faculties must provide data on how the procedures work in practice, and the framework says “one important indicator might be the extent of redistribution of candidates between colleges.” Colleges separately provide an annual “certificate of assurance” that they are running admissions fairly.
That is a more interesting sentence than it first appears. If redistribution between colleges is the health check, then movement between colleges is expected — a system where nobody moved would be evidence that college choice was determining outcomes, which is the thing the framework exists to prevent.
Comparability is the purpose of the whole apparatus, and comparability has a practical consequence: the interview is designed to elicit evidence against named criteria, not to find out whether the tutor likes you. Cambridge describes the same intent from its side, telling candidates that “the people interviewing you are interested in how you work out your answers and how you think.”
Three things follow for preparation. Target the criteria your faculty publishes rather than a generic impression of what is impressive. Expect to be pushed, because “capacity for sustained and cogent argument” cannot be observed in an answer nobody challenged. And accept that content knowledge is the smallest part — Kathryn Scott, interviewing for Biochemistry at Christ Church, says of Oxford’s own published question that “getting the correct answer isn’t important here.”
Oxford interviews online in December 2026; Cambridge’s main period runs 7–18 December 2026. For how to make your reasoning audible against these criteria, see how to think out loud in an Oxbridge interview; for the practical setup, Oxford’s three interview technology tiers; and if your course sits a test first, what replaced what in Oxford’s admissions tests.
Comparability is the point of the whole apparatus, which means the interview rewards the same things in December that it rewards in rehearsal. Our specialist tutors work from published questions, and our subject packs (£180 each) give you the reasoning written out to study between sessions.
Browse the subject packs Or book a free consultationAt Oxford, yes. Its Common Framework for Admissions records “the need for agreed interview criteria and an agreed interview mark scale, with guidelines to be provided by the Undergraduate Admissions Office.” The scale itself is not published, and it is set per subject rather than centrally, but its existence means an interview produces a score that can be compared across colleges rather than an impression held by one tutor.
Not solely the tutor who interviewed you. Oxford’s Common Framework describes “central banding by the relevant faculty at the final offer stage, based on the information already used for shortlisting and the additional information available as a result of interviews and any testing during the interview process” — and then adds that colleges are “to be guided but not bound by such banding.” The faculty ranks; the college decides, within that guidance.
The framework is explicitly designed so that it should not. One of its three stated aims is “to ensure that the prospects of admission are not affected by the college an applicant has chosen or been assigned to through an open application.” Oxford also audits this: faculties report annually, and it names “the extent of redistribution of candidates between colleges” as an important indicator of whether the procedures are working.
Because the framework anticipates it. It refers to candidates being interviewed “where appropriate in two separate colleges, and for these interviews, where possible, to be arranged in advance.” A second interview is a fairness mechanism rather than a verdict on the first one. Applicants to Joint Schools have a separate rule: they “should be interviewed by at least one tutor in each of the subjects that make up the Joint School.”
Several publish them in full. Oxford Law lists three — Application, Reasoning ability and Communication — and defines each. Reasoning ability alone covers analysing and solving problems, drawing fine distinctions, separating the relevant from the irrelevant, accurate observation, sustained and cogent argument, and lateral thinking. It then adds the line most preparation ignores: “Existing knowledge of the law is not a criterion of admission.”
They matter as evidence rather than as a score. Oxford Law puts it precisely: existing achievement in examinations, predicted results and school reports, “as well as performance in the written test and interview, is relied upon mainly as evidence of future potential.” The framework says the same in general terms — applicants are selected on being “well qualified and have the most potential to excel in their chosen course of study.”
Less than candidates assume. Kathryn Scott, who interviews for Biochemistry at Christ Church, says of Oxford’s published question: “We do not expect the student to know the answer to this question straight away, and in fact getting the correct answer isn’t important here.” What is marked is how you apply what you know to something unfamiliar, how you justify conclusions and whether you consider alternatives. Cambridge says its interviewers “are interested in how you work out your answers and how you think.”
By making them concrete. A phrase like “capacity for sustained and cogent argument” is only useful once someone has pushed on your argument and shown you where it gave way. Our specialist tutors run mock interviews that do that, in your subject, using published questions. Our packs are £180 each with worked reasoning rather than model answers, nine subjects have a free sample, and 91% of our students achieve their desired grades.
Free samples in nine subjects show you the format, the full expert packs are £180 each with worked reasoning rather than model answers, and 1-to-1 mock interviews put the criteria under pressure.
See the packs and free samplesLeading Tuition is rated 4.8 out of 5 on Trustpilot, based on 58 reviews from families across the UK. 91% of our students achieve their desired grades.