Do Oxbridge Interview Questions Differ by College? What Changes for 2027 Entry

In January 2026, 1060 pooled applicants were offered a place by a different Cambridge College from the one they applied to, and only 20 were invited to a second interview. Nearly every one of those offers rested on questions asked somewhere else. That single fact settles most of this page: your subject writes the questions, your College arranges the room.

Who Actually Writes the Question You Get Asked?

The question feels like it belongs to the person asking it, and the person asking it belongs to a College. That is the intuition behind the whole worry, and it is the wrong way round. Oxford is explicit about who is in the room: “Applicants will be interviewed by Oxford academic tutors involved in teaching and research here.” Not College tutors. Oxford academic tutors. The same page adds that “typically, interviews are conducted by two tutors, although there may occasionally be more”, and that joint-course applicants “should expect to be interviewed by tutors representing each subject”.

Read that last clause slowly. The unit that has to be represented in the room is the subject. If you have applied for a joint course, Oxford does not promise you a tutor from a particular College; it promises you a tutor who owns each half of your degree. The College supplies the building, the timetable slot and the invitation email. The faculty supplies the academic who knows what a good answer to a History question looks like.

Cambridge frames the interview the same way at the top of its own guidance: interviews there “are academic conversations about the subject you’re interested in studying”. Again, subject. And when Cambridge lists what an interview assesses, every item is subject-shaped — your understanding of your chosen subject area, your readiness to study at a high academic level, your enthusiasm for your subject. None of it is College-shaped. There is no criterion about fitting a particular College’s character, and no published rubric anywhere that a College applies and its neighbour does not.

So the honest reformulation of the question is not “does my College ask different questions?” It is “does the College change anything that reaches the question itself?” That turns out to have a clean, sourced answer, and it is not the answer most preparation advice gives.

Oxford Publishes Its Sample Questions by Subject, Not by College

The strongest evidence sits in plain view on Oxford’s own website, and almost nobody reads it structurally. Oxford maintains a public bank of sample interview questions, and says of them: “These sample interview questions come direct from the tutors who conduct the interviews.”

Look at how that bank is filed. Every entry sits under a subject heading — Biology, Law, Mathematics, Modern Languages, Theology and Religion. Under each subject, Oxford names the tutor who supplied the question and, as an attribution, the College that tutor sits in. The result is an index no College could ever produce on its own:

• The Philosophy, Politics and Economics entry carries questions from tutors attributed to Lincoln College, St Peter’s College, Lady Margaret Hall and Brasenose College. One subject, four Colleges, one published set.
• English Literature draws on tutors at Mansfield College, Hertford College and Regent’s Park College.
• History draws on tutors at Oriel College, Pembroke College and Magdalen College.
• Psychology draws on tutors at St Anne’s College, University College, Brasenose College and New College.

Two entries do not name a College at all. The Law question is attributed to “Ben McFarlane, Faculty of Law”, and one Engineering question to “Byron Byrne, Department of Engineering Science”. Oxford, filing its own question bank, sometimes drops the College attribution entirely and credits the faculty or department instead. It never does the reverse. There is no College-headed section, no “questions asked at Balliol” tab, no mechanism by which a College could contribute a question that is theirs rather than the subject’s.

If questions were a College property, this is precisely the page where that would show. It does not.

What the Winter Pool Proves About College-Specific Questions

The second piece of evidence is an accident of administration, and it is the closest thing to a controlled experiment that admissions offers.

Cambridge states its pooling principle without hedging: “We want to make sure that the best applicants each year are offered a place, no matter which College they originally applied to or were allocated to.” Around 18% of applications received in October 2025 were placed in the winter pool. Then comes the number that matters. In January 2026, 1060 pooled applicants — around one in four of those pooled — were offered a place by a different College to the one they applied to. And in that same January 2026 round, 20 applicants were invited to a second interview.

Set those two figures beside each other. Over a thousand offers were made by Colleges that had never met the candidate, on the strength of an interview conducted by somebody else’s tutors, at somebody else’s College, from questions somebody else chose. A handful of applicants were re-interviewed; the overwhelming majority were not.

That system cannot work if the questions are College-specific. A College reading a transcript or a report from an interview it did not run has to trust that the assessment is portable — that a strong Natural Sciences interview at one College means the same thing as a strong Natural Sciences interview at another. Cambridge built its entire admissions economy on that assumption and publishes the numbers that show it holding. Our page on the Cambridge winter pool covers the mechanics of what happens and when; the point here is narrower and sharper. The pool is the proof.

Oxford runs the same logic in a different shape. Its guidance says: “Sometimes you might get invited to interview by a college you did not apply to. This is part of our reallocation process, where applicants get moved around to make sure everyone interviewed has a similar chance of being made an offer.” The stated purpose of moving candidates between Colleges is to equalise chances. You cannot equalise chances by moving people between Colleges if the Colleges are asking materially different things.

Set by the Subject or Set by the College: The Split, Row by Row

Here is the division as the two universities actually describe it, with the source for each line. Nothing in this table is inferred.

Every row on the left of that split is content. Every row on the right is logistics. The line between them is not a rhetorical flourish; it is where the published sources put it.

Where Colleges Genuinely Change Your Interview Day

It would be dishonest to pretend the College is irrelevant, and Cambridge says something that cuts against a tidy answer: “Each interview is different, depending on the course and College you’ve applied to.” That sentence is on the university’s own interviews page, and it names the College. So what is it referring to?

Everything Cambridge then goes on to itemise is arrangement, not argument. The College decides whether you are online or in the room. The College decides how many interviews you sit — most applicants have one or two, totalling thirty-five minutes to an hour, while some sit three or four. The College tells you whether pre-reading is coming, and when. The College runs the adjustments process if you need one, and the College is where the travel support sits.

Trinity College Cambridge, which publishes unusually detailed guidance, draws the same line explicitly for its own applicants. Having explained that UK-based candidates interview in person and overseas candidates may choose, it writes: “The distinction only affects the format of the interview. The type of questions you will be asked and the level of assessment will be the same whether you are interviewed in person or online.” A College with every incentive to talk up its own distinctiveness instead tells applicants the assessment is invariant.

Trinity does specify kit, and it specifies it by subject: candidates in “Chemical Engineering & Biotechnology, Computer Science, Engineering, Mathematics, Medicine and Natural Sciences” need a graphics tablet. Oxford does the same thing centrally, sorting courses into three technology tiers — tier 1 needs only video conferencing, tier 2 adds a shared whiteboard, tier 3 adds handwriting capture through a stylus. Oxford is careful that in tier 2 interviews “you will not be asked to write out complex mathematical notation using the whiteboard”. The kit follows the subject in both institutions, even though each draws its own subject boundaries.

If you want the College-level format detail for Cambridge specifically, we keep a separate page on Cambridge interview format by College for 2027 entry. This page deliberately does not duplicate it, because format is exactly the thing that does vary and it deserves its own treatment.

Does Your Interviewer’s Research Area Decide the Question?

This is the sophisticated version of the worry, and it is the one repeated most confidently by preparation sites: choose a College whose tutors research what you love, because their research will surface in the questions. It sounds plausible. It has no support in anything either university publishes.

What Oxford actually says about the source of the material is this: “The discussion develops in response to your answers and, in some subjects, may draw on aspects of your application such as your UCAS personal statement or submitted written work.” And on what you may be handed: “Depending on what is relevant for the course you are applying for, you may be given a text, a poem, a graph, or an object, and then asked to answer questions and comment on it.” Note what governs the choice — what is relevant for the course, and what is in your application. Not what the tutor published last year.

The published sample questions bear this out in a way that is easy to check. Read the Biology entries and you find a question about why habitats differ in biodiversity and a question about why animals have stripes, contributed by tutors at two different Colleges. Neither is a specialist probe into that tutor’s research programme. They are deliberately open problems that a strong sixth-former can reason about from A-level foundations, because that is what the interview is for. Oxford states the aim plainly: tutors want to see “your ability to think independently and to engage with new ideas beyond the scope of your school or college syllabus”.

Choosing a College by its tutors’ research interests may be a fine way to pick where you want to live and be taught for three or four years. It is not a way to predict your interview. If you are weighing Colleges for other reasons, our guide on how to choose an Oxford or Cambridge College deals with that decision on its own terms.

What Parents Say

DP

Deborah Prior

★★★★★

Confidence and progress

My 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.

NI

Nicola

★★★★★

Amazing tutors!

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.

S

Saifan Khan

★★★★★

Great Teaching and Focused Learning

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.

Why No Honest College-by-College Question Bank Exists

There is a structural reason you will never find a reliable list of “questions asked at Magdalen last December”, and it is not that nobody has bothered. Oxford requires every shortlisted candidate to accept an interviewees’ code of conduct, and the first clause reads: “I agree not to share in any form the content of interviews with others and including but not limited to candidates or teachers at my school/college or at any another school/college.” The clause covers discussing interview content verbally and disseminating any materials shared before or during interviews. Oxford adds that candidates “will be at serious risk of having any offer withdrawn” if they breach it.

So any College-specific question list circulating online is in one of two states. Either it comes from candidates who broke an undertaking they gave in writing, in which case it is unverifiable and unrepresentative — a handful of half-remembered fragments from a fortnight of thousands of interviews. Or it is reconstructed from the public, subject-filed sample questions, in which case the College label on it is decoration.

The code of conduct also bans researching pre-interview materials online, and bans using AI tools during the interview. Taken together, these rules tell you what the universities think preparation legitimately is: rehearsing the reasoning your subject demands, not memorising a College’s alleged repertoire. Our page on what Oxbridge interviewers actually mark goes into the assessment machinery behind that judgement.

How Our College Pages and Subject Pages Fit Together

We publish two families of interview pages, and readers occasionally wonder why both exist. The answer is now straightforward, and it follows the split above.

That is also why our interview question packs are organised by subject rather than by College. The division that decides what goes in a pack is always the subject, never the College — the whole Oxbridge interviews hub is built the same way. Some subjects run to more than one pack where the material genuinely splits, engineering and maths among them; none of them splits by College. Each pack contains ten questions with the reasoning worked through step by step rather than a model answer to memorise, and nine subjects have a free sample so you can see the format before deciding. Buying by College would be selling you a distinction the universities do not make.

Preparing for the Subject, Not the Porters’ Lodge

The practical consequence is a reallocation of your remaining preparation time. Hours spent hunting for what a particular College asks are hours spent on a variable that does not exist. Hours spent rehearsing the reasoning your subject demands are hours spent on the only thing every source agrees is being assessed.

Three things follow. First, prepare as though you might be interviewed anywhere, because at Oxford you may well be — reallocation and additional interviews at a second College are routine, and Oxford aims to give at least twenty-four hours’ notice of an additional interview. Second, get the logistics right for the College that actually invited you: the format, the kit, the pre-reading window. Those are College-set, they are knowable in advance, and getting them wrong costs you real minutes on the day. Third, practise saying your reasoning out loud, which is the one skill that transfers across every College, every format and both universities. We have a separate guide on how to think out loud in an Oxbridge interview if that is the part you find hardest.

Cambridge’s main interview period runs from 7 December to 18 December 2026, with winter pool interviews around mid to late January 2027, and Oxford interviews online in December 2026. Between now and then, the useful question is not which College you drew. It is whether you can hold a genuine academic conversation about your subject with somebody who has spent their career in it — because that is the conversation you will have, whichever staircase it happens on.

Frequently Asked Questions