Why the process feels arbitrary
Almost every applicant arrives at their first interview holding two beliefs that contradict each other. The first is that an interview is a personality test you either pass or fail. The second is that there is a hidden set of correct answers, and the people who get offers are the ones who found them. Neither is true, and most wasted preparation happens in the gap between them.
What is actually happening is duller and far more learnable. A UK medicine or dentistry interview is an assessment with a mark scheme. Somebody decided in advance which qualities each station measures, wrote descriptions of what a weak, adequate and strong response looks like, and trained an interviewer to score you against them. You never see any of that. You see a stranger with a clipboard and a timer.
That invisibility is the whole problem. When you cannot see the criteria, every judgement looks like taste, and preparation turns into guesswork about what a particular interviewer might like. Once you can see the criteria, the process stops being a lottery and becomes something you can train for deliberately. That is what this reading path is for, and this is its first stop.
The six domains behind almost every station
Schools word their criteria differently, weight them differently, and only some publish them openly. Look across enough of them and the same six areas keep surfacing, because they all trace back to one endpoint: the school has to produce a graduate who is safe to start work as a foundation doctor or a dental foundation trainee, and the regulators — the General Medical Council for medicine, the General Dental Council for dentistry — set out what that graduate must be able to do. Selection is designed backwards from that.
- Motivation and realistic insight into the career. Not how much you want it, but whether you understand what you are signing up for: the training length, the shift patterns, the administrative load, the genuinely unglamorous parts. Describing the job accurately beats describing it warmly.
- Values and ethical reasoning. Whether you notice that a situation has an ethical dimension at all, can name the interests pulling against each other, and can reason between them without collapsing into a slogan.
- Communication, including listening. Clarity, pace, plain language instead of jargon, checking you have been understood — and, in role play, whether you answer what the other person actually said rather than delivering a script over the top of them.
- Teamwork and leadership. Evidence that you can work inside a group where you are not in charge, and that you understand leadership as coordination and responsibility rather than as being the loudest person present.
- Resilience, self-awareness and reflection. Whether you can describe something that went badly, say what you took from it, and show that you changed something as a result. Widely marked, and rarely prepared for.
- Awareness of healthcare as a system. That medicine happens inside an organisation with finite money, competing priorities, a strained workforce and a legal framework around it. You are not expected to be a policy analyst, only to avoid talking as though resources were infinite.
Two of these deserve separating out, because candidates consistently under-prepare them. Listening is scored as a skill rather than as a courtesy, and in a role-play station a marker can usually tell very quickly whether you are processing what you hear or performing an answer you brought with you. Systems awareness is the other: it is what turns a pleasant answer into a mature one, and it turns up uninvited in stations that look like they are about something else entirely.
The values domain is also broader than most applicants assume. It is not a memory test on the four pillars, but you cannot reason well without them, which is why the four pillars of medical ethics sits at the centre of stage three rather than being treated as an optional extra.
Dentistry interviews run on the same six domains with a different centre of gravity. Motivation carries the extra weight of why dentistry rather than medicine, manual dexterity appears as its own line of questioning at some schools, and the systems domain takes in the mixed National Health Service and private reality of general dental practice. Everything else on this path applies unchanged.
How a station is actually marked
Take a typical multiple mini interview. You stand outside a door, read a short prompt, a bell goes, and you have somewhere in the region of five to ten minutes with one interviewer — or with an assessor and an actor, in a role-play station — before the bell moves you on. Circuit lengths and timings vary considerably between schools and are usually set out in your invitation or on the admissions pages, so check yours rather than assuming a national standard.
Behind that door, the interviewer is holding a mark sheet written for that station. It lists the domains the station is designed to probe — often two or three, not all six — with a scale for each and short descriptions of what each point on the scale looks like. Many schools also ask for a global rating: one overall judgement of the candidate at that station, which works as a sanity check on the itemised scores.
- Before the cycle opens, each station is mapped to the domains it will test, so what it measures is fixed rather than improvised on the day.
- You answer. The interviewer listens against the written descriptions and marks what you said, not what they infer you probably meant.
- Each domain in play at that station receives a score on its scale.
- A global rating for the station is often added, and the station total is recorded.
- Station totals are combined across the circuit, sometimes with weightings, into the single number ranked against every other candidate.
The consequence of that last step is the most useful thing you can know on the day. In most MMI circuits the interviewer at station four has no idea what happened at station three. They have not spoken to the previous marker, they frequently have not seen your application, and they are scoring the few minutes in front of them. A station that goes badly is a contained loss.
Each station starts from zero
A marker who cannot see your other stations cannot be prejudiced by them. One poor station lowers one station score; it does not follow you down the corridor. The candidates who lose most from a bad station are the ones who carry it into the next three, which is the most common way a single station ends up costing far more than the station itself was worth.
- Station
- One timed task in an MMI circuit, with its own prompt, its own interviewer and its own mark sheet.
- Circuit
- The full set of stations you rotate through, on a fixed timer with a short reading period at each door.
- Rubric
- The mark scheme for a station: the domains scored, plus written descriptions of what each score looks like.
- Global rating
- One overall score for a candidate at a station, recorded alongside the itemised domain scores.
- Blueprinting
- Mapping stations to the domains they test before the cycle opens, so a circuit covers everything once and nothing twice by accident.
Panel and traditional interviews differ in one respect that changes your tactics. The same two or three people are with you for the whole conversation, so an early wobble stays in the room longer — but you also have time to build, and a panel can circle back to a point you fumbled. The domains being scored are largely the same; what changes is how the evidence is gathered. The formats guide sets out which schools run which.
Why the rubric rewards structure and evidence, not enthusiasm
Enthusiasm on its own earns very little, and that explains a great many surprising rejections. Motivation is scored, but as understanding of the job rather than as appetite for it; there is rarely a box on the sheet for how badly you want it, because wanting it does not distinguish you from anyone else in the corridor. What is scored is whatever the written descriptions describe, and those are almost always phrased in observable terms: did the candidate give a specific example, did they consider more than one perspective, did they acknowledge a limitation, did they check the other person had understood.
This is why structure earns marks out of all proportion to the effort it takes. If you signpost — there are two things pulling against each other here, so let me take them in turn — you have just told the marker where to look on their sheet. An organised answer is not merely nicer to listen to. It is easier to score highly, because the evidence for each domain is visible instead of buried.
Evidence means specifics: a named situation, your actual role in it, what you did, what happened, what you took from it. The claim “I have strong communication skills” cannot be ticked against anything, because it is not observable. “On my care home placement I learned to give a resident with advanced dementia one instruction at a time, because three at once made her anxious” can be.
Why do you want to study medicine?
Answer in three moves and evidence each one. First the origin: a specific, undramatic reason you started looking at medicine, in a sentence or two rather than a paragraph. Second the test: what you did to check the interest was real, and what you saw that confirmed or complicated it — name the placement, the volunteering, the clinician who gave you the honest version. Third the fit: what about the work itself suits you, phrased as a claim someone could disagree with. “I want to help people” scores nothing because it is equally true of teaching, nursing and social work. “I want the long-term relationship a GP has with a family, which I watched across three clinics with the same patient” is a claim about the job. Full treatment in why medicine, without the clichés.
The marker can only tick what you said out loud
Implied insight scores nothing. If you understood that the nurse was de-escalating the relative rather than simply being patient, but all you said was that the visit was interesting, the descriptor stays unticked and you will never know. State the meaning of your example rather than leaving it to be inferred, and state it early — do not save it for a closing sentence the timer takes from you.
Insight is the word that decides the close calls
Read enough selection criteria and one word does more work than any other: insight. It is also the least clearly explained. In practice it means the distance between what happened and what you understood about it — and it is usually what separates a good candidate from an outstanding one, because it cannot be borrowed from anybody else.
Three candidates describe the same afternoon on a stroke unit. The first lists what they saw. The second says it was inspiring and confirmed their desire to study medicine. The third says they watched a consultant spend most of a family conversation not on the diagnosis but on what the patient would have wanted, and that they had not appreciated how much of the work is negotiating uncertainty with people rather than resolving it. Only the third has produced something a marker can score, and it is not because the third candidate saw more.
Insight has a second face pointed at yourself rather than at the work, and that is self-awareness. A candidate who can name a real limitation, describe what it has cost them and explain the specific thing they now do differently is demonstrating exactly the reflective habit that clinical training will demand of them for the next decade. A candidate whose weakness is caring too much is demonstrating that they did not try.
Tell me about a time you were given feedback you did not agree with.
The trap is spending the answer proving you were right. The domain being scored is reflection, not vindication. Set the scene in a sentence, state the feedback plainly without softening it, then say honestly what your first reaction was — most people bristle, and admitting that is not a weakness here. Then make the move that earns the mark: explain how you tested whether the feedback was fair, whether by asking a second person, watching what happened next time, or looking at the outcome rather than your intention. Land on what you kept and what you changed, and be willing to say it turned out to be partly right. If your work experience notes are still a list of places you stood, turning work experience into insight is the conversion process.
They score how you think, not the answer you land on
Ethics and dilemma stations show this most clearly, and they are where confident candidates most often lose marks. The mark sheet is not checking whether you reached the conclusion the school prefers. It is checking whether you identified the competing interests, whether you gave the opposing case its strongest form rather than a straw version, whether you noticed what information you were missing, and whether you were willing to commit to a position at the end while acknowledging what that position costs.
That final part matters more than most candidates expect. Balance is not the same as refusing to answer. Listing both sides and then trailing off completes half the task and scores accordingly. The strongest answers reason across the tension and then land — tentatively, provisionally, with the reason for landing there stated out loud.
A worked example: should a patient’s lifestyle affect their access to treatment?
The case for taking it into account
- Some procedures carry more risk or work less well while a behaviour continues — a clinical argument about likely benefit, not a moral judgement about the person.
- Resources are finite, and a system that ignores expected benefit entirely produces less health from the same budget.
- A short, supported delay paired with help to stop smoking or lose weight can improve the outcome of the operation itself, so it is not purely a rationing device.
The case against
- Behaviour is shaped heavily by circumstances people did not choose — deprivation, addiction, chronic pain, mental illness — so penalising it tends to penalise disadvantage.
- The principle has no clean edge. Once lifestyle counts, it is unclear why it stops at smoking and weight rather than extending to sport injuries or diet.
- The NHS Constitution frames access around clinical need rather than ability to pay, and rationing by perceived deservingness sits awkwardly beside that.
Both columns are markable. Reaching either conclusion is markable. What is not markable is choosing a column in your first sentence and defending it for four minutes, because that tells the interviewer nothing about how you handle a problem where the answer is not already in your head — which is the entire situation you are being selected for.
Where this sits on the path
This is stage one of five. The path is ordered by where you are in the journey rather than by topic: foundations first, then the professional standards and reflective habits of stage two, then ethics and law, then performance under time pressure, and only at the end the frontier conversations about artificial intelligence, genomics, gene editing and assisted dying — which need all four earlier stages underneath them if they are going to sound like anything more than opinions.
The next stop is how the NHS actually works. It is the least glamorous article on this path and the one that quietly improves the most answers, because the systems domain has a habit of appearing in stations that look like they are about something else.