Skip to content

Medical interview practice · Australia

Read a station in silence. Answer it on the clock. Read back what you actually said.

Your first station is read by two AI readers, no card.

2:00 · 7:00Queensland (UQ)

Your answer stays in this browser. Nothing is sent, nothing is saved, and nobody reads it.

Continuing with Google creates your free account — or signs you into the one you already have — and your first station is read by two AI readers, with no credit card and nothing charged. By continuing you agree to the Terms and the Privacy Policy, and to receive emails about your account: practice reminders, streak nudges and a weekly progress recap (unsubscribe in one click).

Sign in, pick your school, and the first station opens.Use email instead— the sign-up form asks for its own agreement.

Show another school’s published station clock
QUEENSLAND (UQ)2:00 · 7:00
  • READ
  • RESPOND
  • ONE-MINUTE REMINDER
13
Programs read
every Australian graduate-entry program, including the one that runs no routine interview
20
Primary sources
each one a university or GEMSAS page, linked on the card it feeds
11 Sep 2026
Last read
the newest verification date on any cell above
54
Stations live
written in-house, each one signed off before it can be dealt
5
Anchors published
the whole scale, in words, on this page
Reader agreement
Agreement between the readers has not yet been measured.
n = 0
Calibration
No reading has been checked against a real interview yet: n = 0. We will publish what we find at n = 30, whichever way it goes.

These pages cover Australian graduate-entry medicine. UK and Irish programs are not read here yet — the record we publish is the one we have verified at the source.

One of our stations, standing

What an MMI station actually looks like

We don't take stations from students — sending us one could cost you your offer.

Station · readingrecorded · no probes

The seat at the community lunch

A hall in a country town runs a lunch every Wednesday for people who live alone. There are twenty-two seats and a waiting list of nine. For a year, Lorraine has come each week with her mother. Her mother has now moved into a care home in the regional centre, an hour away, and Lorraine drives out to see her on Wednesday afternoons. Lorraine rings you, because you keep the list, and asks whether she can keep coming on her own. She says the drive back is the hard part, and Wednesday is the only day she speaks to anyone. The first name on the waiting list is Peter. He has been waiting four months, and he has rung twice in that time to ask where he is up to. There are no spare seats, and no spare volunteers to run a second sitting. Record your answer to two questions. What do you say to Lorraine on this call? What do you then say to Peter?

2:00 to read7:00 to answer
  • Empathy
  • Communication
  • Equity

What the assessor reads this for — our mapping. Pick your school and the same station is read against the words that school publishes.

The marked transcript

What a marked answer looks like when it comes back

The answer below is ours, recorded as a demonstration; no student wrote it. Both readings are ours too — neither is a student's, and neither came from a live seat.

The answer, as typed

I’d start by telling Lorraine I’m glad she rang, and that I can hear this is about more than the meal.

Wednesday is the only day she speaks to anyone — I’d say that back to her before I say anything about the list.

Reader 1Empathy — observed

I can’t promise her the seat today, because the list is what makes this fair and Peter has waited four months.

Reader 2Chooses without naming the cost — partly

I’d ask what the drive back is like, and whether a fortnightly seat, or a lift with another guest, would carry some of it. Then I’d ring Peter myself rather than let him hear it second-hand: tell him he is next, give him a date, and say plainly who else asked and what I decided.

Reader 1

Position and a reason; most people named; one thing unsaid.

Reader 2

Between the anchors for 5 and 7.

15967

Two readers, one anchor apart — this is what rater variance looks like.

An example of the feedback you receive. Not affiliated with ACER.

What a stronger answer does

It says the cost out loud before it chooses: what Peter loses by waiting another week, and what Lorraine loses if the seat goes, in her own words back to her.

Your school's interview

What each Australian medical school publishes about its interview

A dashed rule means not published.

Every minute bar is drawn on one shared scale, so the same length always means the same minutes. All 12 verified 11 Sep 2026.

What each Australian graduate-entry medical school publishes about its interview. A filled cell is a fact the school publishes; a dashed cell is one it does not.
SchoolFormatStationsStation clockModeWindowOffer weightCasperQualitiesHoles
Queensland (UQ)Format: MMIStations: eightStation clock: two to read, seven to respondMode: onlineWindow: MMIs will be held online in September 2026.Offer weight: not publishedCasper: not usedQualities: kindness and compassion, integrity, adaptability, verbal communication skills1
MelbourneFormat: MMIStations: eightStation clock: about five per station, split not publishedMode: not publishedWindow: not published. September/October 2026 - Interviews held (GEMSAS sector calendar; the school's own dates are not published)Offer weight: 50% of the offer rankCasper: not usedQualities: cultural sensitivity, maturity, collaboration, reliability, communication skills2
DeakinFormat: MMIStations: sixStation clock: about five per station, split not publishedMode: onlineWindow: Interviews will be held online in September 2026.Offer weight: 50% of the offer rankCasper: not usedQualities: communication skills, commitment to rural and regional practice, evidence-based practice, self-directed learning, teamwork, motivation for a career in medicine, a commitment to social justice, professionalism, effective use of resources, health promotion0
GriffithFormat: Recorded interviewStations: not publishedStation clock: not publishedMode: recordedWindow: between 9am Monday 21 September and 5pm Friday 25 September 2026 (AEST). You will choose the time and dateOffer weight: 50% of the offer rankCasper: not usedQualities: good communication skills, an empathetic and ethical approach, a reflective manner, problem-solving skills, teamwork, decision-making skills2
Wollongong (UOW)Format: MMIStations: eightStation clock: not publishedMode: onlineWindow: week commencing 17 September 2026.Offer weight: 70% of the offer rankCasper: requiredQualities: communication skills, empathetic and ethical approach, reflective manner, teamwork, decision-making skills in clinical and community contexts, delivery of health care to rural, regional and remote communities, Aboriginal and Torres Strait Islander health1
Notre Dame (Syd)Format: Recorded interviewStations: not publishedStation clock: not publishedMode: recordedWindow: not published. September/October 2026 - Interviews held (GEMSAS sector calendar; the school's own dates are not published)Offer weight: 50% of the offer rankCasper: requiredQualities: not published4
Notre Dame (Frem)Format: Recorded interviewStations: not publishedStation clock: not publishedMode: recordedWindow: not published. September/October 2026 - Interviews held (GEMSAS sector calendar; the school's own dates are not published)Offer weight: 50% of the offer rankCasper: requiredQualities: not published4
Western Australia (UWA)Format: MMIStations: eightStation clock: two to read, five to respondMode: onlineWindow: Online interviews are expected to take place within the period 22 September to 3 October.Offer weight: 50% of the offer rankCasper: not usedQualities: accountability, ethical and moral judgement, communication skills, management skills, problem solving, self-awareness, teamwork/collaboration, conflict resolution, empathy, resilience0
ANUFormat: Two-part interviewStations: not publishedStation clock: not publishedMode: onlineWindow: Interviews will be held in two parts from Friday 4th to Friday 11th September 2026. All interviews are held onlineOffer weight: 50% of the offer rankCasper: not usedQualities: not published3
FlindersFormat: Panel interviewStations: not publishedStation clock: not publishedMode: in personWindow: SA: Monday 21 September to Friday 2 October · NT: Monday 21 to Thursday 24 SeptemberOffer weight: not publishedCasper: not usedQualities: Quality of motivation, Learning Style, Communication skills, Pro-social attitude, Approach to decision-making3
Macquarie (MD)Format: MMIStations: not publishedStation clock: not publishedMode: in personWindow: 26 September 2026 — Interviews with Macquarie University. Note: 2026 interviews will be conducted in personOffer weight: 50% of the offer rankCasper: not usedQualities: not published3
Monash (grad)Format: MMIStations: not publishedStation clock: not publishedMode: onlineWindow: 8, 9 & 10 September 2026 — Interviews held on Zoom.Offer weight: 60% of the offer rankCasper: not usedQualities: not published3

26 of 96 cells are holes: facts the school does not publish anywhere we could read.

The same facts in full, with the page each was read on

Deakin

MMI · online

6 stations

5 min a stationRead and respond split not published
50% of the offer rank

Interviews will be held online in September 2026.

See every school in full

ANU

Two-part interview · online
Stations not published
Station timing not published
50% of the offer rank

Interviews will be held in two parts from Friday 4th to Friday 11th September 2026. All interviews are held online

Sydney (USYD)

No routine interview for the standard domestic stream.

  • “the University reserves the right to use an additional selection tool, such as an interview or Situational Judgement Test (SJT).”
  • “to assess their interest, intention and aptitude for practising medicine in rural and regional areas.”

What no Australian school publishes: its rubric, its scale, its anchors, the weight of any one station, or a pass mark.

Your first station is read by two AI readers, no card.

The instrument

No school publishes its marking scheme. This one is Aptavia's own, and it is published here.

  1. 9

    Position early, people named, cost stated, holds or changes under the probe for a reason said aloud.

  2. 8
  3. 7

    Position and a reason; most people named; one thing unsaid.

  4. 6
  5. 5

    A position by the end; some people named; meets the probe in part.

  6. 4
  7. 3

    Surveys without choosing, or chooses without a reason; hands it up the line.

  8. 2
  9. 1

    Does not engage with the situation as set.

A dashed rung sits between two anchors. The scale is ordered, not measured.

The ladder walked

Written by us as a demonstration; no student wrote it; no reader read it.

The seat at the community lunch

A hall in a country town runs a lunch every Wednesday for people who live alone. There are twenty-two seats and a waiting list of nine. For a year, Lorraine has come each week with her mother. Her mother has now moved into a care home in the regional centre, an hour away, and Lorraine drives out to see her on Wednesday afternoons. Lorraine rings you, because you keep the list, and asks whether she can keep coming on her own. She says the drive back is the hard part, and Wednesday is the only day she speaks to anyone. The first name on the waiting list is Peter. He has been waiting four months, and he has rung twice in that time to ask where he is up to. There are no spare seats, and no spare volunteers to run a second sitting. Record your answer to two questions. What do you say to Lorraine on this call? What do you then say to Peter?

This station is a recorded one and puts no probe, so these three answers are read on position, people and cost alone — the part of each anchor that is about what an answer does when it is pushed is not on the page here.

Surveys without choosing, or chooses without a reason; hands it up the line.

There are good arguments on both sides here. Lorraine has been coming for a year and clearly gets a lot out of it, and Peter has been waiting four months, which is also a long time. The fair thing is to follow the policy, because a list only works if everyone is treated the same. I understand how she feels, and I would tell her that I will take it to whoever runs the lunch and see what they say. I would let Peter know that we are looking into it and will be in touch.

  • “I will take it to whoever runs the lunch and see what they say” — not observed Lands a decision of her own
  • “a list only works if everyone is treated the same” — not observed Gives a reason about these people, not a rule
  • “I understand how she feels” — not observed Empathy enacted, not only declared

Two people are waiting for an answer and neither of them is given one. The rung above it reaches a decision before the turn ends, even if it reaches it late.

A position by the end; some people named; meets the probe in part.

I’d start by letting Lorraine say what Wednesdays mean to her now that her mother has moved, because I wouldn’t want to decide anything while she’s upset on the phone. I’d explain that there are nine people on the list and that Peter has been waiting four months, so I can’t simply keep her seat open. In the end I think the seat has to go to Peter, because that’s what the list is for. With Lorraine I’d look at whether there’s anything else on a Wednesday she could come to. Then I’d ring Peter and tell him he’s in.

  • “In the end I think the seat has to go to Peter” — observed Reaches a position by the end of the turn
  • “there are nine people on the list” — partly Names some of the people affected
  • “that’s what the list is for” — not observed Names the cost of the choice

The decision arrives, and it arrives last and as a rule: what losing Wednesday costs Lorraine is never said to her. The rung above it says the cost out loud, to the person carrying it.

Position and a reason; most people named; one thing unsaid.

I’d tell Lorraine the seat is hers this Wednesday, and say why: she has an hour’s drive back on her own afterwards, and Wednesday is the only day she speaks to anyone. Then I’d say the part she won’t want to hear — that it can’t be hers every week while Peter is first on a list of nine and has rung twice to ask. I’d ask whether a fortnightly seat would carry some of it. Then I’d ring Peter myself, before he hears it from someone else in town: tell him he is next, give him the date he starts, and say plainly that this week the seat went to someone else, and why.

  • “I’d tell Lorraine the seat is hers this Wednesday, and say why” — observed Takes a position and gives a reason
  • “Peter is first on a list of nine and has rung twice to ask” — observed Names most of the people affected
  • “before he hears it from someone else in town” — observed Tells the person it costs, in words, before he hears it elsewhere
  • “whether a fortnightly seat would carry some of it” — partly One thing goes unsaid

It decides, gives the reason to the person it costs, and rings Peter before the town does. What it never says is what a fortnightly seat does to the eight people behind him — the one thing left unsaid, which is exactly what the anchor at seven describes.

The ladder stops at seven. A nine written by us would be a model answer, and the instrument refuses those — the anchors describe what an answer does, not the answer.

Read the five anchors in full

  • Takes a position early and says why; names the people the situation affects and what each stands to lose; states the cost of her own choice; when the probe pushes, either holds the position for a reason she says out loud or changes it for a reason she says out loud.

  • Takes a position and gives a reason; names most of the people affected; answers the probe directly; one thing goes unsaid — the cost of the choice, or one person the choice lands on.

  • Reaches a position by the end of the turn; names some of the people affected; meets the probe in part — restates the first answer rather than engaging the push, or engages without it changing anything she says.

  • Surveys the situation without choosing, or chooses without a reason; names few of the people affected; answers the probe with a rule or a general principle rather than a reason about these people; or hands the decision up the line as the end of her thinking.

  • Does not engage with the situation as set: no position, no one named, the probe unanswered, or an answer to a different question than the one asked.

One station is the least reliable unit in an MMI; the real thing averages eight or more, and a reading of one station is a reading of one station.

Two AI readers, from two different engine families, read what you said — more readers than the real interview gives you, and none of them is a person.

  • Two readers is more than the real interview gives you — one assessor, one station. Two here is a reliability device, not fidelity.

  • Agreement between the readers has not yet been measured.

  • This reading is not calibrated against real interview results yet. When enough students tell us how it went, we will publish what we find — including if it is bad.

“Coaching was ineffective in improving MMI scores, with coached candidates actually having a significantly lower score”
Griffin, Harding, Wilson & Yeomans, Medical Journal of Australia 189(5):270–273 (2008)

That is one Australian cohort at one university in one cycle, and it is the only Australian evidence there is. It is not proof that nothing can help — it is the reason nothing here claims to.

What the evidence says

Does a bank of last year's questions help?

Two studies asked almost exactly that, in real admissions cycles, and published what they found.

Half of 384 applicants in a high-stakes twelve-station circuit were given a station's stem two weeks in advance. It made no measurable difference to how they scored.

“No statistically significant differences in MMI performances were detected. … Predictable violations of MMI security do not unduly influence applicant performance ratings”
Reiter, Salvatori, Rosenfeld, Trinh & Eva, Medical Education 40(1):36–42 (2006)

Sitting interviews at OTHER programmes produced no improvement at all; what gain there was came from re-sitting the same programme, and the authors say plainly what they think it is.

“probably not attributable to improved ability”
Griffin, Bayl-Smith, Duvivier, Shulruf & Hu, Medical Education 53(2):175–183 (2019)

We don't take stations from students — sending us one could cost you your offer. That is a rule we would keep anyway, because the schools bind candidates to confidentiality — but the evidence above is why it costs nothing to keep.

What it costs

What one station costs

Station 1

Read in full by two AI readers.

Station 2

Your words stay sharp. The reading is veiled.

Your first station is read by two AI readers, no card. The next station's reading is produced and shown veiled — your own work, not somebody else's sample — and then the plan decides.

This will not raise your interview score — the only Australian study of coaching found none. What this buys is rehearsal on the clock your school publishes — or UQ's, named on the screen, where your school publishes none — and a transcript of what you actually said, which no school gives back.

Two AI readers, from two different engine families, read what you said — more readers than the real interview gives you, and none of them is a person.

The Station is part of Pro today

$59/month

Every station and every circuit, on your school's clock, with the marked transcript and the days-left plan

Spoken stations are built and switched off. When they are on, this page will say so first.

Every plan and price is on the pricing page.

Your first station is read by two AI readers, no card.

Questions

Fair questions about MMI practice

Will practising here raise my interview score?

No, and nothing on this page will say otherwise. Coaching was ineffective in improving MMI scores, with coached candidates actually having a significantly lower score — Griffin, Harding, Wilson & Yeomans, Medical Journal of Australia 189(5):270–273 (2008). What this buys is rehearsal on the clock your school publishes — or UQ's, named on the screen, where your school publishes none — and a transcript of what you actually said, which no school gives back.

Whose clock does a station run on?

Your school's, where your school publishes one. Where it does not, the station runs on Queensland (UQ)'s published clock — two minutes to read the scenario in silence and seven to respond — and the screen says so by name rather than presenting a number we chose as your school's.

Do I get a mark, and where does it come from?

You get a rating on a scale that is ours, published in full on this page, with a written anchor for every rated step — not a school's mark and not a prediction of one, because no Australian medical school publishes its rubric, scale, anchors or pass mark. Two AI readers, from two different engine families, read what you said — more readers than the real interview gives you, and none of them is a person. They read the same answer blind to each other; if they disagree by more than one anchor, no number is published for that station and you see both readers' receipts side by side instead.

Are these real stations from a university?

No. Every station is written in-house. We do not take stations from students, and we ask you not to send us one — the schools that run these interviews bind candidates to confidentiality, and sharing a station could cost you your offer.

What does it cost to stand at a station?

Your first station is read by two AI readers, no card. The next station's reading is produced and shown veiled — your own work, not somebody else's sample — and then the plan decides. Beyond the floor the Station is part of Pro, and a free trial of Pro starts only when you tap to start it, with no card and nothing charged.

Is my answer recorded, and who can read it?

Your typed transcript is saved to your account so you can read back what you said; it is yours, it travels in your data export, and it is deleted with your account. We keep no recording, and with spoken stations switched off there is no audio at all — a station here is typed. Your answer is never shown to another student and never becomes a station.

My school says Casper is required — is that the same thing?

No. Casper is a separate online situational-judgement test, sat before the interview and run by another company; the schools that require it say so on their own pages and our cards carry that cell with its source. Nothing here is a Casper practice test, and no station on this page is scored the way Casper is.

Is there a free mock MMI I can do online right now?

Yes — on each school's page there is a station you can stand at in your browser, with no account and no sign-in: one of our own scenarios, run on that school's published clock where it publishes one and on UQ's, named on the screen, where it does not, with your own answer back at the buzzer beside the scale we publish. Nobody reads that one; a reading is what an account adds.

Do I need a Google account?

No. Google is one tap and no typing, but the ordinary sign-up form works too: you confirm your address with a 6-digit code, sign in, and land on the same station.

Related reading

Stand at one station and read back what you said.

Your first station is read by two AI readers, no card.

Continuing with Google creates your free account — or signs you into the one you already have — and your first station is read by two AI readers, with no credit card and nothing charged. By continuing you agree to the Terms and the Privacy Policy, and to receive emails about your account: practice reminders, streak nudges and a weekly progress recap (unsubscribe in one click).