Casper versus the MMI: what each one is actually measuring

From 2028 entry, a University of Auckland clinical applicant sits both of these and they are weighted equally: the MMI fair score at 50 percent with a minimum threshold applying, and Casper at 50 percent. They sound like the same thing. Both show you a scenario, both ask what you would do, neither has any content to revise.

They are not the same thing, and the differences are the kind that change what practice is even for.

The short version

Difference one: you are marked against a rubric, or against your peers

The MMI has a published mark scheme. Eight attributes, three assessed per station, and although you are not told which three, you can read all eight in advance. It is criterion referenced: there is a description of what good looks like and reviewers are marking against it.

Casper is not. Acuity describes scores as standardised to represent the relative rankings of each applicant compared to the applicant pool, and as z-scored relative to the strength of your peers’ responses. Scores are grouped into four quartiles.

So on the MMI, an answer that hits the attributes is a good answer. On Casper, an answer is good relative to what everyone else wrote about the same scenario on the same day. There is no bar to clear, only a distribution to sit somewhere in.

Difference two: speaking under a long clock, or writing under a short one

Three minutes of speech is roughly 400 words. Three and a half minutes of typing, split across two questions, is a lot less than that for most people. The MMI rewards having enough to say and pacing it; Casper rewards getting the reasoning down fast and not polishing it.

Acuity is helpful about that last part: raters are instructed to focus on the substance and quality of your response rather than issues like grammar or spelling. So on the typed section, proofreading is time you did not need to spend.

The MMI has the opposite hazard. UoA tells you to aim to use the full three minutes and to bring your thoughts to a close before the timer runs out. Finishing early is not efficiency there, it is an incomplete answer. How to actually fill three minutes is here.

Difference three: health scenarios, or everyday ones

The MMI’s seven domains are published and six of the seven sit squarely in a health or personal-professional context: equity, moral and ethical reasoning, career choice, a healthcare issue, personal insight, personal experience, and Hauora Māori. You can prepare the territory even though you cannot prepare the questions.

Casper’s scenarios are largely not medical. They are situations a person might meet in ordinary life, and they are designed to elicit how you think rather than what you know about healthcare. Acuity’s position is that you do not need anything to prepare beyond their free practice test and the information on their site.

Difference four: one of them exists, and one of them does not yet

This is the practical one. The Auckland MMI is documented down to the ten second transition between stations. Auckland’s Casper is not documented at all: UoA has not published which format variant it has bought, when it is sat, or what it costs, and as of 19 August 2026 the University of Auckland did not appear in Acuity’s own school listing.

Which means everything on this page about Casper describes Acuity’s standard product. Acuity itself says format may vary for select programmes and that test content changes every year. We keep a running page on what is known and what is not, and a list you can join for when it changes.

What actually transfers between them

More than the differences suggest, which is the encouraging part.

• Seeing more than one side. Both reward it. Neither rewards a confident answer that never acknowledged there was a decision to make.
• Saying who is affected and why. The habit of naming the people a decision lands on, rather than only the decision, is the same habit in both.
• Committing. Both are open-response and both are marked by humans who need something to weigh. Refusing to land is not neutrality, it is a blank.
• Working under a clock without freezing. Neither is generous with time and both are one take.

What does not transfer is the delivery half. Nothing in Casper trains you to speak to a lens for three minutes with nobody nodding, and nothing in the MMI trains you to write two coherent answers in three and a half minutes. They are separate skills and each one needs its own reps.

So what should you do about it

If you are applying for 2028 entry, the honest answer is: work on the half that exists. The MMI is published, stable, and practisable today. Casper is a name, a weighting and a promise until UoA publishes more, and preparing for a format nobody has confirmed is not preparation, it is guessing.

And for Casper specifically, do what Acuity says: take the free practice test in your own account. Acuity states in writing that it is not affiliated with any preparation provider and does not endorse their use, and that people who use third-party materials do worse than people who use its free resources. We are a third party, and we are still telling you that.

Start with what the MMI actually is, or read what changed for 2027 and 2028 entry.