Reference

Deutsch Français

OET Speaking: the Profession-Specific Role-Play

The speaking sub-test is a role-play in the candidate's own field. That single fact rules out most of what people instinctively prepare: rehearsed self-introductions, opinion topics, and the fluent monologue that general speaking tests reward.

What it asks for instead is a professional conversation that goes somewhere, which is both narrower and, for a working clinician, more familiar.

How to read this sheet: This page covers one sub-test. The role-plays come from your own profession, so the preparation section is arranged by profession rather than by skill.

Two seat markers facing each other across a table rule, joined by short alternating rules that step from one side to the other.

A role-play has a purpose

In a role-play the candidate is not demonstrating range; they are trying to accomplish something with another person. The shape of a professional exchange follows from that: establishing what the other party understands, delivering information they can act on, checking that it landed, and handling the moment where it does not.

This rewards a set of habits that fluent speakers do not automatically have. Pausing to confirm understanding looks like hesitation on a general speaking test and is competence here. Simplifying vocabulary looks like a limited range there and is appropriate here. The instinct to fill silence with elaboration, which serves a candidate well elsewhere, works against them in a conversation whose purpose is comprehension.

The awarding body's own framing supports reading it this way: the test is designed to assess English language ability rather than medical knowledge, with tasks built so they resemble the situations a candidate will actually be in. The role-play is not a proxy for a clinical exam.

Preparing for your own profession, not for healthcare in general

Speaking is profession-specific, so a pharmacist and a physiotherapist rehearse different conversations. Twelve professions have their own versions, Medicine, Nursing, Dentistry, Dietetics, Pharmacy, Physiotherapy, Podiatry, Radiology, Occupational Therapy, Veterinary Science, Optometry and Speech Pathology: and material that names none of them is material for a job nobody holds.

The most efficient preparation is usually the least exotic: take the three or four conversations that recur most in your own week and run them in English until the shape is automatic. The scenarios that actually appear are ordinary, because ordinary is what the job is made of.

There is a trap in over-rehearsing. A memorised script survives until the other party does something unexpected, which in a role-play they are meant to do. Practising the recovery — asking again, restating, checking: is worth more than practising the opening.

What regulators ask for in speaking

Unlike writing, speaking is an area where the two regulators compared here agree. The Nursing and Midwifery Council requires at least grade B, 350 or above, for reading, listening and speaking, dropping only writing to C+. The General Medical Council requires at least grade B in each testing area, speaking included.

So speaking has no discount anywhere in this comparison, which is worth knowing when allocating preparation time. A nursing applicant can be at the required level in writing while still short in speaking; the reverse arrangement, being strong in speaking and short in writing, is the one that sometimes still passes.

Because requirements are read per sub-test rather than as an average, a strong overall score does not compensate for a speaking result below the stated grade. And where a regulator permits scores from two sittings to be combined, it publishes conditions — including, in at least one case, a worked example of a combination it refuses.

Requirements by regulator

The day itself

Three delivery formats exist: on paper, on computer and at home. A candidate should practise in the one they will sit. A role-play conducted through a screen is a different physical experience from one conducted across a table, even when the task is identical: pauses read differently, and the cues that tell you the other person is about to speak are weaker.

The format also determines when the result arrives. The awarding body states results are typically available within 6 days for the computer and at-home versions and within 12 days for paper. Where a registration deadline is close, that difference decides which format is even available as an option.

One preparation habit is worth naming because it transfers directly. Recording a single role-play and listening back once is more informative than running six without review; the things that cost marks in a professional conversation, talking over the other party, answering a question that was not asked, leaving an instruction ambiguous: are obvious on playback and invisible in the moment.

Finally, the result reports an individual score for Speaking alongside the other three sub-tests plus an overall figure. The speaking number stands on its own in every requirement examined here, so it is worth knowing in advance which of the four is the one carrying the application, and if it is speaking, no amount of work on the other three will move the decision.

Common questions

What form does the speaking test take?

A role-play based speaking assessment specific to the candidate’s profession, as described by the awarding body.

Is it an interview about myself?

No. The task is a professional interaction, not a personal one. Preparation aimed at fluent self-description is aimed at the wrong thing.

Is speaking profession-specific?

Yes, along with writing. Listening and reading are shared across all twelve professions.

What grade do regulators want here?

Both regulators compared on this site ask for grade B in speaking: the Nursing and Midwifery Council as part of its B for reading, listening and speaking, and the General Medical Council as part of its B in every testing area.

Does the delivery format change the speaking test?

The test is offered on paper, on computer and at home. What changes measurably between formats is result turnaround, typically 6 days for computer and at-home against 12 days for paper.