How GAMSAT Section 2 Is Marked: What ACER Publishes

Search for a GAMSAT essay rubric and you will find a dozen confident scoring grids, each with bands and descriptors and a number attached. None of them are ACER's. ACER has not published one.

What ACER does publish about Written Communication marking is short enough to quote in full — and it is more useful than the invented grids, because it is what the examiners are actually asked to consider. Here is the official basis, what it rules out, and what to do with the gap.

The official basis is two sentences long

That is it. Two broad criteria: the thinking, and the language that carries it. Notice the order, and notice that the second is subordinate to the first — ACER says assessment focuses on how ideas are integrated into a thoughtful response to the task, not on prose quality admired on its own.

The two features considered

Under the heading "criteria for the assessment of written communication", ACER names two features of writing and what each covers. They map onto the two broad criteria exactly: thought and content is the thinking, organisation and expression is the control and use of language.

Four bullet points. That is the entire published specification for a section worth a quarter of the standard weighted overall, where Written Communication counts once and the sciences paper counts twice.

Each response is scored independently three times

This single line is the most reassuring fact in the booklet, and almost nobody quotes it. Your Written Communication result is not one reader's opinion on one morning. Independent repeat scoring is how a marking system controls for the variance that any human judgement of writing carries.

It also tells you something about what a single piece of feedback is worth. One reader's verdict on your essay — a friend, a tutor, a model — is one sample of a distribution. It is worth having, and it is not the same thing as a score.

The five questions the examiners consider

ACER goes one step further than the criteria and lists the questions its examiners bring to your essay. They are the closest thing to a rubric that exists, and they are all about what you made of the prompt.

Read alongside them, one more published line changes how the first five minutes should feel: test takers only have 30 minutes to plan and write each essay, and, in ACER's words, the markers do not have unrealistic expectations of what can be produced in that time.

What is not assessed

This is where most preparation advice quietly goes wrong. Three published statements between them retire a lot of received wisdom: you are not assessed on whether your ideas or attitudes are 'correct', control of language is not assessed in isolation, and — the one that surprises everyone — spelling and punctuation errors will not impact your scores.

None of that means writing carelessly is free. Language is assessed insofar as it contributes to the overall effectiveness of the response, so sentences that obscure the thinking still cost you. It means the marks are not sitting in a proofreading pass, and the anxious final minute is better spent on the argument than on commas.

The memorised essay is a live risk, not a myth

ACER states that memorised essays generally score poorly, that pre-prepared responses and responses that do not relate to the topic will receive a low score, and that all responses undergo plagiarism checks — including AI checks, and including checks against previously submitted GAMSAT responses.

The sharpest warning is the one about borrowed text: an essay that leans on statements from a pre-prepared piece, often one supplied by a preparation provider, could be flagged even if you have never seen the original source. Two candidates who studied the same model essay can arrive at the same paragraph honestly. The check does not know that.

So where do all the rubrics come from?

They are in-house. Every band scale, every set of descriptors, every "this essay scored 78" you have read — including ours — is a private construction, because nobody outside ACER can validate one. ACER does not report a per-essay score, and candidates do not retain their essays, so no one can produce a matched pair of your essay and its official mark. Any provider implying otherwise is describing a document that does not exist.

The honest version is the one to look for when you choose a marker: built from the two criteria ACER publishes, with its scale set against the provider's own reviewed examples, and labelled as theirs rather than as ACER's.

Using two criteria as a checklist

Because the published basis is thin, it is easy to hold in your head while you write, which is exactly what a real checklist has to be.

The practical shift is away from form and towards thought. A tidy five-paragraph shell with nothing developed inside it answers the second criterion and fails the first. A genuine idea, explored two levels deep and expressed clearly, answers both — which is why the strongest essays usually have fewer moving parts than candidates expect.

The sitting itself

Two 30-minute tasks in a single 65-minute session, sat remotely under proctoring about three weeks before the test-centre sections. Each task offers four statements on a common theme; the first deals with socio-cultural issues, the second with more personal and social ones. Confirm the logistics against ACER's own instructions for your sitting — those change, and the criteria above do not.

The fastest way to see where your thinking is actually landing is to write one under the clock and have it read. Get your first essay marked free by two independent AI examiners on a rubric built from ACER's two published criteria — with the shape of the reasoning shown, not just a number.

GAMSAT® is a registered trademark of ACER. ACER is not affiliated with, and does not endorse, Aptavia.