GAMSAT trial questions: what to do and what to ignore

ACER's results page says the test may include trial questions, described as a small number, that might not feed into a candidate's score. The worry that follows is natural: should you try to spot them, and should a strange question change how you handle it? Our answer is no to the first and very little to the second. ACER's pages describe no way to identify a trial question, and the test-taking advice they do give works the same for every question in front of you.

What ACER says, and what it leaves out

The statement appears in ACER's explanation of how it keeps the test fair and valid. It says trial questions let ACER confirm that new questions meet the standard it expects of them, and the same passage notes that test data is analysed statistically to check how each question performed. The wording is conditional on both counts: the test may include trial questions, and those questions may not contribute to scores.

What the pages do not say matters as much. They give no figure beyond the word small, do not name a section, and describe no way for a candidate to recognise a trial question while sitting. ACER's preparation page, where the test-taking guidance sits, does not mention trial questions at all.

Why trying to spot them does not pay

Start from what you could actually know mid-test. A question's difficulty, its unfamiliarity and where it sits are things you can observe. Whether it is scored is not, and nothing ACER has published lets you check a guess about it. Time spent deciding is therefore time spent on a judgement you cannot confirm. That is our reasoning rather than an ACER statement, and it leads to a rule we would hold firmly: treat every question as though it counts.

The rule also protects you when the guess would have been wrong. If you decide a question is a trial item and give it less effort, and it turns out to be scored, the loss is yours. If you treat it as scored and it is not, the cost is the effort you spent, and ACER's pacing advice, covered next, already puts a limit on that.

What ACER's advice asks of you on each question

For Humanities and Social Sciences and for Biological and Physical Sciences, both multiple choice, ACER's preparation page gives a short procedure. Keep a steady pace and avoid investing too much time in any single question. If you believe you know an answer, enter it without waiting for certainty, move on, and return later should time allow. ACER's stated reasoning is that every question carries equal value, so attempting as many as possible serves your score.

Notice that the procedure does not ask you to sort questions into scored and unscored. It sorts them by something you can observe: whether you have an answer you are prepared to enter now. For a question where you have no answer at all, ACER's guidance on that page goes no further than moving on, and the next section covers what the no-deduction rule allows. All of this is guidance for the multiple-choice sections. Section 2 has separate ACER advice, which we cover in our piece on length, spelling and punctuation.

Guessing, blanks and the clock

ACER states that a wrong answer carries no deduction. Its results page also describes the raw score as the number of questions answered correctly. Put those two statements together and, under ACER's stated rule, a blank cannot add to that count and a wrong answer cannot subtract from it. An entered guess costs nothing, and a blank gives up the only chance the question offers. That is a direct consequence of what ACER says, and it matches ACER's own advice to attempt as many questions as you can.

For timing, this suggests a practical order. Answer what you can settle at a steady pace, enter a best attempt where you have one, and before time ends make sure no multiple-choice question is left empty. That last step is our recommendation, built on the no-deduction rule, not a quoted instruction. ACER does not say how long counts as too long on one question, so that threshold is yours to set in practice.

ACER's online practice tests for the two multiple-choice sections run at full length with an inbuilt timer: 100 minutes for Humanities and Social Sciences and 150 for Biological and Physical Sciences. They are one place to find your own point for moving on, for example by noting how many questions you flagged to revisit and whether the time you had left was enough to return to them.

When a question seems strange

A question can feel unusually hard, oddly worded or far from what you prepared. Nothing in ACER's material links any of those qualities to trial status. The results page says only that trial questions help check that new questions perform as required. Treating an odd question as evidence that it is unscored is an inference ACER has not supported, and acting on it risks giving up marks on a question that may well count.

So handle it as you would any other. Give it a reasoned attempt, enter your best answer, move on, and return if time allows. Your reaction to the question is something you can observe and later review. Its scoring status is not. If a question looks faulty rather than merely hard, ACER's results page says to alert the supervisor on the day, after which the GAMSAT Office reviews the query and passes it to the test developers. That route does not tell you whether the item was a trial question, and it does not replace answering it.

Where this advice runs out

Several things are not established by the pages ACER has published. How many trial questions a given sitting holds, which section they sit in, and how any one of them is handled in scoring beyond the phrase may not contribute are all unstated. ACER's guidance also gives no figure for time per question, and it makes returning to a question conditional on time permitting, so moving on can leave a question unresolved if the clock runs out. Our rule is a response to that uncertainty, not a claim to have removed it.

If your practice scores have stalled, over-investing in single questions is one thing to check, alongside others. Why repeat sitters plateau looks at whether a recurring reasoning pattern is still costing you marks.

The gap this leaves is on your side of the screen. After a practice set, a question that felt strange gives you a result and an impression, but not a record of how you reasoned, so it is hard to tell whether the question was unusual or whether the same reasoning pattern caught you again. Verantrix can record how you reasoned through practice questions, not only which option you chose, and can identify patterns across your attempts; it cannot tell you whether any question was a trial item. On your next practice set, flag every question that seemed off, write one line on why before you check the answer, and see whether the same reasons keep appearing.