GAMSAT timed practice: when to add the clock, and how

Do both, in that order. Untimed practice is where you build a method you can state, and timed practice is where you find out whether that method survives a clock. The move from one to the other should depend on what your review shows, not on how much of your preparation period has gone.

One GAMSAT preparation company lays out the same order in its published list of preparation tips. Early on, it says, speed is a lesser concern than how you approach a question. It advises untimed work with every question reviewed: if you were wrong, find out why, and if you were right, confirm that your reasoning earned it. Only then does it suggest adding a clock and comparing how your thinking differs between the two conditions. That is one competitor's advice, not guidance from the Australian Council for Educational Research (ACER), and its trigger for adding the clock is the loose one of having got the hang of the questions. An observable trigger is the missing piece, so the rest of this article proposes one. It is our own proposal, and we have no outcome data behind it.

Three problems that look the same in a timed score

A question you get wrong under a clock can be wrong for at least three different reasons. You may not have known what to do, at any speed. You may have known exactly what to do and run out of time doing it. Or you may have a method that works untimed and dropped it once the clock started. A score, or a count of errors, records the same thing in all three cases.

The remedies differ. The first needs more untimed work. The second needs pace built up in steps. The third needs the clock introduced more gently, because the method exists and is not surviving the conditions. Treating all three as a speed problem is how a candidate ends up adding timed sets that do nothing for a gap in content or method. If you have been timing yourself and the questions have stopped teaching you anything, one thing to check is whether your timed sets still leave a reasoning record you can review. A set that produces only a score gives the review nothing to work from.

A ladder with three rungs

Each rung changes one thing. That is the design: if the clock is the only variable that moved between two sets, a change in how you reasoned can be attributed to the clock.

Rung one: untimed reasoning. No clock. The aim is a method you can write down, step by step, for each question type you are working on. Review every question, correct ones included, because a right answer reached by a weak route tells you little about whether the method is sound (correctness and reasoning quality covers that distinction). Here is a test for leaving this rung. It is ours, not an official standard: across several sets in a row, you can state in a sentence why each answer was right or wrong, and the correct answers survive that check.

Rung two: capped-time sets. Add a limit, but set it loose enough that you can still complete your untimed method, then tighten it in steps. We do not have a verified starting figure to give you, so the size of the first cap is yours to choose. The review stays as thorough as it was on rung one. The test for moving on: the reasoning you can reconstruct under the cap matches the reasoning you wrote down untimed, and the errors are of kinds you already know.

Rung three: full timed blocks. Here the timing is the examination's own. Check ACER's current official materials for what that timing is, because we have not taken figures from competitor pages. A full block tells you something about pace and about whether your method holds at full pace, and it needs the same review afterwards.

If a rung makes you stop reviewing, step down one rung rather than abandoning the clock. The aim is for review to continue at every rung, and the rung where it stops is the one to repeat.

Review questions that sort a timed miss

After a capped or timed set, take each miss through these questions, in this order.

• Could I have set out the method before the clock started? Try a fresh question of the same type with no clock. Reusing the same question would test your memory of its answer. If the fresh question also goes wrong untimed, one explanation is a gap in content or method, and the clock was not the cause.
• Did I finish the reasoning, only late? Look at where the time went. If the reasoning is sound but the steps ran past your cap, one explanation is pace, which rung two exists to build.
• Did I use a different method from the one I use untimed? Compare the steps you took with the ones you wrote down on rung one. A step skipped only under the clock is a sign the clock changed your reasoning, and it suggests a smaller step down in the cap.

Reconstructing this from memory has limits. What you remember doing and what you did can differ, which is one reason to write the review down while it is fresh rather than recall it later. What a structured mistake review involves is covered separately.

What this does not settle

The ladder costs sets spent below full pace, and we cannot say how many sets each rung needs. That varies by candidate, and we hold no data that would put a number on it. It also does not replace checking the examination's actual conditions with ACER. If a sitting is close and there is no time for three rungs, the same review questions still apply to whatever timed sets you can run.

The step this leaves open is the evidence. A marked set shows which options you chose and whether they were right, but not how you got there, so telling a timing problem from a content gap rests on your memory of your own reasoning. Verantrix can record how you reasoned through a question, not only which option you chose, which gives the review questions something written to compare between an untimed set and a capped one. On your next capped set, record your reasoning on each question, then sort every miss into the three problems and see which one accounts for the most.