Exam Preparation — mocks, finals and revision planning

Two things stand clearly above the rest. In the most cited review of the field — Dunlosky, Rawson, Marsh, Nathan and Willingham, Improving Students’ Learning With Effective Learning Techniques, published in Psychological Science in the Public Interest in 2013 — ten common techniques were rated for utility. Practice testing and distributed practice were the only two rated high. Interleaved practice, elaborative interrogation and self-explanation were rated moderate. Summarisation, highlighting, the keyword mnemonic, imagery for text and rereading were all rated low. Two high, three moderate, five low — and the five low ones are, between them, a fair description of how most students revise.

Because the studies do not find it works. The review’s summary of the evidence is that most studies have shown no benefit of highlighting over and above the benefit of simply reading the material. It feels productive, it produces a visibly worked-on page, and it leaves the reader passive. The same applies to rereading, which the review found does not consistently boost performance and concludes that other techniques should be used in its place. Neither is harmful in itself. The cost is the hours — a student who highlights for twelve weeks has spent twelve weeks not testing themselves.

Closing the book and trying to produce the answer from memory, then checking. Not rereading the notes, not copying them out, not looking at a worked example and nodding — producing it. Past papers under timed conditions are the exam-term version. Flashcards, self-quizzing and answering a question before turning the page are the daily version. The effect is one of the most replicated findings in the field, and it holds across ages and subjects; a 2017 meta-analysis of practice testing by Adesope, Trevisan and Sundararajan in the Review of Educational Research is the standard reference. It is also the technique students most reliably dislike, because it feels harder than rereading. That is rather the point.

Spreading the same total study time across more sessions rather than massing it into fewer. Four hours as four one-hour sessions across four weeks beats four hours in one evening, for the same four hours. It is the second of the two high-utility techniques, and it is the one that a revision timetable can actually deliver — which is why we spend time on the timetable rather than treating it as admin. The practical consequence for an exam year is that starting earlier is worth more than working harder later, and that is a scheduling decision rather than a motivational one.

For tomorrow, sometimes. For a final examination in May, no — and it is the direct opposite of the distributed-practice finding. The honest position is that massed study before a test can produce a short-term result that decays fast, which is why it feels like it works and why students keep doing it. What we would say to a family is narrower and more useful: if there are four weeks left, the plan should be spaced across those four weeks; if there are four days left, then technique and past papers on the highest-weight topics will do more than trying to learn the syllabus again.

There is research associating all-night study with lower grades, though we are careful about the numbers — the widely quoted study reports the direction of the association from a sample of around a hundred students, and we could not verify the specific figures often attached to it, so we will not publish them. The mechanism is not controversial: sleep matters for consolidating what was learned that day, and an exhausted candidate reads questions badly. This is one of the few pieces of advice that costs nothing and that both test publishers and teachers agree on.

The evidence does not support teaching to a learning style. This is one of the most persistent ideas in education and one of the least supported, and it matters because time spent converting notes into a preferred format is time not spent testing. What is true underneath it is less exciting but more useful: different subjects genuinely have different best formats — a diagram is the right representation for a mechanism, a worked example for a method, a timeline for a sequence — and that is a fact about the material rather than about the child. We will not build a revision plan around a learning-style label.

Mixing topics within a session rather than doing all of topic A then all of topic B — alternating question types in a maths paper, for example, rather than doing thirty of the same kind. It was rated moderate rather than high in the Dunlosky review, on the basis that it is promising but the evidence for it was more limited and more recently explored. We use it, and we describe it accurately: it helps particularly where the exam requires choosing which method applies, which is precisely what blocked practice hides. It is not in the same evidential class as retrieval practice, and we will not pretend it is.

They are the strongest single tool in an exam term, because they are retrieval practice and timed rehearsal at once. The mistake we see most is doing them open book, or marking them generously, or not marking them at all — at which point they become rereading with extra steps. Done properly, that means under time, without notes, then marked against the actual mark scheme rather than against a feeling. Reading the mark scheme is a distinct skill and it is teachable quickly: students routinely lose marks they knew the content for, by not answering in the shape the scheme rewards.

Usually one of three things, once a genuine gap in the syllabus has been ruled out, and they need different responses. It may be technique — knowing the content but not how the marks are awarded, which is fixable quickly with mark schemes and timed papers. It may be timing — running out of time on the last question every time, which is fixable with pacing practice. Or it may be exam anxiety, where performance under pressure falls well below what the same child produces at home. The research on test anxiety consistently finds a negative association with performance, and there is evidence that it can be reduced. We stay firmly on the practical side of that line: familiarity, rehearsal under realistic conditions and a plan for the first five minutes. Where anxiety looks more serious than that, the right referral is not a tutor and we will say so.

For the IB, the November 2026 session runs from 26 October to 13 November 2026, and the May 2027 session runs from 23 April to 18 May 2027. For Cambridge International, the November 2026 series closed final entries on 16 August 2026 with late entries to 21 September, and results are released on 7 January 2027 for International AS and A Level and 14 January for IGCSE. For Pearson Edexcel, the October 2026 International A Level series closed on 28 August 2026, January 2027 entries close on 16 October 2026, and June 2027 international entries close on 21 March 2027. Mocks in Swiss international schools typically fall in January or February. The Matura examination period is cantonal and varies. Tell us the date and we size the package to it.

Fewer than families fear if the plan is right, and more than they hope if it is not. The most common useful shape we see is five to ten hours around the mocks — diagnosing what went wrong rather than covering everything — followed by weekly work on the two or three specific things the mocks exposed. Twenty hours across a subject makes sense when the syllabus genuinely has not been covered, which is a teaching problem rather than a revision one. If a provider tells you the answer is always more hours, be suspicious; the evidence says the return comes from how the time is spent at least as much as from how much of it there is.

Usually a mock paper or a recent test, marked properly against the scheme, with the tutor and the student working out together where the marks went. That is deliberately unglamorous and it is the highest-value hour in the whole package, because almost every family arrives with a theory about what is wrong and the paper frequently says something different — that it is timing rather than knowledge, or one topic rather than the subject, or the way answers are written rather than what is in them. Everything after that is planned against what the paper showed.

Send the most recent mock or test paper, with the marks on it, and tell us the date of the real examination. Those two things are enough for us to say what the actual problem is and how many hours it is — and occasionally to say that the content is fine and what is needed is a revision timetable your child will follow, which is a conversation rather than a package. We come back with a tutor who has taught that specification, typically within 48 to 72 hours. Nothing is charged to enquire and nothing is charged to be assigned. Book a tutor, or send a WhatsApp message on +41 76 213 1843.