Revising with glossary entries or with questions?

A glossary entry builds a notion: it explains what a tool is, what it is for, how it is set up and where it goes wrong. It is made to be read once carefully, then reread quickly.

A question measures: it says whether the notion is available at the moment you have to decide. That is information reading cannot produce, because a text you understand as you read it gives the illusion of being acquired.

The mistake is to believe that one replaces the other. Revising with entries only produces a candidate who recognises everything and decides nothing; practising with questions only produces someone who memorises answers without understanding why.

Start with the questions, not the entries. It is the opposite of what most candidates do, and it is what changes the most. A set played before any revision gives a map of the topics that resist — which is worth several hours of general reading.

Then read the entries for the topics you got wrong, those only. Revision time is limited; spending it on what is already acquired is the most common waste of a preparation.

Play again, and see whether the gap has moved. A topic that stays weak after reading does not call for a rereading: it calls for another angle — an example, a calculation redone by hand, a discussion.

Finish in exam conditions. Limited time, correction at the end, no going back. It is the only step that tests time management, and it cannot be improvised on the day.

A right answer found by elimination or by chance is worth less than a wrong answer understood. That is why the correction is read systematically, including on the questions you got right: that is where the nuance lies that will come back in another form.

It is also the only way to spot fragile successes. A candidate who tracks their score does not know how many right answers they owe to luck; one who reads the corrections knows, and therefore knows where they really stand.