What 10 years of Scholarship papers actually repeat

Every Scholarship "important topics" list you can find was written from memory. This one is a count: 500 questions from 20 papers across 10 sittings (2014 to 2025), each tagged with the specific thing it asks.

The finding that surprised us

387 distinct topics, 355 of them appearing in a single sitting, and only 9 appearing in three or more. The Scholarship paper is deliberately broad: it tests whether a child can think about something they have not seen before, which is precisely why a narrow "important topics" drill disappoints families every year.

What this means: drilling a handful of topics is not enough. Do the topics below first, but do not do only them.

The topics that come back most often

Ordered by how many sittings asked the topic. Two questions in two different years is a stronger signal than two questions in one year.

Reasoning tags dominate the top of that table, and that is the practical takeaway for a parent: the questions that repeat are the ones about handling information, not the ones about remembering a fact. Those are trainable, and they are trainable in fifteen minutes a day.

Questions per sitting

The questions we have curated, by sitting:

Two sittings are missing from that table (2016 and 2020) because we have not finished checking those papers against their marking schemes. We would rather have a gap than a wrong answer in front of a ten-year-old.

Method

One tag per question, assigned during curation and checked alongside the answer against the published marking scheme. Tags are specific rather than broad, which is why the distinct count is high; a coarser scheme would give a shorter, friendlier list that told you less. Both papers from each sitting are included.

One question from the pile