Spacing and mixing: how to schedule maths revision

5 min read

Two scheduling decisions change how much maths revision survives to the exam: how far apart sessions sit, and whether topics are blocked or mixed.

Two decisions that do the heavy lifting

Most revision advice concerns what to study. Two decisions about scheduling matter at least as much: how far apart the sessions sit, and whether you practise one topic at a time or several mixed together. Both have been tested directly in maths classrooms, and both point away from how students naturally revise.

Spacing: the same hours, spread out

Cepeda and colleagues (2006) synthesised a large body of research comparing massed practice, which is everything in one sitting, against distributed practice, which is the same total time spread across days or weeks. Spacing won consistently, and the advantage grew as the delay before the final test grew. For an exam months away, spacing is not a small optimisation.

The same 2013 review that rated practice testing highly gave distributed practice its other high utility rating. The practical translation is unglamorous: four thirty-minute sessions across a week beat one two-hour session, even though the total time is identical.

A workable default is to revisit a topic roughly a day after first meeting it, again about a week later, and again a month later. The exact intervals matter far less than the fact that old topics keep coming back instead of being marked as finished.

Interleaving: the maths-specific finding

Interleaving means mixing question types within a practice set rather than doing twenty of the same kind in a row. This has been tested repeatedly in maths, which makes the evidence unusually direct.

Rohrer and Taylor (2007) found that shuffling maths problems improved later test performance compared with practising them in blocks. Rohrer, Dedrick and Stershic (2015) replicated the effect, and a randomised controlled trial in real classrooms (Rohrer, Dedrick, Hartwig and Cheung, 2020) found interleaved practice produced substantially higher scores on a delayed test than the blocked practice used in ordinary homework.

The reason is straightforward once stated. If a worksheet is headed 'quadratic equations', the student never has to work out which method applies, because the heading has already answered the hardest question. Exams remove the heading. Blocked practice trains execution while quietly skipping selection; interleaved practice trains both.

The catch worth knowing about

Interleaved and spaced practice both feel worse while you are doing them. Accuracy during the session drops, sessions feel disjointed, and students often conclude the method is not working. In the studies, performance during practice and performance on the later test move in opposite directions. Judging a revision method by how smooth it feels is exactly the mistake to avoid.

What this looks like in practice

• Build mixed problem sets that pull from several topics, rather than working through a chapter in order.
• Keep finished topics in rotation. A topic covered in September should still appear in a set in November.
• Use past papers early, not only at the end. They are interleaved by design.
• Prefer short, frequent sessions to long, rare ones, especially in the months before an exam.
• Expect a lower success rate during practice than a blocked worksheet would give you. That is the trade being made deliberately.

None of this requires extra hours. It is the same revision time, ordered differently, which is what makes it one of the highest-return changes a student can make.