Start with a diagnostic, not a marathon

A full paper can tell you where marks are leaking, but doing full paper after full paper often repeats the same mistakes. Begin with one paper or a mixed diagnostic set. Mark it, then group every lost mark by topic and by mistake type.

Keep the categories concrete. “Algebra” is too broad to guide the next session. “Changing the subject when the variable appears twice” or “forgetting the final inequality sign after dividing by a negative” gives you something you can actually practise.

  • Knowledge gap: you did not know the method or fact.
  • Method gap: you knew the topic but chose the wrong approach.
  • Execution error: arithmetic, notation, units, or copying went wrong.
  • Exam error: you misread the command, skipped a part, or managed time badly.

Practise one weak skill in a focused set

Once a gap is clear, switch from whole papers to questions from that topic. A useful set is long enough to reveal a pattern but short enough to review properly in the same session. Six well-chosen questions with careful corrections beat twenty rushed questions that you never revisit.

Mix easier and harder versions of the skill. Start by rebuilding the method, then include unfamiliar wording or multi-step questions so you learn to recognize when the method applies rather than memorizing one visual pattern.

Use the mark scheme as feedback, not a script

Finish the attempt before opening the answer. Then compare each line of your work with the mark scheme or worked solution. Do not stop at whether the final answer matches. Find the first line where your reasoning diverged and write one short correction beside it.

If the mark scheme uses a different valid method, check whether yours would still earn the method marks. If you cannot tell, ask a teacher or tutor. The goal is to understand what a complete, creditworthy solution needs to show.

  • Write the missing step in your own words.
  • Redo the question from a blank page without copying.
  • Save the question if the mistake is likely to recur.
  • Record one rule for next time, not a vague note to “be careful.”

Retest mistakes on a delay

A corrected answer can feel obvious while the mark scheme is still open. That is not the same as being able to solve it independently. Return to the question after a delay, hide the previous work, and solve it again from scratch.

If the same mistake returns, narrow the skill further or use an easier example before trying the exam question again. If the method holds, mix that question with other topics so you must identify the right approach without a topic label giving it away.

Move back to timed papers when the gaps are smaller

Use full papers to test selection, stamina, and timing after focused practice has improved the weak areas. Sit them under realistic conditions, keep a record of marks by topic, and compare the pattern across several papers rather than obsessing over one score.

The loop is simple: diagnose, target, review, retest, then simulate. A past paper is not just a mock exam. It is a map of what to practise next.

Frequently asked questions

Should I do past papers by topic or by year?

Use questions by topic to repair a specific weakness and full papers by year to test timing, question selection, and performance across the whole course. Most students need both modes at different points in the same revision cycle.

How many past papers should I do?

There is no useful universal number. Stop counting papers and track whether recurring mistakes are disappearing. A smaller number of papers that you mark, correct, and retest is more useful than a large stack completed without review.

When should I look at the mark scheme?

Look after you have made a complete attempt or reached a genuine stopping point. Opening it too early turns problem solving into copying; waiting until the end lets the mark scheme explain exactly where your independent method broke down.