A past paper is a dress rehearsal, not a script to memorise. The useful part is the demand it makes while you cannot see the mark scheme: time, wording, and a blank page. The useless part is learning that in 2022 the third question was about a particular case, then hoping the paper asks it again in the same words.
If you read the questions and the answers in one sitting, you mostly practise recognition. You leave feeling covered. The next unseen paper does not care. This is a way to use the papers you have so that the skill transfers.
Before you start, decide the job of this sitting
Not every sitting has the same job. Pick one.
- A timed paper, or a timed section, under exam conditions. No notes, no scheme, a clock, and the tools you are allowed in the real exam.
- An untimed diagnostic on a topic you have just studied, used to find holes. Still no scheme until you have answered.
- A technique drill on one question type, repeated with new numbers or a new source text, after you have already sat one honestly.
Doing all three at once is how an evening becomes "I looked at a paper." Write the job at the top of the page. If the job is a timed paper, do not pause to check a definition. If the job is a diagnostic, do not punish yourself for going slowly. Speed is a later sitting.
How to sit the paper
Put the mark scheme in another room, or in a closed file. Open only the question paper and the permitted data booklet. Set a timer for the minutes that section would get, not for "until I feel done."
Read the whole section once, quickly, and star the questions you can start. Begin with those. This is not cheating. It is what a calm candidate does when the paper is in front of them. Leave a question you cannot start within a minute, mark it, and move. Coming back with ten minutes left is more useful than spending twenty minutes inventing a method you do not have.
Write as you would in the exam. Show the step the marker needs, not a private shorthand you will "explain if asked." If the question says "compare," write the comparison, not two summaries placed next to each other. If it says "calculate," keep the unit in the working.
When the timer ends, stop. Draw a line under what you produced in time. You may continue in a different colour if you want to see what you knew but could not reach. Do not mix the two. The line is the information. Everything below it is extra practice, not part of the score you are allowed to believe.
Marking without lying to yourself
Mark only after the line is drawn. Use the scheme as a critic, not as a story you now adopt.
For each mark you missed, write one of four labels, and only one.
- Did not know. The idea was absent.
- Knew, but too vague, or missing the term, condition, or unit the scheme wanted.
- Misread. You answered a nearby question, ignored a "not," or used the wrong figure from the stem.
- Method. You knew the idea and still took a step that does not follow, or you could not finish in time.
A total score is the least interesting line on the page. A paper that is 60 percent "did not know" needs a return to the source. A paper that is 60 percent "misread" needs slower reading of stems, not another week of notes. A paper that is 60 percent "method" needs fresh problems of that type, done without the scheme. The label chooses the next hour. The percentage is only a mood.
Do not award yourself a mark because your answer was "basically" the scheme. If the scheme wanted a direction, a named condition, or a unit, and you did not write it, it is a miss. Kindness here is how the same miss survives until the real exam.
What to do the next day, not the same night
The same night, the paper is contaminated. You have just seen the scheme, so another immediate attempt mostly recalls the paragraph you read twenty minutes ago. That can feel like learning. It is a short-term echo.
The next day, with the scheme closed again, redo only the questions you missed, or a new question that demands the same skill. For a "did not know," spend ten minutes on the source first, then answer a narrower prompt from memory. For a "misread," cover the question and rewrite what it is actually asking in one line before you solve it. For a "method" miss, do two fresh items of that type. Do not reread your old wrong answer until you have tried again.
Keep a short list of the labels, not a shrine to the paper. If you want a fuller record of why a miss happened, an exam mistake log is built for that. The paper sitting and the log are partners. The sitting produces the evidence. The log stops you from remembering only the score.
How not to memorise the paper
Some repetition of a specific question is fine. Memorising it is what happens when the question, and not the skill, becomes the thing you can perform.
- After you can do a missed question cleanly, change a number, a name, a date, or a condition and do it again. If you cannot, you learned the script.
- If two years ask the same skill in different clothes, practise the skill once from each paper rather than copying both model answers into a document you reread.
- Do not highlight the model answer and call it notes. Write the rule that would have earned the mark, in one sentence, then close the scheme.
- Avoid doing the same paper every weekend until it feels friendly. Friendliness with one PDF is not readiness for the next PDF.
- If a mark scheme essay is beautiful, steal the structure, not the sentences. Claim, reason, evidence, limit. Then write a different essay under a timer.
Teachers reuse ideas. They rarely reuse paragraphs. Train for the idea.
A rhythm that uses a small pile of papers
If you have three papers, do not burn all three this week.
Use the oldest as a timed diagnostic now. Mark it with labels. Spend the next few sessions repairing the labels with the source and with fresh questions, which can come from the middle paper's relevant items only, not from sitting that whole paper early. Save the newest paper for a later timed sitting, when the repairs have had a few days to exist. One clean later paper tells you more than three papers read with the scheme beside you.
If you only have one paper, protect it. Do it timed, once. Mine the misses for a week. Do not keep redoing it until you can recite it. Invent sibling questions instead: same command word, different content.
Between papers, ordinary study still matters. A past paper does not introduce a topic you have never met. It reveals whether you can use a topic you have met. If half the paper is "did not know," you are not in a past-paper phase yet. You are in a first-learning phase, and another paper will only manufacture anxiety.
Timing, nerves, and the blank first minute
The first minute of a timed paper often feels empty even when the knowledge is there. Give that minute a job so you do not fill it with panic. Write the command word, the number of marks, and the first step you intend to take. Then take it. If the step is wrong you will find out when you mark. If you spend the minute staring, you learn nothing except that you can freeze.
Practise this start on short sections through the term, not only on the Saturday before the exam. A clock is a skill. You will not acquire it by reading about time management.
When you review timing, note where the minutes went. A question worth a few marks that ate half the section is a strategy error, and it gets the label "method" even if the fragment you wrote was correct. Next time, the drill is to leave earlier. That is as trainable as a formula.
What a good evening looks like
Ninety minutes is enough for a section, not a heroic full paper plus a rewrite plus a new topic.
- Ten minutes to set the rules: job, timer, scheme out of reach.
- The timed section itself.
- A short break where you do not open the scheme yet. Stand up.
- Marking with four labels, and a one-line rule for each miss.
- Stop. The repairs are tomorrow, with the scheme closed.
If you still have energy, do not start a second paper. Add one sibling question for the ugliest miss, from memory, and then leave it. You are building a habit of attempting unseen work. You are not trying to become a person who has read every archived PDF.
Questions about past papers
Should I look at the mark scheme before I revise a topic?
No. Attempt first, even if the attempt is short. The scheme is how you mark, not how you start.
What if my course has no published papers?
Use class tests, textbook exercises you have not annotated, or questions you wrote last week and have not looked at since. The rule is the same: answer, then check.
Is a low score a reason to abandon past papers?
It is a reason to change the next hour. A low score full of "did not know" means return to the source. It does not mean the paper was a waste.
How many papers should I finish before the exam?
Enough timed sections to meet the clock and the command words, plus repairs in between. A stack of unread model answers is not a number worth counting.
After you have sat a section yourself, you canturn your own notes into a fresh quiz on Oyrivaso the next questions are not the same items you just marked.
