Loading article…
S3 is the second turning point in a Rwandan student's life. Past papers are the highest-leverage tool you have — and most kids use them completely wrong. Here is how to do it right.
S3 is the second moment in a Rwandan kid's life when one exam decides what comes next. P6 decided which secondary school you got into. S3 decides whether you continue to A-level (S4–S6), join TVET, or repeat a year. Around 180,000 students sit S3 every year in Rwanda. The pass rate hovers around 92%. But like P6, the real question isn't passing — it's the aggregate, because that decides your combination.
I've had this conversation with a lot of S3 students. Most of them are doing one specific thing wrong. They're collecting past papers but not using them — and the way they use them, when they finally do, leaves half the value on the table.
This is a long post, but if you're an S3 student or a parent of one, please read all of it. The difference between “passed” and “placed in MCB at a strong school” is in here.
The S3 exam is set by NESA, aligned to the REB curriculum, and structured almost identically to the year before's paper. Question types repeat. Topics repeat. The wording changes; the underlying structure barely does.
This is why a single past paper, done properly, is worth more than ten chapters of textbook reading. You're practicing the exact format, the exact pace, the exact kind of question you'll see on the day. And you're building something more valuable than knowledge: familiarity with the test. Familiarity reduces panic. Panic is what costs students marks.
There are three real sources:
I'm biased, obviously. But the question isn't which source you use — it's whether you actually do them.
The S3 exam covers nine subjects. They're all required:
You can't skip any. But your time isn't equal across them. The conventional wisdom says give equal time to each subject. The reality, after watching hundreds of S3 students prepare:
The wrong way: open the past paper, read it through casually, glance at the answer key, think “ah yes I would have gotten that”, close it, feel productive.
This is worthless. You haven't practiced retrieval. You haven't practiced under time pressure. You haven't practiced sitting with discomfort. You've essentially read a quiz the way you'd read a magazine.
Sit at a table. Set a timer for the exact duration of the real exam. Phone in another room. No music with words. No snacks within reach. If you can't make 90 minutes uninterrupted, don't do the paper — wait until you can.
If you don't know a question, write “skip” next to it and move on. Come back to it before the timer ends. Don't look at the answer key. The instinct to peek is the thing destroying the value of past papers — the discomfort of not knowing is exactly the discomfort that's training your brain.
Use the official mark scheme. Be strict. If you wrote “kinetic” but the answer is “kinetic energy”, that's a half-mark, not a full mark. The strictness of your marking is what trains you to write precise answers under pressure.
This is the step almost everyone skips. Pull every wrong question into a notebook. Re-do it from scratch the next day. Re-do it again three days later. Re-do it again two weeks later. By the time the real exam happens, those questions feel automatic. This step is where past papers actually pay off.
After every five papers, look at your wrong-question list. Which topics keep coming up? Those are where you focus your textbook study. Past papers are the diagnostic; textbooks are the treatment for the specific ailments past papers identify. People do this in the wrong order — they study topics they think they should study, then test what they happen to know.
Minimum: 5 full papers per subject in the eight weeks before the exam. Better: 10. The kids I've seen jump from C to A in S3 all did 10+ per subject. The kids who didn't typically did 1 or 2.
The ratio matters more than the absolute number. If you do 10 papers but mark them lazily, you'll get less from them than 5 papers marked strictly with full re-do.
The math paper is split between calculation questions and word problems. Calculation questions are easier to drill — you can do 20 in an hour. Word problems are where most kids lose marks because they don't set them up right. For every word problem you get wrong, write out the “translation” — what the words mean in math notation — before you re-attempt. Half the time the original error was at translation, not at calculation.
The sciences in S3 reward precision. Memorise units. Memorise the prefixes. The number of marks lost on physics for missing units (writing “5” instead of “5 m/s”) is staggering. In chemistry, balance every equation, even when not asked. In biology, draw labels for every diagram, even when the question asks only for the name.
The composition section is where you can pick up easy marks. Pick a structure and stick with it: introduction, three body paragraphs, conclusion. Plan it for two minutes before writing. Write the conclusion even if you're running out of time — half a conclusion is better than none.
The literature questions reward depth. Pick three of the prescribed texts, read them properly, and write a one-page summary of each — themes, characters, key passages. You can usually predict which texts will appear based on what came up in the last three years. Inkoranya, Imihigo, and several others rotate predictably.
Maps and timelines. Both subjects test maps and timelines repeatedly. If you know the major dates of Rwandan history (1885, 1959, 1962, 1973, 1990, 1994, 2000, 2003, 2017) and you can sketch a map of Rwanda's provinces and major cities, you've already locked in 15+ marks across both papers.
The questions are conceptual and predictable. Define a concept, give an example, explain a benefit, identify a challenge. The structure of the answers is more important than the content — examiners are looking for completeness, not creativity.
If you start eight weeks out, here's the rhythm:
Same advice as P6: don't teach, mark. If you have time, mark your kid's past papers. The act of timely feedback matters more than the quality of explanation. Most kids fail not because they don't understand, but because they don't know what they don't understand.
Don't finish reading this and tell yourself you'll start tomorrow. Pick your weakest subject. Open Ganzaa's past papers library. Pick last year's paper. Set a timer. Sit with it.
That's the start. The rest is just doing it again, fifty more times, with honest marking. Most of S3 prep is unglamorous, repetitive, and uncomfortable. The kids who do it anyway are the ones who get placed where they want.
I've tested the major learning apps available to Rwandan students. Here is the honest, opinionated review — what works, what doesn't, and what to skip entirely. With my biases declared up front.
I've sat at the back of P6 classrooms in Kigali, Musanze and Rubavu for two years. Here is what actually moves the needle on the leaver exam — not generic advice, the real thing.
Ganzaa is Rwanda's online learning platform — free for individual students, with REB-aligned practice and past papers. Schools get the first term free.