How to Ace OxfordAQA IGCSE Computer Science Paper 2: Concepts and principles of computer science

How to ace OxfordAQA IGCSE Computer Science Paper 2: Concepts and principles of computer science

Paper 2 is where oxfordaqa igcse computer science paper 2 tests everything Paper 1 does not: the theory behind algorithms, data representation, hardware, networks and security. No code editor here, just written answers, and that changes what "good technique" looks like. This guide is deliberately practical: paper facts, question format, mark scheme habits, and a plan for the weeks before you sit it. Treat it as your igcse 9210 paper 2 tips reference, built around one oxfordaqa computer science exam technique at a time rather than vague general advice.

The paper facts

Duration: 120 minutes.
Total marks: 80.
Content: concepts and principles of computer science, covering algorithms, data representation, computer systems, networks, cyber security, relational databases and web page design, but no on-screen programming.

Eighty marks, 120 minutes, roughly a mark a minute once you build in time to read questions and check your answers. That is tighter than it sounds, because several questions on this paper ask for extended written explanations, and a rushed three-line answer to a six-mark question is a guaranteed way to leave marks on the table.

What the question format looks like

Expect short, fact-recall questions early in most sections, worth one or two marks each, followed by longer questions that ask you to explain, describe or compare. Algorithms and data representation questions often include a calculation, such as converting a number between bases or working out the size of an image file in bits. Computer systems and networks questions lean more on written explanation: describing what a component does, or explaining why a particular network setup or security measure is appropriate for a given scenario. Relational databases and web page design questions tend to combine a small worked example, such as a table of data or a fragment of HTML, with questions about it.

The command word at the start of each question tells you how much to write. "State" or "identify" wants a short, direct answer, often a single word or phrase. "Describe" wants you to give more detail about what something is or does. "Explain" wants you to give a reason, typically linking a cause to an effect: not just what happens, but why. Answering an "explain" question with only a "describe"-level answer is one of the most common ways marks are lost on this paper, because the "why" is where the marking points actually sit.

Worked example: matching the answer to the command word
Q: Explain why a hard disk drive is generally slower than a solid state
drive when reading data.
A "describe" answer might say "an HDD uses spinning magnetic disks and a moving read/write head, while an SSD has no moving parts." That is accurate but incomplete for an "explain" question, because it does not say why that difference matters. The marking point is the causal link: "the HDD's read/write head has to physically move to the correct location on the spinning disk before data can be read, which takes measurable time, whereas an SSD accesses data electronically with no physical movement required, so it responds faster." State the mechanism, then state the consequence. That second half is what "explain" is asking for.

How the mark scheme rewards answers

The oxfordaqa igcse computer science mark scheme for this paper is built from discrete marking points, and longer questions typically list several acceptable points, awarding one mark each up to the maximum for the question. A six-mark "explain" question rarely needs six separate ideas; it more often needs two or three well-developed points, each worth two marks for stating the point and then developing it with a reason or example. Padding an answer with repeated versions of the same idea, phrased differently, will not pick up additional marks; the mark scheme is looking for distinct points.

Calculation questions in algorithms and data representation are marked with method marks as well as an answer mark, so showing your working matters even when you are confident of the final number. If a base-conversion or binary arithmetic answer is wrong but the working shows a correct method with one small slip, method marks are often still available; a bare wrong answer with no working earns nothing.

Common pitfalls on Paper 2

Pitfall	Why it costs marks	Fix
Under-answering "explain" questions	A description without a reason misses the marking points built around cause and effect	After every "explain" answer, check you have stated a reason, not just a fact
Vague terminology	General computing terms picked up outside the syllabus, rather than the specification's own terms, can miss the exact marking point	Use the specification's own vocabulary: for example, "relational operators" rather than "comparison symbols"
No working shown on calculations	Method marks are lost even when the process was correct, if it is not visible on the page	Show every step of a base conversion or binary calculation, not just the final value
Running out of time on later sections	Networks, cyber security and databases sit later in the paper and are easy to shortchange under pressure	Check the clock against the mark total roughly a third and two-thirds through the paper
Ignoring the number of marks available	A two-mark answer needs two distinct points; a one-line answer to a four-mark question is structurally incomplete	Count the marks, then count how many separate points your answer actually makes

The spread of topics, and where marks concentrate

Paper 2 draws on the full theory side of the specification, so it pays to know roughly what each area tends to demand from you, rather than revising every topic to the same depth in the same way.

Algorithms: representing and tracing algorithms, working out the efficiency of a given approach, and comparing searching and sorting methods. Expect trace tables and short comparison questions.
Data representation: number bases, converting between them, binary arithmetic, character encoding, and how images and sound are represented and compressed. This is the most calculation-heavy part of the paper.
Computer systems: hardware, software, Boolean logic and systems architecture. Expect a mix of recall and explanation, often tied to a labelled diagram or component list.
Computer networks: how networks are structured and how devices on them communicate. Expect scenario-based questions asking you to justify a choice of network setup.
Cyber security: threats and the methods used to detect and prevent them. Expect "explain why" questions that want a mechanism, not just a name.
Relational databases and SQL: database design and simple structured query language. Expect a small sample table followed by questions that build on it.
Web page design: HTML structure and CSS styling. Expect a short fragment of markup or a style rule to read or complete.

Because every one of these areas can appear, do not gamble on question-spotting or skip a whole area because it "usually gets fewer marks." A no-nonsense revision plan covers all seven, then spends the remaining time on whichever two or three you personally find weakest.

Structuring your time and your answers

Read the whole paper before committing to an order, and answer the sections you find most comfortable first, whether that is data representation calculations or database questions, to bank marks early and build momentum. For longer written answers, jot a one-line plan in the margin before writing in full: what is the point, what is the reason, is there an example to attach. This costs seconds and prevents the drifting, unfocused answers that use up space without hitting the marking points.

Leave real time for the final section of the paper. A frequent, avoidable mistake is spending too long perfecting early calculation answers and then rushing the last page of extended writing, which is usually where the highest-value marks for explanation and comparison sit.

Practice strategy

Work through oxfordaqa igcse computer science past papers by topic first, then by full timed paper closer to the exam. Topic-by-topic practice lets you drill a weak area, such as network topologies or cyber security threats, without the pressure of the clock; full timed papers then test whether you can apply that knowledge at the pace the real exam demands.

Mark every practice answer against the published mark scheme, point by point, not just right or wrong overall.
Keep a running list of command words you tend to under-answer, and rewrite two or three practice answers to those command words at full "explain" depth.
Redo base conversions and binary arithmetic from memory regularly, since fluency here saves time that longer written questions need.
Read past examiner reports where available; they describe, in plain terms, exactly where real candidates lost marks on real questions.

These are the same oxfordaqa igcse computer science exam tips that apply across every board's theory paper, but applied here specifically to what Paper 2 actually tests: match your answer depth to the command word, show your working, use the specification's own vocabulary, and manage your time against the mark total rather than the question count. Direct preparation like this, focused on habits rather than just content, is what turns a paper full of theory into a straightforwardly scorable exam.

The week before the exam

In the final week, stop trying to learn new material and switch entirely to retrieval practice: closed-book self-testing on definitions, calculations and command-word answers, checked against the mark scheme immediately after. Redo at least one full past paper under exact timed conditions in this last week, sitting it in one block rather than spread across several days, so your stamina and pacing match what the real 120 minutes will demand. Sleep and a clear head on the day do more for a theory paper like this than one extra late night of cramming; recall under pressure depends on it.