AS CAIE Computer Science: Exam Techniques and Marking Criteria — AS CAIE 计算机：答题技巧与评分标准

📚 AS CAIE Computer Science: Exam Techniques and Marking Criteria | AS CAIE 计算机：答题技巧与评分标准

Scoring well in the CAIE AS Computer Science (9618) exam requires more than just knowing the syllabus — it demands a strategic approach to how you answer questions. Understanding command words, mark scheme expectations, and paper-specific techniques can make the difference between a B and an A. This guide breaks down the essential strategies for both Paper 1 (Theory Fundamentals) and Paper 2 (Fundamental Problem-solving and Programming), helping you maximize every mark.

在 CAIE AS 计算机科学（9618）考试中取得好成绩，不仅需要掌握考纲内容，更需要策略性的答题方法。理解指令词、评分标准的要求，以及各试卷的专项技巧，往往是拿到 A 而不是 B 的关键。本指南将为你详细解析 Paper 1（理论基础）和 Paper 2（问题解决与编程基础）的核心应试策略，帮助你抓住每一分。

1. Understanding the Exam Structure | 理解考试结构

The CAIE AS Computer Science qualification consists of two compulsory papers. Paper 1 (Theory Fundamentals) is a 1-hour-30-minute written paper worth 75 marks, covering topics from information representation to system software, security, and ethics. Paper 2 (Fundamental Problem-solving and Programming) is a 2-hour practical programming paper worth 75 marks, requiring you to write, trace, and debug code — predominantly in pseudocode. Both papers carry equal weight and are taken in the same exam series.

CAIE AS 计算机科学由两门必考试卷组成。Paper 1（理论基础）是 1 小时 30 分钟的笔试，满分 75 分，涵盖信息表示、系统软件、安全与伦理等主题。Paper 2（问题解决与编程基础）是 2 小时的编程实践卷，满分 75 分，要求编写、追踪和调试代码——主要以伪代码形式出现。两卷权重相同，在同一考试季进行。

Paper 1 is divided into two sections. Section A (approximately 30 marks) contains short-answer questions that test breadth across the syllabus. Section B (approximately 45 marks) has structured questions that may draw from multiple topics and often require extended writing — for example, comparing processor architectures, explaining how an operating system manages memory, or evaluating the ethical implications of a computing technology. Paper 2 is a single structured paper where you will write pseudocode solutions, complete trace tables, identify and correct errors in given code, and answer questions about program design.

Paper 1 分两部分。Section A（约 30 分）为简答题，考察知识面的广度。Section B（约 45 分）为结构化题目，常涉及多个主题的综合，并需要扩展性作答——例如比较处理器架构、解释操作系统如何管理内存，或评估某项计算技术的伦理影响。Paper 2 是统一的结构化试卷，你需要编写伪代码、完成追踪表、找出并修正代码中的错误，以及回答程序设计的相关问题。

2. Mastering Command Words | 掌握指令词

Every question on the CAIE AS Computer Science paper contains command words — specific verbs that tell you exactly what the examiner expects. Misinterpreting a command word is one of the most common and costly mistakes students make. Here are the key command words and what they demand from you:

CAIE AS 计算机科学试卷中的每道题都包含指令词——这些动词准确告诉了你考官的期望。误解指令词是学生最常见且代价最高的错误之一。以下是关键指令词及其要求：

State / Identify / Name — These require a short, factual answer with no explanation. A single word or phrase is often sufficient. For example: “State the binary representation of denary 42.” The answer is simply “00101010”. Do not waste time explaining how you got there.

State / Identify / Name（指出/识别/命名）——这类题目要求简短的事实性答案，无需解释。通常一个词或短语就足够了。例如：”指出十进制 42 的二进制表示。”答案只需写”00101010″。不要浪费时间解释推导过程。

Describe — You need to state the key characteristics or features of a concept. Unlike “state”, a “describe” answer requires several points linked together. For instance: “Describe the role of the Control Unit in a processor.” Your answer should mention fetching instructions, decoding them, generating control signals, and coordinating the ALU and memory — not just “it controls the processor.”

Describe（描述）——你需要陈述概念的关键特征或属性。与”state”不同，”describe”类答案需要将多个要点串联起来。例如：”描述处理器中控制单元的作用。”你的答案应包括提取指令、解码指令、生成控制信号、协调 ALU 和内存——而不仅仅是”它控制处理器”。

Explain — Go beyond description to give reasons or causes. “Explain” questions demand the “why” and “how”. Use connecting phrases like “because”, “therefore”, and “this means that”. Example: “Explain why the operating system uses paging for memory management.” You need to describe what paging is AND explain why it solves problems like external fragmentation and allows programs larger than physical memory to run.

Explain（解释）——超越描述，给出原因或因果关系。”Explain”类题目要求回答”为什么”和”如何”。使用连接词如”because”、”therefore”和”this means that”。例如：”解释为什么操作系统使用分页进行内存管理。”你需要描述什么是分页，并且解释为什么它能解决外部碎片问题，以及为什么它允许大于物理内存的程序运行。

Compare / Contrast — Identify similarities AND differences between two or more concepts. A common pitfall is only listing differences. The examiner expects a balanced treatment. Use comparative language: “Both use…”, “X is faster than Y because…”, “Unlike X, Y…”. For high marks, structure your comparison point by point rather than describing each item separately.

Compare / Contrast（比较/对比）——识别两个或多个概念之间的相似和不同之处。常见误区是只列出区别。考官期望平衡地处理两个方面。使用比较性语言：”两者都使用…”、”X 比 Y 更快，因为…”、”与 X 不同，Y…”。为获得高分，应逐点比较，而不是分别描述每个项目。

Evaluate / Discuss — These are the highest-order command words. You need to present arguments for and against, then reach a reasoned conclusion. In the CAIE context, “evaluate” means making a judgment based on given criteria. For example: “Evaluate the use of SSDs over HDDs in a file server.” You must discuss advantages (speed, durability, lower power) AND disadvantages (higher cost per GB, limited write cycles), then conclude which is more suitable for the specific scenario described.

Evaluate / Discuss（评估/讨论）——这是最高阶的指令词。你需要呈现正反两方面的论据，然后得出有理有据的结论。在 CAIE 的语境中，”evaluate”意味着基于给定标准做出判断。例如：”评估在文件服务器中使用 SSD 而非 HDD。”你必须讨论优点（速度、耐用性、低功耗）和缺点（每 GB 成本较高、写入次数有限），然后针对所述的具体场景得出哪种更合适的结论。

3. How Marks Are Awarded: Decoding the Mark Scheme | 评分解码：读懂评分标准

CAIE mark schemes follow a consistent logic that, once understood, can dramatically improve your answer quality. Each mark on the scheme corresponds to a specific, verifiable point. The examiner is not judging the overall “quality” of your answer — they are ticking off whether you have made each of the required points. This means your goal is to make it as easy as possible for the examiner to award marks.

CAIE 的评分标准遵循一套一致的逻辑，一旦理解，就能显著提升你的答题质量。评分标准上的每一分都对应一个具体、可验证的要点。考官不是在评判你答案的整体”质量”——他们是在逐条核对你是否写出了每一个要求的得分点。这意味着你的目标是让考官尽可能容易地给你分。

Explicit over implicit. Never assume the examiner will infer what you meant. If the mark scheme says “1 mark for identifying that the parity bit can only detect, not correct, errors”, you must explicitly say those words — an answer like “parity checks are limited” may be true but will not earn the mark. Be direct, be precise, and always use the correct technical vocabulary from the syllabus.

明确优于隐含。永远不要假设考官会推断出你的意思。如果评分标准写着”识别奇偶校验位只能检测错误而无法纠正——1 分”，你必须明确写出这些词——像”奇偶校验有局限性”这样的回答虽然没错，但拿不到分。要直接、精确，并始终使用考纲中的正确技术术语。

Bullet points and structure. For Section B questions worth 4–8 marks, structure your answer into clear paragraphs or numbered points. This doesn’t just make your answer more readable — it directly mirrors the mark scheme, which is itself a list of markable points. When an examiner sees a well-structured answer with clear technical points, marking becomes faster and more accurate, and you are less likely to lose marks through omission.

使用要点和结构化答题。对于 Section B 中值 4–8 分的题目，将答案组织成清晰的段落或编号要点。这不仅让答案更易读——它直接对应评分标准，而评分标准本身就是一个计分点的列表。当考官看到一个结构清晰、技术要点明确的答案时，评分变得更快更准确，你也不太可能因遗漏而丢分。

The mark scheme also reveals which points are “independent” and which are “dependent”. An independent mark can be earned regardless of previous answers; a dependent mark (often marked with “A1” for accuracy or “ft” for follow-through) depends on a correct preceding step. This is especially relevant in Paper 2 trace table questions, where one wrong value early on can cascade. In such cases, the examiner applies “error carried forward” (ECF) where possible — but only if your working is clear and the error is consistent.

评分标准还揭示了哪些点是”独立”的，哪些是”依赖”的。独立得分点不论前面的答案正确与否都可以拿到；依赖得分点（通常标记为 A1 即准确性分，或 ft 即结转分）取决于前一步的正确性。这在 Paper 2 的追踪表题目中尤其重要——早期的一个错误值可能引起连锁反应。在这种情况下，考官会尽可能使用”错误结转”（ECF）给分——但前提是你的解题过程清晰且错误是一致的。

4. Paper 1 Strategy: Theory with Precision | Paper 1 策略：精准作答理论题

Paper 1 rewards breadth and precision. With 75 marks in 90 minutes, you have just over one minute per mark — this means you cannot afford to write long introductions or repeat yourself. Every sentence should carry a markable point, and you should complete shorter questions rapidly to reserve time for the extended Section B answers.

Paper 1 考察的是知识的广度和精确度。75 分、90 分钟意味着每分钟约 0.8 分——你写不出冗长的引言，也不能重复自己。每句话都应该承载一个可计分的要点，你应该快速完成简答题，为 Section B 的扩展性答案留出时间。

Information Representation (Chapter 1). Expect at least one binary/hexadecimal conversion, a question on binary addition or two’s complement, and a question about representing text, sound, or images. These are process-heavy marks — practice until you can convert without hesitation. For image and sound representation, learn the formulas for file size calculation: for images, width × height × colour depth (in bits); for sound, sample rate × sample resolution × duration × number of channels. Show your working clearly, as method marks are often available even if the final answer is wrong.

信息表示（第 1 章）。预计至少有一道二进制/十六进制转换题、一道二进制加法或补码题，以及一道关于文本、声音或图像表示的题目。这些是过程分密集型题目——练习到可以毫不犹豫地转换的程度。对于图像和声音的表示，记住文件大小计算公式：图像 = 宽度 × 高度 × 色深（bit）；声音 = 采样率 × 采样精度 × 时长 × 声道数。清晰地展示解题步骤，因为即使最终答案错误，方法分通常仍然可以获得。

Processor Fundamentals (Chapter 3) and System Software (Chapter 6). These chapters frequently appear in Section B as extended-response questions. Be ready to explain the Von Neumann architecture (MAR, MDR, PC, CIR, ACC, Control Unit, ALU) and how they interact during the fetch-decode-execute cycle. For operating systems, master the following: memory management (paging, segmentation, virtual memory), process scheduling (round-robin, priority-based, shortest job first), and interrupt handling. Use diagrams where appropriate — a simple labelled sketch of the Von Neumann architecture can earn multiple marks even before you write a word of explanation.

处理器基础（第 3 章）和系统软件（第 6 章）。这些章节在 Section B 中常以扩展性回答的形式出现。准备好解释冯·诺依曼架构（MAR、MDR、PC、CIR、ACC、控制单元、ALU）以及它们在取指-解码-执行周期中如何交互。对于操作系统，掌握以下内容：内存管理（分页、分段、虚拟内存）、进程调度（轮转调度、优先级调度、最短作业优先）和中断处理。在适当的地方使用图示——一个简单标注的冯·诺依曼架构草图可以在你写任何解释文字之前就拿到多分。

Security, Privacy, and Data Integrity (Chapter 8) and Ethics (Chapter 9). These topics often appear as 6–8 mark evaluate/discuss questions. The key is to structure your answer around specific technical mechanisms rather than vague statements. For example, when asked about data security, don’t just say “use encryption” — specify that asymmetric encryption uses a public key for encryption and a private key for decryption, mention typical key lengths, and explain what threats it protects against. For ethical questions, use a framework: identify the stakeholders involved, the potential harms and benefits, and any relevant legislation (e.g., Data Protection Act 2018, Computer Misuse Act 1990 in the UK context).

安全、隐私与数据完整性（第 8 章）和伦理（第 9 章）。这些主题通常以 6–8 分的评估/讨论题形式出现。关键在于围绕具体的技术机制组织答案，而非泛泛而谈。例如，当被问及数据安全时，不要只说”使用加密”——明确指出非对称加密使用公钥加密、私钥解密，提及典型的密钥长度，并解释它能防范哪些威胁。对于伦理问题，使用框架分析法：识别涉及的各方利益相关者、潜在的危害和收益，以及相关法规（例如英国的 2018 年数据保护法和 1990 年计算机滥用法）。

5. Paper 2 Strategy: Programming and Problem-Solving | Paper 2 策略：编程与问题解决

Paper 2 is where many students find the greatest challenge — but also the greatest opportunity. The exam tests both your ability to understand and write pseudocode and your problem-solving skills. The 2-hour duration for 75 marks gives you approximately 1.6 minutes per mark, slightly more generous than Paper 1, but the questions require deeper thinking.

Paper 2 是许多学生觉得最具挑战性的部分——但也是最大的得分机会。考试既测试你理解和编写伪代码的能力，也测试你的问题解决能力。2 小时、75 分，约 1.6 分钟/分，略比 Paper 1 宽松，但题目需要更深入的思考。

Trace tables are guaranteed marks — master them. Trace tables typically appear in every Paper 2 and can account for 5–10 marks. The process is systematic: create columns for every variable and output, step through the code line by line, and record the value of each variable after each change. Pay special attention to loop conditions — a common mistake is forgetting to update the loop variable at the correct point or misreading the termination condition. If a trace table has many rows, check whether the question allows you to write “cycles” or abbreviated entries for repeated patterns — CAIE mark schemes sometimes accept “repeats for next 4 iterations” with a note of the final value, saving you time.

追踪表是稳拿的分数——彻底掌握它。追踪表几乎每份 Paper 2 都会出现，可占 5–10 分。过程是系统性的：为每个变量和输出创建列，逐行遍历代码，记录每次变化后每个变量的值。特别注意循环条件——常见错误包括忘记在正确的位置更新循环变量，或误读终止条件。如果追踪表有很多行，检查题目是否允许你为重复模式写”循环”或缩略条目——CAIE 评分标准有时接受”重复后续 4 次迭代”并注明终值，帮你节省时间。

Pseudocode writing — be precise and consistent. CAIE publishes an official pseudocode guide with specific syntax for variable declaration, assignment, selection, iteration, input/output, and file handling. The examiner expects you to follow this syntax. Key points: use DECLARE for declarations, OUTPUT for printing, INPUT for reading, and the standard IF-THEN-ELSE-ENDIF and FOR-TO-NEXT or WHILE-DO-ENDWHILE structures. Indent your code clearly — while not directly marked, indentation makes your logic clear and helps the examiner follow your algorithm, increasing the likelihood of earning method marks even when the exact output is wrong.

伪代码编写——精确且一致。CAIE 发布了官方的伪代码指南，规定了变量声明、赋值、选择、迭代、输入/输出和文件处理的具体语法。考官期望你遵循这套语法。关键点：使用 DECLARE 声明变量，OUTPUT 打印，INPUT 读取，以及标准的 IF-THEN-ELSE-ENDIF 和 FOR-TO-NEXT 或 WHILE-DO-ENDWHILE 结构。清晰地缩进你的代码——虽然不直接计分，但缩进使你的逻辑清晰，帮助考官理解你的算法，增加拿到方法分的可能性，即使最终输出有误。

Error identification and correction. Paper 2 often includes a section where you are given buggy code and asked to identify and correct errors. Approach this systematically: first, trace through the provided code with sample test data to understand what it actually does. Then, compare this with what the specification says it should do. Common error types in CAIE exam questions include: off-by-one errors in loop bounds, using the wrong logical operator (AND instead of OR), missing initialisation of variables before use, incorrect file handling mode (READ vs WRITE vs APPEND), and logic errors where the algorithm does not account for edge cases like empty inputs or boundary values.

错误识别与修正。Paper 2 经常包含一段有错误的代码，要求你识别并修正错误。系统性地处理：首先，用示例测试数据追踪给定代码，理解它实际做了什么。然后，将此与题目规定的预期行为进行比较。CAIE 考试中常见的错误类型包括：循环界限的差一错误、使用错误的逻辑运算符（AND 而非 OR）、变量使用前未初始化、错误的文件操作模式（READ vs WRITE vs APPEND），以及算法未考虑边界情况的逻辑错误，如空输入或边界值。

6. Time Management and Exam Technique | 时间管理与考场技巧

Effective time management is perhaps the single most impactful skill you can develop. For Paper 1, allocate approximately 1 minute per mark. This means a 3-mark short question should take no more than 3 minutes, and an 8-mark extended question roughly 8 minutes. If you are stuck on a question after spending your allocated time, move on — an unanswered 1-mark question you never reach is worse than an incomplete 3-mark question where you at least picked up partial marks.

有效的时间管理可能是你能培养的最具影响力的单项技能。对于 Paper 1，大约按每分钟 1 分分配时间。这意味着 3 分的简答题不应超过 3 分钟，8 分的扩展题大约 8 分钟。如果你在某题上超时卡住，继续前进——一个你根本没来得及做的 1 分题比一个你至少拿了部分分的 3 分未完成题更糟。

The three-pass approach. On your first pass through the paper, answer every question you are confident about. On the second pass, tackle the questions you know how to do but need more time for. On the third pass, address the difficult questions where you are unsure. This guarantees you collect the “low-hanging fruit” first and do not run out of time on easy marks while struggling with a hard question early on. For Paper 2, this approach is even more crucial — a challenging trace table early in the paper should not consume time from the programming question later that you can solve comfortably.

三遍答题法。第一遍阅读试卷时，回答所有你有把握的题目。第二遍，处理你知道怎么做但需要更多时间的题目。第三遍，解决你不确定的难题。这保证了你先摘取”低垂的果实”，不会因为早期与一道难题死磕而耗尽时间，丢了容易拿的分。对于 Paper 2，这种方法更为关键——试卷早期的一道复杂追踪表不应消耗掉你后面能轻松解决的编程题的时间。

Read the question — twice. Before writing anything, read the full question carefully. Underline the command word and any constraints or conditions. A question that says “Write pseudocode to input 10 numbers and output the largest” is different from “Write pseudocode to input numbers until 999 is entered and output the largest”. The difference between a fixed loop (FOR loop) and a conditional loop (WHILE loop) is often the key to the entire question. Missing that distinction costs you most or all of the marks, regardless of how elegant your pseudocode is.

读题——读两遍。在写任何东西之前，仔细通读整道题目。划出指令词和任何约束条件。一道说”编写伪代码输入 10 个数字并输出最大值”的题目不同于”编写伪代码输入数字直到输入 999 为止并输出最大值”。固定循环（FOR）与条件循环（WHILE）之间的区别往往是整道题的关键。错过这个区别会让你丢掉大部分甚至全部分数，无论你的伪代码写得多么优美。

7. Common Mistakes and How to Avoid Them | 常见错误与避免方法

Mistake 1: Confusing bits and bytes in calculations. When calculating file sizes, remember that file size is typically measured in bytes, but colour depth and sample resolution are given in bits. Always convert: 1 byte = 8 bits. Many students lose easy marks by forgetting to divide by 8 at the end, producing answers that are 8 times too large.

错误一：在计算中混淆 bit 和 byte。计算文件大小时，记住文件大小通常以字节（byte）为单位，但色深和采样精度以位（bit）为单位。始终进行转换：1 byte = 8 bits。很多学生忘记在最后除以 8，得到的结果大了 8 倍，丢了送分题。

Mistake 2: Describing instead of explaining. When the question asks you to “explain”, describing what happens is insufficient. You need to explain WHY it happens. For example, “The operating system uses virtual memory” is description. “The operating system uses virtual memory so that programs larger than physical RAM can execute by swapping pages between RAM and secondary storage” is explanation. Always check: have I answered why?

错误二：描述而非解释。当题目要求”explain”时，仅描述发生的事情是不够的。你需要解释它为什么会发生。例如，”操作系统使用虚拟内存”是描述。”操作系统使用虚拟内存，以便大于物理 RAM 的程序能够通过在 RAM 和辅助存储之间交换页面来执行”才是解释。始终检查：我回答了为什么吗？

Mistake 3: Poor pseudocode variable naming. Using single-letter variables like “a”, “b”, “x” makes your code hard to follow — both for the examiner and for yourself. Use meaningful names: “Total”, “LargestValue”, “StudentName”. This also demonstrates professional practice, which CAIE explicitly values in the syllabus objectives. If the question specifies variable names, use exactly those names — do not rename them.

错误三：伪代码变量命名不当。使用单字母变量如”a”、”b”、”x”让你的代码难以理解——对考官和你自己都是如此。使用有意义的名称：”Total”、”LargestValue”、”StudentName”。这也展示了专业实践能力，这正是 CAIE 考纲目标中明确重视的。如果题目指定了变量名，就使用正好那些名称——不要重新命名。

Mistake 4: Ignoring edge cases in programming solutions. A pseudocode solution that works for the “happy path” may fail the mark scheme if it ignores edge cases. Always consider: what if the input is empty? What if all numbers are negative? What if the search target is not found? What if the array index is out of bounds? Adding a single IF statement to handle an edge case can earn a mark that separates a good answer from a great one.

错误四：编程方案中忽略边界情况。一个在”理想路径”下能运行的伪代码方案，如果忽略了边界情况，可能不符合评分标准。始终考虑：如果输入为空怎么办？如果所有数字都是负数怎么办？如果搜索目标未找到怎么办？如果数组索引越界怎么办？添加一个 IF 语句来处理边界情况，就能拿到区分好答案和优秀答案的那一分。

Mistake 5: Answering what you know, not what is asked. This is especially common in “evaluate” questions. If the question is “Evaluate the use of cloud storage for a multinational corporation”, do not write everything you know about cloud storage. Focus specifically on the scenario — discuss latency across geographic regions, data sovereignty laws in different countries, the cost of bandwidth for large data transfers, and the reliability needs of a multinational. Irrelevant information wastes time and does not earn marks, even if technically correct.

错误五：回答你知道的，而不是题目问的。这在”evaluate”类题目中尤其常见。如果题目是”评估云存储对跨国公司的使用”，不要把你知道的关于云存储的所有内容都写出来。聚焦具体场景——讨论跨地理区域的延迟问题、不同国家的数据主权法律、大数据传输的带宽成本，以及跨国公司的可靠性需求。不相关的信息浪费时间且不得分，即使技术上正确。

8. Revision Strategy: Working Smarter | 复习策略：聪明备考

The most effective revision for CAIE AS Computer Science is past-paper practice with active mark scheme engagement. This means not just doing past papers, but marking your own answers against the official mark scheme and understanding exactly why you lost marks. A common pattern: students complete 10 past papers but make the same mistakes on the 10th as they did on the 1st. The key is to do fewer papers but learn more from each one.

CAIE AS 计算机科学最有效的复习方法是结合评分标准的真题练习。这意味着不仅仅是做真题，还要对照官方评分标准自己批改，并确切理解为什么丢了分。常见模式：学生做了 10 份真题，但在第 10 份上犯了和第 1 份相同的错误。关键在于少做题、多总结。

For Paper 1, create topic-specific quick-reference cards. Each card should contain the key points that earn marks for that topic. For example, your “Von Neumann Architecture” card should list every register, the buses (data, address, control), the fetch-decode-execute cycle steps in order, and common factors affecting performance (clock speed, bus width, cache size, number of cores). Make these cards exam-ready — the points on them should match the granularity of mark scheme points.

对于 Paper 1，制作主题专项速记卡。每张卡应包含该主题的得分要点。例如，你的”冯·诺依曼架构”卡应列出每一个寄存器、总线（数据、地址、控制）、取指-解码-执行周期步骤顺序，以及影响性能的常见因素（时钟速度、总线宽度、缓存大小、核心数量）。让这些卡片达到应试标准——上面的要点应与评分标准中计分点的粒度相匹配。

For Paper 2, establish a weekly programming routine. Write pseudocode for common patterns: linear search, binary search, bubble sort, finding minimum/maximum, counting occurrences, validating input, reading from and writing to files, and manipulating both 1D and 2D arrays. Time yourself: can you write a complete binary search algorithm with proper initialisation, loop condition, mid-point calculation, comparison logic, and boundary updates in under 5 minutes? Automated, fluent pseudocode writing frees up mental capacity during the exam for the actual problem-solving.

对于 Paper 2，建立每周编程常规。为常见模式编写伪代码：线性搜索、二分搜索、冒泡排序、查找最小值/最大值、计数出现次数、验证输入、文件读写，以及一维和二维数组的操作。给自己计时：你能在 5 分钟内写出一个完整的二分搜索算法，包含正确的初始化、循环条件、中点计算、比较逻辑和边界更新吗？自动化的、流畅的伪代码写作能在考试中为实际问题解决腾出脑力。

9. Exam Day Checklist | 考试日清单

On the day of the exam, arrive with a clear plan. Bring multiple black or blue pens, a pencil and eraser for diagrams and trace tables, a ruler for drawing neat tables and architecture diagrams, and a clear, non-programmable calculator for binary/hexadecimal conversions. Read the instructions on the front of the paper carefully — they specify which questions are compulsory and any options you may have. Before writing, scan the entire paper and identify the questions you are most confident about, then begin with those. And finally: if you finish early, use every remaining minute to check your answers. Many students lose marks to careless arithmetic errors that a second look would catch.

考试当天，带着清晰的计划到达考场。携带多支黑色或蓝色笔、用于图表和追踪表的铅笔和橡皮、用于画整齐表格和架构图的尺子，以及一台透明的非可编程计算器用于二进制/十六进制转换。仔细阅读试卷首页的说明——它们规定了哪些是必答题以及你可能有的选择题。在动笔之前，浏览整份试卷，找出你最自信的题目，然后从那些开始。最后：如果你提前完成，利用每一分钟剩余时间检查答案。很多学生因为粗心的算术错误丢分，而第二次检查就能发现。

Remember: the CAIE AS Computer Science exam is designed to test your understanding of computing principles and your ability to apply them. The examiners want to award marks — your job is to make it easy for them to do so. Be clear, be precise, and show your working. Every mark counts, and with the right technique, you can achieve the grade you deserve.

记住：CAIE AS 计算机科学考试旨在测试你对计算机原理的理解和应用能力。考官希望给你分数——你的工作就是让他们容易做到这一点。要清晰、精确，展示你的解题过程。每一分都重要，有了正确的技巧，你就能取得应得的成绩。