📚 Exam Technique and Marking Criteria for Year 10 Cambridge Computer Science | 剑桥10年级计算机科学：答题技巧与评分标准
Mastering Cambridge IGCSE Computer Science requires more than just knowing the theory and programming concepts; you must also understand how to present your answers to meet the examiners' expectations. This article provides essential exam techniques, highlights common pitfalls, and explains the marking criteria so you can boost your score.
掌握剑桥 IGCSE 计算机科学不仅需要了解理论和编程概念，还必须理解如何呈现答案以满足考官的期望。本文提供必备的答题技巧，指出常见陷阱，并解释评分标准，助你提高分数。
1. Understanding the Exam Papers | 了解试卷结构
The Cambridge IGCSE Computer Science (0478) assessment consists of two written papers. Paper 1 covers Computer Systems and is worth 75 marks. It tests data representation, hardware, software, networking, security, and ethics.
剑桥 IGCSE 计算机科学（0478）的评估包含两份笔试。试卷 1 涵盖计算机系统，满分 75 分。它考查数据表示、硬件、软件、网络、安全以及伦理。
Paper 2 focuses on Algorithms, Programming and Logic, also worth 75 marks. You will be asked to write pseudocode, draw flowcharts, trace algorithms, and solve logic problems. Both papers last 1 hour 45 minutes and feature a mix of short-answer and structured questions.
试卷 2 侧重于算法、编程与逻辑，同样是 75 分。你将被要求编写伪代码、绘制流程图、追踪算法并解决逻辑问题。两份试卷的时长均为 1 小时 45 分钟，题型包括简答题和结构性问题。
Knowing the paper structure helps you allocate mental effort. Paper 1 often requires precise definitions and comparison, while Paper 2 demands step-by-step logic and problem-solving. Always check the number of marks per question to gauge how much detail to include.
了解试卷结构有助于合理分配精力。试卷 1 通常要求准确的定义和比较，而试卷 2 要求逐步的逻辑推演和问题解决。务必根据题目分值判断需要包含多少细节。
2. Command Words and Their Meanings | 指令词及其含义
Command words tell you exactly what the examiner expects. Misinterpreting a command word is one of the most common reasons for losing marks. The table below lists frequently used command words and their requirements.
指令词准确告诉你考官的期待。误解指令词是丢分最常见的原因之一。下表列出了常见的指令词及其要求。
| Command Word | Meaning | 中文含义 |
|---|---|---|
| State / Give / Name | Provide a brief answer, no explanation needed. | 陈述 / 给出 / 命名（简明回答） |
| Describe | Give a detailed account, typically including characteristics. | 描述（详细说明特征） |
| Explain | Give reasons or causes, linking cause and effect. Use 'because'. | 解释（用“因为”来阐述因果） |
| Calculate | Work out a numerical answer, showing steps for partial credit. | 计算（展示步骤以获取过程分） |
| Draw / Design | Produce a diagram, flowchart, or structured algorithm. | 绘制 / 设计（生成图表或算法） |
| Identify / Select | Pick the correct option or recognisable feature. | 识别 / 选择（挑出正确项） |
For an 'Explain' question, always write in full sentences and use clear causal connectors like 'because', 'therefore', or 'as a result'. For 'State', a single word or short phrase may be accepted, but ensure it is technically accurate.
对于“解释”题型，一定要用完整的句子，并使用“因为”“因此”“结果是”等清楚的因果连接词。对于“陈述”，一个词或短语即可，但必须确保术语准确无误。
3. Structuring Answers for Maximum Marks | 构建答案以获得最高分
Even when you know the correct information, poor structure can cost you marks. Examiners mark positively, looking for evidence of understanding rather than deducting points for every error.
即便你知道正确信息，糟糕的答案结构也可能导致失分。考官采用正面评分，寻找理解的证据，而并非对每个错误都扣分。
For longer 5–8 mark questions, break your answer into logical parts. Use bullet points or numbered steps if the question asks for a sequence, such as in an algorithm trace or a list of stages. Keep your handwriting legible and leave space between answers so corrections can be added clearly.
对于5–8分的较长的题目，将答案分解成合乎逻辑的部分。如果题目要求序列，例如追踪算法或列出阶段，可使用要点或编号步骤。保持字迹清晰，答案之间留出空隙，以便清楚地进行修正。
In comparison questions, use phrases like 'whereas' or 'on the other hand' to highlight differences. For example, when comparing RAM and ROM, structure your answer: 'RAM is volatile and can be written to, whereas ROM is non-volatile and holds firmware.' This directly addresses the command word 'compare'.
在对比题中，使用“而”“另一方面”等短语来突出差异。例如，比较 RAM 和 ROM 时，这样组织答案：“RAM 是易失性的且可写入，而 ROM 是非易失性的且存放固件。”这直接回应了指令词“比较”。
4. Using Technical Terminology Accurately | 准确使用技术术语
Cambridge examiners award marks for correct use of subject-specific vocabulary. Casual language like 'the brain of the computer' is not acceptable; you must write 'central processing unit (CPU)'.
剑桥考官对于准确使用学科词汇给予分数。像“计算机的大脑”这样的随意用语是不被接受的，你必须写出“中央处理器（CPU）”。
Always spell acronyms out in full on first use if the question contains a bracketed abbreviation, e.g. 'Arithmetic Logic Unit (ALU)'. Do not coin your own abbreviations. Be precise: 'data bus' is not the same as 'address bus'.
如果题目带有括号缩写，首次使用时一定要写出全称，例如“算术逻辑单元（ALU）”。不要自创缩写。务必精确：“数据总线”和“地址总线”并不相同。
Common errors include confusing 'input device' with 'output device', writing 'megahertz' when you mean 'megahertz (MHz)', and using 'internet' and 'World Wide Web' interchangeably. The internet is the global network infrastructure; the web is a service that uses it.
常见错误包括混淆“输入设备”与“输出设备”，在你指的是“兆赫（MHz）”时写成“兆赫”，以及交替使用“互联网”和“万维网”。互联网是全球网络基础设施，而万维网是使用它的服务。
5. Tackling Pseudocode Questions | 处理伪代码问题
Pseudocode questions test your algorithmic thinking, not your knowledge of a specific programming language. Use the Cambridge-approved pseudocode style: keywords in uppercase (e.g., INPUT, OUTPUT, IF, THEN, ELSE, ENDIF, FOR, TO, NEXT, WHILE, DO, ENDWHILE, REPEAT, UNTIL).
伪代码问题考查的是你的算法思维，而非特定编程语言的知识。使用剑桥认可的伪代码风格：关键字大写（例如 INPUT、OUTPUT、IF、THEN、ELSE、ENDIF、FOR、TO、NEXT、WHILE、DO、ENDWHILE、REPEAT、UNTIL）。
Indent code blocks inside loops and conditional statements to show structure. For example:
在循环和条件语句内部缩进代码块以展示结构。例如：
FOR Count ← 1 TO 10
OUTPUT Count
NEXT Count
Always initialise variables before using them. Use '←' for assignment. Comment your pseudocode only if the question asks for it. Avoid mixing syntaxes from different languages; stick to one consistent convention.
始终在使用变量前将其初始化。用“←”进行赋值。仅在题目要求时为伪代码添加注释。不要混用不同语言的语法；坚持使用一致的规则。
6. Drawing Flowcharts | 绘制流程图
Flowcharts must use the standard symbols: a rounded rectangle (or oval) for Start/End, a rectangle for process steps, a parallelogram for input/output, and a diamond for a decision. Arrows should clearly show the flow from top to bottom and left to right.
流程图必须使用标准符号：圆角矩形（或椭圆）表示开始/结束，矩形表示处理步骤，平行四边形表示输入/输出，菱形表示判断。箭头应清晰显示从上到下、从左到右的流向。
Label decision branches with 'Yes' and 'No' or 'True' / 'False'. Do not leave branches unlabelled. Each symbol should contain concise, actionable text, such as 'Read temperature' rather than 'Input from sensor'.
在判断分支上标注“是”和“否”或“真”/“假”。不要留下未标注的分支。每个符号内应包含简洁可操作的文字，如“读取温度”而非“来自传感器的输入”。
When a flowchart is assessed, examiners look for correct logic, completeness of loops, and proper termination. Avoid overlapping lines; if a flowchart becomes messy, redraw it neatly. A clear flowchart can earn full marks even if the pseudocode equivalent has minor errors.
当流程图被评估时，考官关注逻辑的正确性、循环的完整性以及恰当的终止。避免线条重叠；如果流程图变得杂乱，就重新整洁地绘制。清晰的流程图即使其对应的伪代码有轻微错误，也可能获得满分。
7. Data Representation Calculations | 数据表示计算
You must be comfortable converting between binary, denary, and hexadecimal, and performing binary addition. Always show your working. For example, convert denary 45 to binary:
你必须能熟练地在二进制、十进制和十六进制之间进行转换，并能执行二进制加法。始终展示你的计算过程。例如，将十进制 45 转换为二进制：
45₁₀ = 32 + 8 + 4 + 1 = 2⁵ + 2³ + 2² + 2⁰ = 00101101₂
When adding binary numbers, write the carry row explicitly:
进行二进制加法时，应明确写出进位行：
1011₂
+ 0110₂
───────
10001₂
Be aware of overflow errors: if an 8-bit register holds a result that exceeds 255, the overflow bit will be set, and the answer wraps around. For hexadecimal, use the letters A–F for values 10–15. Conversion can be checked by splitting the binary into nibbles (4-bit groups).
要注意溢出错误：如果一个 8 位寄存器存放的结果超过 255，溢出位将被置位，且答案会回绕。对于十六进制，使用字母 A–F 表示数值 10–15。转换可以通过将二进制划分为半字节（4 位一组）来检查。
8. Memory, Storage and Data Units | 内存、存储与数据单位
= 2²⁰ bytes, 1 GB = 2³⁰ bytes, 1 TB = 2⁴⁰ bytes. Use the correct symbols and powers of 2.
你需要知道各单位之间的关系：1 nibble = 4 bits，1 byte = 8 bits，1 KB = 2¹⁰ bytes = 1024 bytes，1 MB = 2²⁰ bytes，1 GB = 2³⁰ bytes，1 TB = 2⁴⁰ bytes。要使用正确的符号和 2 的幂次。
h, the uncompressed size is 1024 × 768 × 24 = 18,874,368 bits. Convert to bytes by dividing by 8, then to KB by dividing by 1024.
计算文件大小时，应用公式：文件大小 = 分辨率（宽 × 高）× 色彩深度。对于一个 1024 × 768 像素、24 位色彩深度的图像，未压缩大小为 1024 × 768 × 24 = 18,874,368 bits。除以 8 转换为字节，再除以 1024 转换为 KB。
Differentiate between primary memory (RAM, ROM) and secondary storage (HDD, SSD, optical). Know that RAM is volatile and holds currently running programs and data, while ROM stores the bootstrap loader and is non-volatile.
区分主存储器（RAM, ROM）和辅助存储器（HDD, SSD, 光盘）。要知道 RAM 是易失性的，存放当前运行的程序和数据，而 ROM 存储引导程序，是非易失性的。
9. Logic Gates and Truth Tables | 逻辑门与真值表
You will be asked to complete truth tables for circuits with up to three inputs, and to draw logic circuits from Boolean expressions. Use standard symbols: a D-shaped AND gate, curved-back OR gate, and triangle with a bubble for NOT.
你会被要求完成最多三输入的电路真值表，并根据布尔表达式绘制逻辑电路。使用标准符号：D 形的与门、弧形的或门，以及带小圆圈三角的的非门。
es. For an expression like Q = (A AND B) OR C, create columns for A AND B, then OR with C.
书写真值表时，按二进制顺序列出所有输入组合（000、001、010……）。逐步推算出中间门的状态。对于表达式 Q = (A AND B) OR C，先创建 A AND B 列，再与 C 做 OR 运算。
A AND 1 = A, A OR 1 = 1. These are useful for simplifying expressions in longer questions.
记住标准的布尔等式：A AND 0 = 0，A OR 0 = A，A AND 1 = A，A OR 1 = 1。这些在较长题目中简化表达式时很有用。
10. Avoiding Common Mistakes | 避免常见错误
Many marks are lost through avoidable slips. In pseudocode, forgetting to close a loop with ENDWHILE or NEXT is a frequent error. Always check that every IF has a matching ENDIF and every FOR has a NEXT.
许多失分都是由于可以避免的疏忽造成的。在伪代码中，忘记用 ENDWHILE 或 NEXT 结束循环是一个常见错误。务必检查每个 IF 都有匹配的 ENDIF，每个 FOR 都有匹配的 NEXT。
In hardware questions, do not confuse a switch with a router or a hub. A switch forwards data to a specific device using MAC addresses, whereas a hub sends data to all ports. Mixing up HTTP and HTTPS or their default ports (80 and 443) will lose marks.
在硬件题目中，不要混淆交换机和路由器或集线器。交换机使用 MAC 地址将数据转发到特定设备，而集线器将数据发送到所有端口。混淆 HTTP 和 HTTPS 或它们的默认端口（80 和 443）会丢分。
In calculations, forgetting to show working can result in zero credit even if the final answer is correct. Always write intermediate steps. When tracing an algorithm, use a trace table with columns for each variable, and update row by row.
在计算题中，忘记展示过程可能导致即使最终答案正确也得不到分。始终写出中间步骤。追踪算法时，使用带有各变量列的追踪表，逐行更新。
11. Time Management Strategies | 时间管理策略
With 75 marks in 105 minutes, you have approximately 1.4 minutes per mark. In practice, allocate roughly 1 minute per mark for shorter questions and save more time for the extended algorithm design tasks.
在 105 分钟内完成 75 分，大约每分对应 1.4 分钟。实际上，较短的题目可按每分 1 分钟左右分配，将更多时间留给扩展性的算法设计任务。
Read through the entire paper during the first 3 minutes. Identify questions you find easy and answer them first to build confidence. Circle command words and key data. For a 10-mark programming scenario, spend no more than 15–18 minutes planning and writing.
在最初 3 分钟内通读整份试卷。找出你擅长的题目先做，建立信心。圈出指令词和关键数据。值 10 分的编程场景题，计划与写作时间不应超过 15–18 分钟。
If you are stuck on a question, mark it and move on. Never leave a question blank: even an incomplete trace or partial pseudocode can earn method marks. At the end, use any remaining time to review spelling of key terms and to check for missing ENDIF or NEXT statements.
如果被一道题卡住，做个标记并继续做下去。绝不要让题目空着：即使是不完整的追踪或部分伪代码也可能得到方法分。最后，利用剩余时间检查关键术语的拼写，查看是否有遗漏的 ENDIF 或 NEXT 语句。
12. How Marks Are Awarded (Mark Schemes) | 评分方式（评分方案）
Cambridge uses a points-based mark scheme. Each mark point corresponds to a specific piece of knowledge or a step in the solution. Some marks are 'A marks' (accuracy marks), which depend on obtaining a correct result only if the method is clear. 'M marks' (method marks) are awarded for a valid approach even if the final answer is wrong.
剑桥采用分点评分方案。每个得分点对应特定的知识内容或解答的一个步骤。有些是“A 分”（准确分），只有在方法清楚的情况下才能因正确的结果得到。而“M 分”（方法分）则奖励给有效的方法，即使最终答案错误也能获得。
'B marks' stand for independent marks that do not depend on previous results. For example, stating the correct definition of a firewall might be a B1 mark. In multi-part questions, you can still earn follow-through marks if you use an incorrect earlier result consistently.
“B 分”是独立分值，不依赖之前的结果。例如，准确陈述防火墙的定义可能是一个 B1 分。在分题小题中，如果你一致地使用了一个之前错误的结果，仍可能得到后续跟进分。
To maximise marks, always present your reasoning clearly. Examiners cannot award method marks for messy or illegible working. When a question says 'Show your working', it is because part of the marks are for the process, not just the answer. Treat every number on the answer line as an opportunity to demonstrate your logic.
为了获得最高分，始终清晰地展示你的推理过程。对于杂乱或字迹难以辨认的步骤，考官无法给方法分。当题目说“展示你的计算过程”时，意味着部分分数是给过程的，而不仅仅是答案。把答案纸上每一个数字都视为展示逻辑的机会。
Published by TutorHao | Computer Science Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图