📚 IGCSE OCR Computer Science: Top Techniques for Full Marks | IGCSE OCR 计算机科学：满分答题技巧
Securing top marks in IGCSE OCR Computer Science requires far more than just knowing the content. It demands a clear understanding of what examiners look for in every answer, how to deploy subject-specific keywords, and how to present your working and logic in a way that leaves no mark unclaimed. This guide walks you through the essential answering techniques that consistently help students move from a grade 6 or 7 to a grade 9.
在IGCSE OCR计算机科学中拿到满分，远不止是知道内容那么简单。你需要清楚地理解考官在每道题中寻找什么，如何运用学科关键词，以及如何展示你的推导过程和逻辑，从而不让任何一个分数溜走。本指南将为你梳理那些能稳定帮助学生从6-7分跨越到9分的核心答题技巧。
1. Understanding the Mark Scheme | 理解评分标准
Every question on an OCR Computer Science paper is designed around a specific set of mark points. The wording of the question tells you exactly how many distinct points to make; for example, a 'State two reasons...' question is worth 2 marks and needs two separate, simple statements. An 'Explain why...' question worth 4 marks expects a step-by-step reasoning chain, often with a correct conclusion.
OCR计算机科学试卷上的每一道题都围绕着特定的得分点设计。题目措辞会明确告诉你需要给出多少个独立的观点；例如，'State two reasons...'（给出两个理由）分值为2分，需要两个单独、简单的陈述。而一道价值4分的'Explain why...'（解释为什么）题目，则期待你给出逐步展开的推理链，通常还需要给出正确的结论。
For 'state' or 'identify' questions, keep answers brief and factual; do not add extra explanation unless asked.
对于'state'或'identify'类型的题目，保持答案简短且基于事实；除非题目要求，否则不要额外解释。
For 'describe' questions, focus on what happens or what something does; you may need to mention components or stages.
对于'describe'类型的题目，聚焦于发生了什么或某物做什么；你可能需要提到组件或阶段。
For 'explain' questions, use linking words like 'because', 'therefore', 'this leads to' to build a logical sequence.
对于'explain'类型的题目，使用诸如'because'、'therefore'、'this leads to'等连接词，构建一个逻辑序列。
For 'compare' questions, always address both similarities and differences, and use comparative language such as 'whereas' or 'however'.
对于'compare'类型的题目，始终要同时处理相同点和不同点，并使用'whereas'或'however'等比较性语言。
2. Using Precise Keywords | 使用精确关键词
OCR Computer Science mark schemes rely heavily on specific technical vocabulary. Writing about a 'quick memory inside the processor' will not earn the mark reserved for 'cache memory'. Similarly, referring to 'the part of the CPU that does calculations' should be 'ALU'. Train yourself to use the exact term every time.
OCR计算机科学的评分方案严重依赖特定的技术词汇。写出'处理器内部的快速存储器'不会拿到为'高速缓存（cache memory）'预留的分数。同样地，提及'CPU中进行计算的部分'，应该表述为'ALU（算术逻辑单元）'。训练自己每次都使用精确术语。
Replace 'the software that manages hardware' with 'operating system'.
用'operating system（操作系统）'替代'the software that manages hardware（管理硬件的软件）'。
Use 'volatile' and 'non-volatile' when discussing RAM and ROM, rather than 'loses data when power is off'.
讨论RAM和ROM时使用'volatile（易失的）'和'non-volatile（非易失的）'，而不是'turns off时丢失数据'。
In networking, distinguish between 'packet switching' and 'circuit switching' by name, not by vague descriptions of journey.
在网络中，要用名称区分'packet switching（分组交换）'和'circuit switching（电路交换）'，而非模糊地描述传输过程。
For programming, stick to OCR pseudocode keywords like
INPUT,OUTPUT,IF...THEN...ELSE...ENDIF, andFOR...NEXTexactly as specified in the syllabus.对于编程，必须严格遵循考纲指定的OCR伪代码关键字，例如
INPUT、OUTPUT、IF...THEN...ELSE...ENDIF和FOR...NEXT。
3. Structuring Long-Answer Responses | 结构化长答案作答
Six-mark and eight-mark questions in OCR papers often require extended writing. Examiners award marks for clarity, logical progression, and the presence of both technical detail and a valid conclusion. A well-structured answer uses short paragraphs or bullet points, each delivering one clear idea, and finishes with a concluding sentence that ties back to the question.
OCR试卷中6分和8分的题目通常需要展开论述。考官根据清晰度、逻辑递进以及技术细节和有效结论的存在与否来给分。一份结构良好的答案会使用短小段落或项目符号，每个段落传递一个清晰的思路，并以一句回扣题目的总结句收尾。
Begin by restating the question in your own words as a topic sentence, showing the examiner you understand the focus.
开始时用自己的话重述问题作为主题句，向考官表明你理解了聚焦点。
Present each technical point in a separate sentence; do not bury two distinct marks in one long, rambling statement.
每个技术点用独立的句子呈现；不要将两个不同的得分点埋没在一个冗长松散的陈述中。
Use connective phrases such as 'One consequence of this is...', 'In contrast...', 'Furthermore...' to link ideas smoothly.
使用诸如'One consequence of this is...'、'In contrast...'、'Furthermore...'等连接短语来顺畅衔接观点。
Always end with a short evaluation that summarises the overall impact or answers the 'why' part of the question.
始终以一段简短的评估收尾，总结整体影响或回答题目中的'why'部分。
4. Showing Working in Calculation Questions | 在计算题中展示步骤
OCR calculation questions – on file sizes, binary shifts, logic gate outputs, or compression ratios – often award method marks even if the final answer is wrong. Never just write the answer on its own. Write down every intermediate step, label each value, and keep units beside numbers so the examiner can follow your thinking.
OCR的计算题——涉及文件大小、二进制移位、逻辑门输出或压缩比——即便最终答案错误，也常常会给出方法分。绝不要只写出答案本身。写下每一个中间步骤，为每个数值加标签，并在数字旁保留单位，以便考官能跟上你的思路。
Example: Sound file size = sample rate × bit depth × duration (seconds)
示例：音频文件大小 = 采样率 × 位深度 × 时长（秒）
rt all units to match (e.g., sample rate 44.1 kHz = 44100 Hz, duration 3 minutes = 180 s).
步骤1：统一所有单位（例如，采样率44.1 kHz = 44100 Hz，时长3分钟 = 180秒）。
Step 2: Write the formula and substitute: 44100 × 16 × 180 = ... bits.
步骤2：写出公式并代入：44100 × 16 × 180 = ... 位。
Step 3: Convert to bytes, kilobytes, or megabytes as the question requests, dividing by 8, then by 1024.
步骤3：按题目要求转换为字节、千字节或兆字节，除以8，再除以1024。
binary multiples; OCR accepts 1 kB = 1000 bytes or 1 KiB = 1024 bytes, but be consistent.
始终标明你用的是十进制还是二进制倍数；OCR接受1 kB = 1000字节或1 KiB = 1024字节，但必须保持一致。
5. Algorithm Tracing and Dry-Running | 算法追踪与桌面检查
Many Paper 2 questions ask you to trace an algorithm with given test data. To guarantee full marks, produce a neat trace table with columns for each variable and any output. Update the table row by row, exactly as the computer would, and do not skip steps – even if the logic seems obvious, the marks are in the trace.
许多Paper 2题目要求你用给定的测试数据追踪一个算法。要保证满分，须制作一个整洁的追踪表，为每个变量和任何输出设立列。逐行更新表格，完全如计算机那样执行，并且不要跳步——即使逻辑看起来显而易见，分数就在于追踪过程。
Create a table with column headings matching the variables in the algorithm: e.g.,
count,total,output.创建一个表格，列标题与算法中的变量对应：例如
count、total、output。Write the initial values in the first row before any loop starts.
在第一行写任何循环开始前的初始值。
After each iteration of a loop, insert a new row with updated values; if a condition changes the flow, mark which branch was taken.
每次循环迭代之后，插入一行更新后的值；如果某个条件改变了流程，标注进入了哪个分支。
If the algorithm produces output, include an extra column and write each value as it would appear on screen, in quotation marks if it is a string.
如果算法产生输出，加入额外一列，并像屏幕上显示的那样写出每个值，若为字符串则加上引号。
For algorithms that ask 'What is the purpose of this algorithm?', state exactly what the output represents in context, e.g., 'It calculates the sum of all even numbers in the array.'
对于询问'该算法的目的是什么？'的题目，要准确地说明输出在上下文中代表什么，比如'它计算数组中所有偶数的和。'
6. Writing Pseudocode for Full Marks | 伪代码满分书写
When a question asks you to write a pseudocode solution, the examiner checks for logical correctness, adherence to OCR syntax, and efficiency. Full marks go to algorithms that handle boundary conditions, use consistent indentation, and include comments where helpful. Do not mix real programming languages with pseudocode; stick to the OCR guide.
当题目要求编写伪代码解决方案时，考官会检查逻辑正确性、是否遵循OCR语法以及效率。能够处理边界条件、使用一致缩进并在有用处包含注释的算法，才能拿到满分。不要将真实编程语言与伪代码混用；严格依循OCR指南。
Always declare variables with a statement like
count ← 0before using them.使用变量前始终用类似
count ← 0的语句声明。Use
INPUTandOUTPUTfor user interaction, andFOR i ← 1 TO 10for fixed loops; for condition-based loops useWHILE condition DO ... ENDWHILE.用
INPUT和OUTPUT实现用户交互，用FOR i ← 1 TO 10表示固定循环；对于基于条件的循环，使用WHILE condition DO ... ENDWHILE。Include an
IF temp > 30 THEN ... ELSE ... ENDIFstructure, notELSE IF(which is not OCR syntax).包含
IF temp > 30 THEN ... ELSE ... ENDIF结构，而不是ELSE IF（这并非OCR语法）。After writing your pseudocode, dry-run it with both normal, edge, and invalid data to verify correctness.
写完伪代码后，分别用正常值、边界值和无效值进行桌面检查以验证正确性。
7. Binary, Hex and Data Representation | 二进制、十六进制与数据表示
Questions on number conversion, binary addition, and character encoding are straightforward marks if you present your working methodically. For conversions, show division-by-two remainders for denary to binary, or grouping of binary bits into nibbles for binary to hex. For binary addition, align the columns, carry bits forward, and check for overflow.
关于数制转换、二进制加法和字符编码的题目，如果你能有条不紊地展示演算过程，就是极易得分的项目。在十进制转二进制时，展示除二取余的过程；二进制转十六进制时，展示将二进制位分组为半字节。对于二进制加法，对齐数位，进位，并检查溢出。
Denary to binary: example for 45 → 45 ÷ 2 = 22 rem 1, 22 ÷ 2 = 11 rem 0, … → read remainders backwards: 1011012.
十进制转二进制：以45为例 → 45 ÷ 2 = 22 余1, 22 ÷ 2 = 11 余0, … → 反向读取余数：1011012。
Binary to hex: group 1011 01012 into nibbles → B 5 16. Always specify the base clearly.
二进制转十六进制：将1011 01012分组为半字节 → B 5 16。始终清晰注明基数。
For two's complement, remember to flip the bits and add one when dealing with negative numbers.
对于二进制补码，处理负数时记住将各位取反后加一。
When asked about ASCII vs Unicode, mention the number of bits (7/8 vs up to 32) and the range of characters each can represent, using keywords like 'internationalisation'.
当被问及ASCII与Unicode时，要提到位数（7/8位与最多32位）以及各自能表示的字符范围，并运用'国际化（internationalisation）'等关键词。
8. Hardware & Software Explanation Questions | 硬件与软件解释题
Explanation questions on hardware components often require you to connect a part's function to overall system performance. Describing that 'the CPU has a clock speed of 3 GHz' is not enough; you must explain that a higher clock speed means more fetch-decode-execute cycles per second, leading to faster program execution.
关于硬件组件的解释题，常常要求你将某个部件的功能与整个系统的性能联系起来。仅仅描述'CPU的时钟速度为3 GHz'是不够的；你必须解释更高的时钟速度意味着每秒有更多的取指-译码-执行周期，从而使得程序执行更快。
When describing the fetch-decode-execute cycle, name the registers involved: PC (program counter), MAR, MDR, CIR, and accumulator.
描述取指-译码-执行周期时，要说出涉及到的寄存器：PC（程序计数器）、MAR、MDR、CIR和累加器。
For virtual memory questions, explain that the OS moves inactive data from RAM to the hard disk when RAM is full, and that excessive use causes disk thrashing, slowing the system.
对于虚拟内存的题目，要解释当RAM已满时操作系统将不活跃数据从RAM移至硬盘，而过度使用会导致磁盘颠簸（thrashing），减慢系统速度。
In software topics, distinguish between application software and system software, and give clear examples of each: a word processor vs an operating system or utility software.
在软件主题中，区分应用软件与系统软件，并为每一类提供清晰的例子：文字处理软件与操作系统或实用工具软件。
When discussing the role of a compiler vs an interpreter, mention that a compiler translates the entire source code into machine code before execution, whereas an interpreter translates and executes line by line; link this to speed and error detection.
讨论编译器与解释器的作用时，要提到编译器在执行前将整个源代码翻译为机器码，而解释器则逐行翻译并执行；将此与速度和错误检测联系起来。
9. Avoiding Common Pitfalls | 避免常见陷阱
Even well-prepared students lose marks through simple exam-technique errors. The most frequent include not reading the command word, giving a description when an explanation is needed, and confusing units in data transfer questions. Being aware of these traps can save 5-10 marks across a paper.
即使是准备充分的学生，也会因简单的考试技巧错误而丢分。最常见的问题包括未认真阅读指令词，在需要解释时给出了描述，以及在数据传输题中混淆单位。意识到这些陷阱，一整张试卷可以挽救5-10分。
Trap 1: Writing a list of points when the question asks for an evaluation. An evaluation requires a judgement supported by reasons.
陷阱1：题目要求作评估时写了一系列要点。评估需要得出一个由理由支持的判断。
Trap 2: Using a brand name like 'Windows' instead of the general term 'operating system' unless the question specifically asks for an example.
陷阱2：使用'Windows'等品牌名而非通用术语'operating system'，除非题目明确要求举例。
Trap 3: Forgetting to label axes or using abstract variable names in trace tables. Examiners need to see precisely what x and y represent.
陷阱3：在追踪表中忘记标定轴名称或使用了抽象的变量名。考官需要明确看到x和y究竟代表什么。
Trap 4: Neglecting the ethical or legal aspects in 'discuss' questions about technology. Mention the Data Protection Act or Computer Misuse Act where relevant.
陷阱4：在关于技术的'discuss'题目中忽略了伦理或法律层面。在相关时提及《数据保护法》或《计算机滥用法》。
10. Time Management in the Exam | 考试中的时间管理
OCR Paper 1 and Paper 2 each last 1 hour 30 minutes, but the question styles are different. Paper 1 mixes short-answer and a long 6/8-mark question; Paper 2 has a strong algorithmic focus with a programming task. A sensible per-mark time allocation is about 1.2 minutes per mark, leaving 5-10 minutes for checking.
OCR的Paper 1和Paper 2各持续1小时30分钟，但题目风格不同。Paper 1混合了简答题和一道6/8分的长题目；Paper 2则侧重算法，并包含一个编程任务。合理的时间分配大约是每1分给1.2分钟，留出5-10分钟用于检查。
Complete the short 1-2 mark questions first, because they build confidence and secure easy marks quickly.
先完成1-2分的短题，因为它们能帮你建立信心并快速锁定容易拿到的分数。
For algorithm questions, spend no more than 15 minutes on writing pseudocode; if stuck, outline the logic and move on, returning later.
对于算法题，花在编写伪代码上的时间不超过15分钟；如果遇到困难，先写出逻辑大纲然后继续往下做，稍后再回来完善。
Use the 'mark number' as a guide: a 4-mark question should take about 5 minutes; do not write a full page when two well-structured sentences are enough.
以'题分数'为指引：一道4分的题目应花费约5分钟；当两个结构良好的句子就足够时，不要写上满满一页。
Reserve the final 5 minutes to re-read all short answers for missing keywords and check that any trace tables are complete.
保留最后5分钟重读所有简答题，查找缺失的关键词，并检查追踪表是否填完整。
11. Revision Strategies That Lead to a Grade 9 | 通往9分的复习策略
A grade 9 in IGCSE OCR Computer Science is built on active, exam-focused revision, not passive reading. The most successful students practise answering past paper questions under timed conditions, self-mark using the official mark scheme, and maintain a 'keyword log' of technical terms they have missed. They also teach tricky concepts to someone else, which forces clarity of thought.
IGCSE OCR计算机科学的9分是建立在主动的、以考试为中心的复习之上，而非被动阅读。最成功的学生会模拟限时条件练习历年考题，用官方评分方案自行批改，并为自己答错的技术术语维护一份'关键词日志'。他们还会把难懂的概念教给别人，从而迫使自己把思路理清。
Create flashcards for all key definitions from the syllabus: one side with the term, the other with a concise, exam-ready definition and an example.
为考纲中所有关键定义制作闪卡：一面是术语，另一面是简洁、可用于考试的释义和一个例子。
Practice converting between denary, binary and hex until you can do it mentally; this will save 5-10 minutes in the exam.
反复练习十进制、二进制和十六进制之间的转换，直到能心算完成；这会在考试中节省5-10分钟。
For the programming paper, write at least three pseudocode solutions every week, focusing on OCR-style scenarios such as validation, counting, and linear search.
对于编程卷，每周至少编写三个伪代码解决方案，聚焦于OCR风格的场景，如验证、计数和线性搜索。
Collect a bank of 'explain why' model answers from past papers and analyse the structure: point, technical reason, consequence.
从历年试卷中收集'explain why'类型的标准答案库，并分析其结构：观点、技术原因、后果。
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