📚 Year 12 WJEC Computer Science: Exam Techniques and Marking Criteria | WJEC 计算机科学 Year 12：答题技巧与评分标准
Mastering exam technique is as important as understanding the content for WJEC Year 12 Computer Science. The combination of Unit 1’s written paper and Unit 2’s on-screen programming test demands not only solid technical knowledge but also the ability to interpret questions, structure answers to meet mark scheme expectations, and manage time effectively. This guide breaks down the key strategies you need to convert your knowledge into marks across both units.
对于 WJEC 计算机科学 Year 12 来说，掌握答题技巧与理解学科内容同等重要。单元 1 的笔试与单元 2 的上机编程考试相结合，不仅要求扎实的技术知识，还要求你具备解读题意、按照评分标准构建答案以及有效管理时间的能力。本文将逐一剖析关键策略，帮助你在两个单元中将知识转化为分数。
1. Understanding the Exam Structure | 了解考试结构
Unit 1: Fundamentals of Computer Science is a 2-hour written paper worth 100 marks, accounting for 62.5% of the AS qualification. It covers computer architecture, communication, data representation, data structures, algorithms, principles of programming, software engineering, operating systems, databases, networks, and web technology. You will encounter a mix of multiple-choice, short-answer, and extended-writing questions. Unit 2: Practical Programming to Solve Problems is a 2-hour on-screen exam worth 60 marks (37.5% of AS). You will be assessed on your ability to design, code, test, and evaluate solutions in a chosen high-level language, using a pre-released scenario.
单元 1：计算机科学基础，笔试 2 小时，满分 100 分，占 AS 成绩的 62.5%。内容涵盖计算机体系结构、通信、数据表示、数据结构、算法、编程原理、软件工程、操作系统、数据库、网络和网页技术。题型包括选择题、简答题和扩展写作题。单元 2：通过编程解决实际问题，上机考试 2 小时，满分 60 分（占 AS 的 37.5%）。你将基于预先发布的题目场景，在所选择的高级语言中完成设计、编码、测试和评价，以展示你的实践能力。
2. Command Words and Their Meanings | 指令词及其含义
Describe: Give a detailed account of a concept, process, or feature, covering its main characteristics. Do not explain why; simply say what it is or what it does.
描述： 详细说明一个概念、过程或特征，涵盖其主要特性。无需解释原因，只需说明它是什么或它做什么。
Explain: Go beyond description by giving reasons or showing how something works. Often requires linking cause and effect or using an example to clarify the mechanism.
解释： 在描述的基础上给出原因或展示某事物如何运作。通常需要联系因果关系或举例阐明其机制。
Compare: Identify similarities and differences. A good comparison uses comparative language and may be structured around key characteristics. State both sides, not just one.
比较： 识别相似之处和不同之处。高质量的比较应使用对比性语言，并围绕关键特征展开。要说明两面，而非仅涉及其中一面。
Evaluate: Make an evidence-based judgement, considering both strengths and weaknesses. Use phrases like 'an advantage is... however, a drawback is...' and finish with a justified conclusion.
评价： 基于证据做出判断，兼顾优点与缺点。使用诸如“优点是……然而缺点是……”的表述，并以有据可依的结论收尾。
Justify: Give reasons for a choice or design decision. You must argue in favour of something, using relevant technical criteria such as efficiency, maintainability, or suitability for purpose.
论证： 说明某个选择或设计决策的理由。你必须从技术标准（如效率、可维护性或适用性）出发进行辩护。
Analyse: Break down a topic into its component parts and examine how they interrelate. Often used when examining algorithms, data structures, or system behaviour.
分析： 将主题分解为组成部分并检查其相互关系。常见于考察算法、数据结构或系统行为之时。
3. Mastering Short-Answer Questions | 掌握简答题
Short-answer questions in Unit 1 usually target recall of facts and straightforward understanding. Examiners expect precise technical language. Rather than writing 'a fast memory close to the CPU', write 'cache memory is a small, high-speed storage located on or near the CPU that stores frequently accessed instructions and data to reduce average memory access time'. Mark schemes often award marks for key terms: mention 'Harvard architecture', 'separate buses for data and instructions', and 'pipelining' when asked about processor improvements. Link each sentence to the question’s command word to stay focused.
单元 1 的简答题通常考查对事实的回忆和清晰的理解。考官期待考生使用准确的技术语言。不要写“靠近 CPU 的快速内存”，而应写“高速缓存（cache）是位于 CPU 内部或附近的小容量高速存储器，用于存放频繁访问的指令和数据，以降低平均存储访问时间”。评分方案经常给关键术语赋分：被问到处理器改进时，要提及“哈佛架构”、“指令与数据分离的总线”和“流水线”。使每句话都紧扣指令词，以便保持聚焦。
When a question asks for 'one disadvantage of a bus topology', a single well-expressed point with justification scores full marks. Avoid writing a list of vague ideas hoping one sticks; this wastes time and rarely earns credit. Instead, practice producing a clear disadvantage and then immediately explaining why it matters, e.g., 'If the backbone cable fails, the entire network goes down, causing a single point of failure.'
当题目要求写出“总线拓扑的一个缺点”时，一个表述清晰并有理有据的要点即可得满分。避免罗列模糊的想法以碰运气；这样做既浪费时间又很少得分。相反，练习先说一个明确的缺点，紧接着解释其重要性，如“若主干电缆故障，整个网络便会瘫痪，形成单点故障。”
4. Tackling Extended Response and Evaluation | 攻克扩展问答与评价题
A typical 'Evaluate the use of cloud storage' or 'Discuss the ethical impacts of AI' question can carry 6–10 marks. Structuring your response using the PEEL paragraph model (Point, Evidence, Explanation, Link) impresses examiners. Start with a clear point, support it with a specific technical example (e.g., 'cloud service provider data centres use redundant RAID arrays'), explain how this impacts users or businesses, and link back to the question. Aim for at least two balanced perspectives before a concluding judgement.
典型的“评价云存储的使用”或“讨论人工智能的伦理影响”这类题目可能占 6–10 分。使用 PEEL 段落模型（观点、证据、解释、联系）组织回答能赢得考官好感。先提出明确的观点，然后用一个具体的技术例子支持（如“云服务商的数据中心采用冗余 RAID 阵列”），接着解释这对用户或企业有何影响，最后回扣问题。在给出有据可依的结论之前，至少要从两个平衡的视角展开讨论。
Evaluation questions reward a genuine conclusion that weighs the arguments: 'Overall, while cloud storage offers scalability and reduced local hardware costs, its reliance on stable internet connectivity and long-term subscription fees may offset initial savings for some small businesses.' Avoid generic statements like 'it depends on the situation' without a final, reasoned stance.
评价题青睐真正的结论——对论点进行权衡：“总体而言，虽然云存储提供了可扩展性并降低了本地硬件成本，但它对稳定互联网连接的依赖以及长期订阅费用可能会抵消某些小企业最初的节省。”避免在没有最终合理立场的情况下说出“视情况而定”这样的泛泛之谈。
5. Algorithm Tracing and Dry Runs | 算法跟踪与手工执行
Questions that ask you to trace an algorithm require a methodical approach. Construct a trace table with columns for each variable and output. Update values step by step as you walk through the pseudocode. WJEC questions often include loops, 1D/2D arrays, and conditional statements. Begin by noting initial values, then iterate line by line. If the algorithm uses a WHILE loop, carefully check the condition at each iteration. Never skip steps or attempt to guess the final output without building the trace—marks are awarded for partial tracing and correct intermediate values.
要求跟踪算法的题目需要你采用条理清晰的方法。为每个变量和输出构建跟踪表（trace table）。逐行走查伪代码时，同步更新数值。WJEC 的题目经常包含循环、一维/二维数组和条件语句。从记录初始值开始，然后一行一行迭代。如果算法用了 WHILE 循环，务必在每次迭代时检查条件。切勿跳过步骤或不经构建跟踪表就猜测最终输出——评分会照顾到局部的正确中间值。
Dealing with nested loops or recursion can be tricky. Draw extra columns for inner loop counters and keep the trace clean. An integer variable storing a sum should be updated like sum ← sum + arr[index]. For an algorithm finding the maximum value in an array, your trace will commonly show comparisons and swaps of a max variable. Presenting a neatly labelled trace table makes it simple for an examiner to award all available method marks.
处理嵌套循环或递归时可能棘手。为内层循环计数器增绘额外列并保持跟踪表清晰。存储总和的整型变量应按 sum ← sum + arr[index] 的方式更新。对于寻找数组最大值的算法，你的跟踪表通常会展示 max 变量的比较与更新。呈现一个标注清晰的跟踪表，能让考官轻松判给你所有方法分。
6. Writing High-Quality Pseudocode | 编写高质量伪代码
WJEC does not enforce a strict pseudocode syntax, but your code must be unambiguous, properly indented, and apply consistent conventions. Use meaningful variable names like customerAge or totalCost, not single letters unless the context is obvious. Capitalize keywords (IF, WHILE, FOR, ENDIF, OUTPUT) to make structure clear. Show nesting through indentation, and ensure every IF has a matching ENDIF.
WJEC 不强求严格的伪代码语法，但你的代码必须无歧义、正确缩进并遵循一致的约定。使用有意义的变量名，如 customerAge 或 totalCost，而不要用单字母，除非上下文显而易见。将关键词大写（IF、WHILE、FOR、ENDIF、OUTPUT）以使结构清晰。通过缩进展示嵌套，并确保每个 IF 都有对应的 ENDIF。
Efficiency and clarity both matter. When designing an algorithm to search a list, a linear search might be coded as:
index ← 0
found ← false
WHILE index < length AND found = false
IF list[index] = target THEN
found ← true
ELSE
index ← index + 1
ENDIF
ENDWHILE
OUTPUT index
Adding brief comments to explain the logic can strengthen an answer, but do not over-narrate. Ensure your algorithm handles edge cases, such as an empty list or target not present. The mark scheme often rewards robust algorithm design with boundary-test consideration.
效率与清晰度都重要。在设计搜索列表的算法时，线性搜索可编码如下：
index ← 0
found ← false
WHILE index < length AND found = false
IF list[index] = target THEN
found ← true
ELSE
index ← index + 1
ENDIF
ENDWHILE
OUTPUT index
添加简短的注释来解释逻辑可以为答案增色，但不要过度叙述。确保算法处理边界情况，如空列表或目标不存在。评分方案通常会奖励考虑了边界测试的稳健算法设计。
7. Data Structures and Representation: Key Concepts in Mark Schemes | 数据结构与表示：评分方案中的核心概念
Questions on data structures require precise definitions and the ability to compare operations. For example, when asked to 'compare an array and a linked list', mention that arrays offer O(1) random access because elements are stored contiguously, but insertion and deletion may require shifting elements, resulting in O(n) worst case. Linked lists support O(1) insertion at the head but have O(n) search time. Using Big O notation with Unicode superscripts such as O(n²), O(log n), or O(1) can impress examiners, provided it is used correctly. Be ready to draw diagrams of stacks (LIFO), queues (FIFO), and binary trees for tracing questions.
有关数据结构的题目需要精准的定义和比较操作的能力。例如，当要求“比较数组与链表”时，要提到数组因其元素连续存储而支持 O(1) 的随机访问，但插入和删除可能需要移动元素，最坏情况为 O(n)。链表支持在头部 O(1) 的插入，但查找时间为 O(n)。使用带 Unicode 上标的大 O 符号，如 O(n²)、O(log n) 或 O(1)，若使用正确可以打动考官。做好准备绘制栈（LIFO）、队列（FIFO）和二叉树的示意图以应对追踪题。
lculation questions. Show your working clearly: when converting a negative number into two' combine. In bitmap image calculation, express size as width × height × colour depth in bits, then convert to bytes. Clear working ensures you receive method marks even if a final answer contains a minor arithmetic slip.
二进制、十六进制和浮点数等数据表示主题常出现在计算题中。清晰地展示你的计算过程：在将负数转换为补码时，写出中间二进制步骤。对于浮点数，分离出尾数和阶码，应用阶码偏移量，再组合。在位图图像计算中，用 宽度 × 高度 × 色深（比特）表示大小，再转换为字节。清晰的运算过程能确保即使最终答案出现小算术失误，你仍能获得方法分。
8. Unit 2 On-Screen Programming Exam: Practical Strategies | 单元 2 上机编程考试：实操策略
Unit 2 requires you to work from a pre-released scenario, often involving file input/output, data structures, and user interaction. Before the exam, thoroughly explore the scenario and practise implementing the core functionality in your chosen language. In the exam, allocate time for reading the task booklet, planning on paper, coding, and testing. A common mistake is to start coding immediately without a plan, leading to disorganised solutions that fail to meet all assessment criteria. Spend the first 5–10 minutes sketching a plan with procedure names, parameters, and intended logic.
单元 2 要求你基于预先发布的场景进行编程，通常涉及文件输入/输出、数据结构和用户交互。考前要充分钻研场景，并用所选语言练习实现核心功能。在考试中，合理分配时间用于阅读任务手册、在纸上规划、编写代码和测试。一个常见错误是不经规划就直接编码，导致解决方案杂乱不全，无法满足所有评分标准。花前 5-10 分钟勾勒规划，包括过程名、参数和预期逻辑。
Marking criteria for Unit 2 prioritise a working program, but also assess use of programming constructs, data validation, maintainability, and appropriate commenting. Ensure your code handles invalid input gracefully using conditional checks or exception handling. Use meaningful function decomposition to break the problem into smaller tasks. A program that compiles/runs and produces correct output for sample data forms the basis of passing marks; adding robust error handling and efficiency elevates the score into higher bands. Test with boundary values and unexpected input to demonstrate thoroughness.
单元 2 的评分标准优先考虑可运行的程序，同时也评估编程结构的使用、数据验证、可维护性和恰当的注释。确保你的代码能通过条件检查或异常处理优雅地应对无效输入。运用有意义的函数分解将问题拆分为更小的任务。能够编译/运行并针对样本数据产生正确输出的程序是及格的基础；添加健壮的错误处理和高效实现可将分数推入更高等级。通过边界值和意外输入进行测试以展示全面性。
9. Interpreting Mark Schemes and Band Descriptors | 理解评分方案与等级描述
WJEC mark schemes separate marks into different assessment objectives: AO1 (knowledge and understanding), AO2 (application), and AO3 (analysis and evaluation). For extended questions, band descriptors define what constitutes a high-, middle-, or low-level response. A top-band answer demonstrates 'detailed knowledge and understanding, sophisticated analysis, and a well-supported evaluation', while a low-band response 'may show limited understanding with few technical terms'. Familiarise yourself with these phrases and aim to show depth by connecting theoretical concepts to real-world examples.
WJEC 的评分方案将分数划分至不同的评估目标：AO1（知识与理解）、AO2（应用）和 AO3（分析与评价）。对于扩展型问题，等级描述界定了高、中、低水平答案的构成。顶级答案展示“详尽的知识与理解、精到的分析以及有充分支撑的评价”，而低水平回答“可能理解有限且术语使用较少”。熟悉这些措辞，并通过将理论概念与现实示例联系起来，力争展现深度。
You can download past mark schemes and examiner reports from the WJEC website. Use them to self-assess your practice answers. Notice how marks are distributed: a 6-mark 'explain' question may give 2 marks for identifying factors, 2 for developing explanations, and 2 for linking to a context. By aligning your answer structure with this pattern, you make it easy for the examiner to allocate each mark. Highlight key terms and judgments in your answer so they stand out.
你可以从 WJEC 网站下载历年评分方案和考官报告，用来对自己的练习答案进行自我评估。留意分数的分配方式：一道 6 分的“解释”题可能是 2 分给识别因素，2 分给展开解释，2 分给联系上下文。将你的答案结构与此模式对齐，便能让考官轻松分配每 1 分。在答案中突出关键术语和判断，使其更加醒目。
10. Avoiding Common Mistakes and Final Tips | 避免常见错误与最后建议
Many students lose marks by not reading the question fully or by ignoring the command word. If the question says 'using an example, explain...', you must embed a concrete example; otherwise, you cannot access full marks. Similarly, writing too much for a 2-mark question wastes time that could earn marks elsewhere. Tailor your answer length to the marks available—roughly a sentence and a supporting technical point per mark in short questions.
许多学生因未完整阅读题目或忽视指令词而失分。如果题目说“用一个例子解释……”，你就必须嵌入具体例子；否则无法获得全部分数。同样，为一道 2 分的题目洋洋洒洒写上一堆，只会浪费本可在别处得分的时间。根据可获得的分数调整答案长度——简答题中，大致对应 1 分写一句附带一个支撑技术要点。
In programming and algorithm questions, forgetting to initialise variables, using the wrong loop condition, or omitting end-of-file checks are frequent errors. Double-check your pseudocode by running a mental test or a quick trace. In theory questions, avoid vague language: replace 'it makes the system better' with 'it increases fault tolerance by duplicating critical components'. Finally, sleep well before the exam and have a time management plan: allocate minutes based on mark weighting. Keep an eye on the clock, and if stuck on a tough question, move on and return later. Consistent, disciplined technique will elevate your grade significantly.
在编程和算法题中，忘记初始化变量、使用错误的循环条件或遗漏文件结束检查是常见错误。通过在大脑中运行或快速追踪来复查你的伪代码。在理论题中，避免含糊的语言：将“它使系统更好”替换为“它通过复制关键组件提高了容错能力”。最后，考前保证充足睡眠，并制定时间管理计划：依据分值权重分配时间。留意时钟，若在某道难题上卡住，先跳过，稍后再回头处理。持之以恒、训练有素的答题技巧将显著提升你的等级。
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