📚 Year 13 WJEC Computer Science: Unit Test Mock Paper Analysis | Year 13 WJEC 计算机：单元测试模拟卷解析
Mock exams are a vital part of your revision toolkit. This in-depth walkthrough of a typical Year 13 WJEC Computer Science unit test will help you decode question styles, spot common pitfalls, and master the level of detail required for top marks. We will break down each topic area, providing model analysis and bilingual explanations that align closely with the WJEC specification.
模拟考试是复习工具箱中的重要利器。这份针对典型 Year 13 WJEC 计算机科学单元测试的深入解析，将帮助你拆解题型、发现常见陷阱，并掌握获得高分所需的答题深度。我们将逐一分解每个主题领域，提供贴合 WJEC 考纲的示范分析与中英双语讲解。
1. Algorithms and Data Structures | 算法与数据结构
A classic question asks: "Explain how a stack can be used to evaluate a Reverse Polish Notation (RPN) expression. Use the example 5 3 + 8 2 - *." The algorithm pushes operands onto a stack. When an operator is encountered, the top two operands are popped, the operation is performed, and the result is pushed back. For 5 3 +, 5 and 3 are pushed; '+' pops 3 and 5, pushes 8. Then 8 and 2 are pushed; '-' pops 2 and 8, pushes 6. Finally '*' pops 6 and 8, pushes 48.
经典题型：“解释如何使用栈计算逆波兰表达式（RPN）。以示例 5 3 + 8 2 - * 说明。” 算法将操作数压入栈。遇到运算符时，弹出栈顶两个操作数执行运算，再将结果压回栈中。对于 5 3 +，压入 5 和 3；‘+’ 弹出 3 和 5，压入 8。接着压入 8 和 2；‘-’ 弹出 2 和 8，压入 6。最后 ‘*’ 弹出 6 和 8，压入 48。
Time complexity questions often feature merge sort versus quick sort. Merge sort guarantees O(n log n) in all cases, while quick sort degrades to O(n²) if the pivot is poorly chosen. In a mock exam, you may be asked to draw the merge sort tree for a list of numbers, or explain why quick sort is often preferred in practice despite its worst-case scenario, due to smaller constant factors and in-place sorting.
时间复杂度题目常涉及归并排序与快速排序的对比。归并排序在任何情况下都保证 O(n log n)，而快速排序在基准值选取不佳时会退化为 O(n²)。模拟卷中可能要求你为数字列表画出归并排序树，或解释为何快速排序尽管有最坏情况，仍因常量因子较小且可原地排序而被广泛使用。
Binary search tree (BST) traversal is another core topic. Be prepared to trace in-order, pre-order, and post-order traversals on a given BST. A typical error is confusing the visit order; remember: in-order produces sorted output, pre-order gives the root first, and post-order is useful for deleting the tree.
二叉搜索树（BST）遍历是另一个核心主题。准备好对给定的 BST 追踪中序、前序和后序遍历。典型错误是混淆访问顺序；记住：中序产生排序输出，前序先输出根节点，后序常用于删除树。
2. Programming Techniques and Recursion | 编程技术与递归
ct code but also explain the base case and the recursive case. For example, in Python: def fib(n): if n <= 1: return n; else: return fib(n-1) + fib(n-2). A follow-up question will ask about the inefficiency of this approach due to overlapping subproblems, and you should propose memoisation or an iterative solution.
常见编程任务是编写计算第 n 个斐波那契数的递归函数。答案中不仅需要提供正确的代码，还要解释基准情形与递归情形。例如 Python 函数：def fib(n): if n <= 1: return n; else: return fib(n-1) + fib(n-2)。追问往往会涉及该方法因重叠子问题而效率低下，你需要提出记忆化或迭代解决方案。
File handling and exception management are frequently tested. You might be given a text file containing sensor readings and asked to write a program that reads the file line by line, converts strings to integers, and catches ValueError exceptions for corrupt data. Always use a try-except block and ensure the file is properly closed using a context manager (with open(...)).
文件处理与异常管理经常考查。题目可能给出一个包含传感器读数的文本文件，要求编写程序逐行读取文件、将字符串转换为整数，并捕获损坏数据引发的 ValueError 异常。务必使用 try-except 块，并利用上下文管理器（with open(...)）确保文件正确关闭。
An integrated question might combine recursion with data structures, such as writing a recursive function to count the number of leaf nodes in a binary tree. The algorithm must handle an empty tree, check if both left and right children are null, and recurse on subtrees.
综合题目可能将递归与数据结构结合，例如编写递归函数计算二叉树中的叶子节点数。算法需要处理空树，检查左右子节点是否均为空，并对子树进行递归调用。
3. Computer Architecture and Hardware | 计算机体系结构与硬件
WJEC mock papers heavily feature the Von Neumann architecture and its bottlenecks. A standard question asks you to describe the fetch-decode-execute cycle, explicitly mentioning registers such as the Program Counter (PC), Memory Address Register (MAR), Memory Data Register (MDR), and Current Instruction Register (CIR). Ensure you explain how the PC increments and how the MDR acts as a buffer.
WJEC 模拟卷大量考查冯·诺依曼架构及其瓶颈。标准题目要求描述取指-译码-执行周期，明确提到程序计数器（PC）、内存地址寄存器（MAR）、内存数据寄存器（MDR）和当前指令寄存器（CIR）。确保解释 PC 如何递增，以及 MDR 如何充当缓冲区。
RISC versus CISC comparisons are a must. Know that RISC processors have a small, simple instruction set, execute one instruction per clock cycle typically, and rely heavily on compiler optimisation. CISC processors have complex, multi-cycle instructions that can reduce code size but require more complex decoding hardware. You may be asked to compare power consumption, pipelining suitability, and use cases like embedded systems versus general-purpose computing.
RISC 与 CISC 的对比是必考内容。要知道 RISC 处理器指令集小而简单，通常每个时钟周期执行一条指令，并高度依赖编译器优化。CISC 处理器具有复杂的多周期指令，可减少代码长度，但需要更复杂的解码硬件。题目可能要求比较功耗、流水线适用性，以及嵌入式系统与通用计算的用例。
Pipelining hazards should be understood at a conceptual level. Data hazards, control hazards, and structural hazards can stall the pipeline. Mock questions often provide a short assembly code snippet and ask you to identify which hazard will occur and how it could be resolved, for instance by inserting NOPs or forwarding.
需要在概念层面理解流水线冒险。数据冒险、控制冒险和结构冒险可能导致流水线停顿。模拟题常会提供一小段汇编代码，要求你识别将发生的冒险类型，并提出解决方案，如插入空操作指令（NOP）或采用数据转发。
4. Networking and the Internet | 网络与互联网
A popular question presents a scenario where a web browser requests a page, and asks you to explain the journey through the TCP/IP protocol stack. Start from the application layer (HTTP), move down to transport (TCP segmentation and port numbers), then network (IP addressing and routing), down to link and physical layers. Addressing with MAC addresses and ARP should be included when moving between subnets.
热门题型会展示一个网页浏览器请求页面的场景，要求你解释通过 TCP/IP 协议栈的传输过程。从应用层（HTTP）开始，向下到传输层（TCP 分段和端口号），再到网络层（IP 寻址和路由），直至链路层和物理层。在跨子网传输时，需提及 MAC 地址和 ARP 协议。
Network security threats and countermeasures are a staple. You must distinguish between packet sniffing, man-in-the-middle attacks, and DDoS attacks. For a 6-mark question, you could discuss how packet sniffing can be mitigated by using encryption (HTTPS) and switches instead of hubs. DDoS mitigation might include traffic scrubbing and rate limiting.
网络安全威胁与对策是必考内容。你需要区分数据包嗅探、中间人攻击和分布式拒绝服务攻击（DDoS）。针对一道 6 分题，可以讨论通过使用加密（HTTPS）和交换机代替集线器来缓解数据包嗅探，DDoS 缓解措施可能包括流量清洗和速率限制。
The OSI model's seven layers and their functions are testable. Remember the mnemonic "Please Do Not Throw Sausage Pizza Away" for Physical, Data Link, Network, Transport, Session, Presentation, Application. Make sure you can map real-world protocols like Wi-Fi (802.11) to the Data Link layer, and DNS to the Application layer.
OSI 模型的七层及其功能也是考查点。记住助记口诀“Please Do Not Throw Sausage Pizza Away”对应从物理层到应用层。确保能将现实协议如 Wi-Fi（802.11）映射到数据链路层，将 DNS 映射到应用层。
5. Databases and SQL | 数据库与 SQL
SQL query writing is a guaranteed section. Given a relational schema with tables such as Student, Course, and Enrolment, you might be asked to "List all students who enrolled in 'Computer Science' after 2023, showing their names and course titles." The query will involve JOIN operations on foreign keys, a WHERE clause with a date filter, and perhaps an ORDER BY. Be careful with INNER JOIN versus LEFT JOIN semantics.
SQL 查询编写是必考环节。给定包含 Student、Course 和 Enrolment 等表的关系模式，可能会要求“列出 2023 年后注册‘计算机科学’课程的所有学生，显示姓名和课程名称”。查询将涉及基于外键的 JOIN 操作、带有日期过滤的 WHERE 子句，可能还有 ORDER BY。注意区分 INNER JOIN 与 LEFT JOIN 的语义。
Normalisation is frequently examined through a poorly designed table exhibiting anomalies. A typical task: "Normalize the following unnormalised table into 3NF." You must identify repeating groups to achieve 1NF, remove partial dependencies for 2NF, and eliminate transitive dependencies for 3NF. Show the resulting tables with primary keys underlined and foreign keys indicated.
规范化常通过一个存在异常的设计糟糕的表格来考查。典型任务：“将以下未规范化的表格规范化至 3NF。” 你必须识别重复组以达成 1NF，消除部分依赖达到 2NF，消除传递依赖达到 3NF。展示结果表，主键加下划线，外键标出。
Entity Relationship Diagrams (ERDs) are another visual question. You'll be given a business scenario and asked to draw an ERD showing entities, attributes, relationships, and cardinality. Use Crow's Foot notation as per WJEC preference. Don't forget to mark the primary key attribute for each entity.
实体关系图（ERD）是另一种可视化题型。你会得到一个业务场景，要求绘制 ERD 展示实体、属性、关系和基数。按照 WJEC 喜好使用 Crow's Foot 表示法。别忘了为每个实体标出主键属性。
6. Object-Oriented Programming | 面向对象编程
Inheritance and polymorphism are core OOP concepts. A mock question might provide a base class 'Vehicle' with methods 'move()' and 'honk()', and ask you to define a subclass 'Car' that overrides 'move()' and adds a 'fuelType' attribute. In your explanation, you must clarify how dynamic binding allows the correct overridden method to be called at runtime, showcasing polymorphism.
继承与多态是核心 OOP 概念。模拟题可能给出一个基类 ‘Vehicle’，包含方法 ‘move()’ 和 ‘honk()’，要求定义子类 ‘Car’，覆盖 ‘move()’ 并添加 ‘fuelType’ 属性。解释中必须阐明动态绑定如何在运行时调用正确的覆盖方法，体现多态性。
Encapsulation and abstraction are often compared. Encapsulation is the bundling of data with methods that operate on that data, and restricting direct access to an object's internal state (using private attributes and public getters/setters). Abstraction is about hiding complex implementation details and exposing only essential features. You could illustrate by writing a class with private attributes and public methods that manipulate them.
封装与抽象经常被比较。封装是将数据与操作数据的方法捆绑在一起，并限制对对象内部状态的直接访问（使用私有属性和公共 getter/setter）。抽象则是隐藏复杂实现细节，仅暴露必要功能。你可以通过编写一个包含私有属性和公共操作方法的类来加以说明。
Design patterns may appear in the higher-tier paper. The Singleton pattern ensures only one instance of a class exists, which is useful for a database connection pool. The Factory Method pattern defines an interface for creating an object but lets subclasses decide which class to instantiate. Be ready to sketch a simple class diagram for a given pattern.
设计模式可能出现在高级试卷中。单例模式确保一个类只有一个实例，适用于数据库连接池。工厂方法模式定义一个用于创建对象的接口，由子类决定实例化哪个类。准备好为给定模式绘制简单的类图。
7. Computational Thinking and Problem Solving | 计算思维与问题解决
Decomposition, pattern recognition, abstraction, and algorithm design form the four cornerstones. A question might present a complex task like sorting a list of customer orders by multiple criteria, and ask you to break it down step by step. Decompose by identifying sub-tasks: loading data, selecting comparison logic, implementing a stable sort, and output formatting. Recognise patterns like using lambda functions for custom sorting, and abstract away the data loading details into a function.
分解、模式识别、抽象和算法设计是四大基石。题目可能呈现一项复杂任务，如按多个标准对客户订单列表排序，并要求逐步分解。通过识别子任务来分解：加载数据、选择比较逻辑、实现稳定排序、输出格式化。识别模式，如使用 lambda 函数进行自定义排序，并将数据加载细节抽象为一个函数。
Heuristic methods versus optimal algorithms are discussed in terms of tractability. For the travelling salesman problem, an optimal solution is O(n!) making it intractable for large n. Hence heuristics like the nearest neighbour algorithm are preferred. Mock exams expect you to explain the trade-off between accuracy and time efficiency.
在可计算性方面会讨论启发式方法与最优算法。对于旅行商问题，最优解为 O(n!)，对于较大 n 难以处理，因此首选最近邻居算法等启发式方法。模拟考试期望你解释准确性与时间效率之间的权衡。
Finite State Machines (FSMs) are used to model sequential logic. Given a vending machine specification, draw the state transition diagram showing states such as Idle, CoinInserted, ItemSelected, and DispenseItem. Transitions should be labelled with the event that triggers them. Be precise with initial and accepting states.
有限状态机（FSM）用于对顺序逻辑建模。给定自动售货机规范，绘制状态转换图，展示 Idle、CoinInserted、ItemSelected 和 DispenseItem 等状态。转换应标注触发事件。准确标明初始状态和接受状态。
8. Legal, Moral, Ethical, and Cultural Issues | 法律、道德、伦理与文化议题
The Data Protection Act 2018 (incorporating GDPR) is a recurring theme. A scenario might describe a company storing customer health data. You need to explain the eight principles, focusing on lawful and transparent processing, purpose limitation, and data minimisation. A high-mark answer critically evaluates the impact of a data breach and discusses the role of the Information Commissioner's Officer (ICO).
2018 年《数据保护法》（纳入 GDPR）是反复出现的主题。题目场景可能描述一家公司存储客户健康数据。你需要解释八大原则，重点关注合法、透明处理，目的限制和数据最小化。高分答案会批判性地评价数据泄露的影响，并讨论信息专员办公室（ICO）的作用。
The Computer Misuse Act 1990 covers unauthorised access, unauthorised access with intent to commit further offences, and unauthorised acts with intent to impair computer operation. Distinguish clearly between these three sections. A moral dimension question could involve an employee who accesses a colleague's emails without permission out of curiosity; identify which offence this is and suggest the potential professional consequences.
1990 年《计算机滥用法》涵盖未经授权的访问、意图实施进一步犯罪的未经授权访问，以及意图损害计算机操作的未授权行为。清晰区分这三个条款。涉及道德层面的问题可能包括员工出于好奇未经允许访问同事电子邮件；辨认这属于哪一类犯罪，并说明潜在的职业后果。
Ethical dilemmas around AI and automation often appear. You must discuss accountability when a fully autonomous vehicle causes an accident. Who is responsible — the developer, the manufacturer, or the passenger? Frame your answer around transparency, algorithmic bias, and the societal need for regulation. Link to broader cultural implications such as job displacement.
围绕人工智能与自动化的伦理困境常见。必须讨论当全自动驾驶汽车引发事故时，责任归谁——开发者、制造商还是乘客？围绕透明度、算法偏见和社会监管需求来组织答案。联系就业替代等更广泛的文化影响。
9. Big Data and Distributed Systems | 大数据与分布式系统
Big Data is defined by the 3Vs: Volume, Velocity, and Variety. A mock question often provides a dataset like live social media feeds and asks you to explain why traditional relational databases struggle with it. Highlight the semi-structured and unstructured data varieties, the need for distributed storage like Hadoop HDFS, and the MapReduce programming model for parallel processing.
大数据由 3V 定义：规模（Volume）、速度（Velocity）和多样性（Variety）。模拟题常提供类似实时社交媒体信息流的数据集，要求解释传统关系型数据库为何难以处理。强调半结构化和非结构化数据的多样性，对 Hadoop HDFS 等分布式存储的需求，以及用于并行处理的 MapReduce 编程模型。
Data warehousing and data mining concepts are tested. A data warehouse is a central repository integrating data from multiple sources, optimised for queries and analysis (OLAP) rather than transaction processing (OLTP). Data mining involves pattern discovery using techniques such as clustering, classification, and association rule mining. Be able to link these to business intelligence and how insights can drive decision-making.
数据仓库与数据挖掘概念是考点。数据仓库是集成多源数据的中央存储库，针对查询和分析（OLAP）而非事务处理（OLTP）进行优化。数据挖掘涉及使用聚类、分类和关联规则挖掘等技术进行模式发现。要能将其与商业智能联系起来，并说明洞见如何推动决策。
10. System Software and Operating Systems | 系统软件与操作系统
The role of an operating system in resource management is a classic: memory management, process scheduling, file management, and input/output control. You need to compare paging with segmentation for memory management. Paging divides memory into fixed-size pages, avoiding external fragmentation but suffering from internal fragmentation. Segmentation uses variable-sized segments that match logical units, but can cause external fragmentation.
操作系统在资源管理中的作用是经典考点：内存管理、进程调度、文件管理和输入/输出控制。需要比较分页与分段内存管理。分页将内存划分为固定大小的页面，避免了外部碎片，但存在内部碎片。分段使用与逻辑单元匹配的可变大小段，但可能引起外部碎片。
Interpreters versus compilers: know the differences thoroughly. An interpreter translates and executes high-level code line by line, making it easier for debugging but slower in execution. A compiler translates the entire source code into machine code before execution, yielding faster runtime performance but a longer initial compile time. Mock questions may ask you to justify the choice in a safety-critical embedded system versus a web development environment.
解释器与编译器：要彻底掌握区别。解释器逐行翻译并执行高级代码，便于调试但执行较慢。编译器在执行前将整个源代码转变为机器码，运行时性能更快但初始编译时间更长。模拟题可能要求你说明在安全关键的嵌入式系统与网页开发环境中，应如何合理选择。
11. Software Development Methodologies | 软件开发方法论
Waterfall and Agile are contrasted head-to-head. Waterfall is linear and sequential with distinct phases (requirements, design, implementation, verification, maintenance). Agile is iterative and incremental, with customer collaboration and continuous feedback. For a 10-mark essay, you must discuss a project's suitability, for instance, choosing Waterfall for a well-defined government system with fixed requirements, and Agile for a startup app that needs to adapt quickly.
瀑布模型与敏捷开发需要正面对比。瀑布模型是线性、顺序的，具有明显区分阶段（需求、设计、实现、验证、维护）。敏捷开发是迭代和递增的，强调客户协作和持续反馈。对于 10 分论述题，需要讨论项目适用性，例如，需求固定的成熟政府系统适合瀑布模型，而需要快速适应的初创应用适合敏捷开发。
Testing strategies in each methodology are important. Waterfall typically delays most testing until the verification phase, increasing the cost of fixing late-discovered errors. Agile advocates for test-driven development (TDD), where tests are written before the code, and continuous integration ensures regular builds and instant feedback. The mock might ask you to write a simple unit test for a function that validates an email address.
每种方法论中的测试策略也很重要。瀑布模型通常将大部分测试延迟到验证阶段，增加了修复后期发现错误的成本。敏捷开发提倡测试驱动开发（TDD），即先写测试再写代码，并通过持续集成保证定期构建和即时反馈。模拟卷可能要求你为验证电子邮件地址的函數编写简单的单元测试。
12. Mock Exam Strategy and Common Pitfalls | 模拟考试策略与常见陷阱
Time management is critical. Allocate roughly one minute per mark. For a 4-mark definition question, spend no more than 4 minutes outlining key terms with a short example. Save the lengthy 10-mark discussion questions for last, ensuring you have a solid structure: introduction, arguments, counter-arguments, conclusion. Always leave 5 minutes for review at the end.
时间管理至关重要。大致按每分钟一分的节奏分配时间。对于 4 分的定义题，花不超过 4 分钟概述关键术语并配简短示例。将冗长的 10 分讨论题留到最后，确保结构稳固：引言、论点、反论点、结论。最后务必留出 5 分钟复查。
Carefully reading command words is non-negotiable. 'State' requires a brief answer; 'Describe' needs detailed characteristics; 'Explain' requires reasons and effects; 'Discuss' demands a balanced review. A common pitfall is writing everything you know about a topic without focusing on the command word, which will not earn full marks.
仔细审读指令词至关重要。‘State’ 要求简短回答；‘Describe’ 需要详细特征；‘Explain’ 需要原因和影响；‘Discuss’ 要求平衡评述。常见陷阱是不顾指令词而把某个主题的所有知识都写上去，如此无法获得满分。
Finally, always provide concrete WJEC-compliant examples. If asked about data validation, don't just say "range check" — write "a range check on a date of birth field ensures the year is between 1900 and 2024." This specificity demonstrates application and can lift your answer into the highest mark band.
最后，始终提供符合 WJEC 要求的具体示例。若被问及数据验证，不要只说“范围检查”——要写出“对出生日期字段进行范围检查，确保年份在 1900 至 2024 之间”。这种具体性体现了应用能力，可将答案推至最高分段。
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