📚 Pre-U WJEC Computer Science: Exam Techniques and Marking Criteria | WJEC 预科计算机科学：答题技巧与评分标准
Succeeding in WJEC Pre-U Computer Science requires more than just knowing the syllabus — you must understand how marks are awarded and how to structure your answers effectively. This guide breaks down key exam techniques and the marking criteria for each question type, helping you maximise your score.
在 WJEC 预科计算机科学考试中取得好成绩，不仅需要掌握课程内容，还必须理解评分标准并组织好答案。本指南分解了关键答题技巧和各类题型的评分细则，帮助你最大化得分。
1. Understanding the Marking Scheme | 理解评分标准
WJEC uses a combination of point-based marking for short-answer questions and levels-of-response marking for extended writing. For point-based marks, each correct fact or step earns 1 mark, so answer concisely and avoid unnecessary information.
WJEC 对简答题采用按点给分，对论述题采用层级评分法。在按点给分中，每个正确的事实或步骤得 1 分，所以请简洁作答，避免无关内容。
In levels-based marking, examiners look for overall quality, structure, and depth. This means your extended answers must show a logical sequence, accurate use of terminology, and balanced discussion where required. Always read the indicative content provided in past mark schemes to understand what gains top-band marks.
在层级评分中，考官看重整体质量、结构和深度。这意味着你的论述题必须展示逻辑顺序、术语的准确使用，并在需要时进行平衡的讨论。务必阅读往年评分标准中的指示性内容，了解如何获得最高档分数。
For programming and algorithm questions, marks are often split between correct logic, syntax, and handling edge cases. Even if your code doesn’t compile perfectly on paper, clearly structured pseudocode can still secure high marks if the examiner can follow your reasoning.
对于编程和算法题，分数通常分布在逻辑正确性、语法以及边界情况处理上。即使书写的代码在纸上不能完美编译，结构清晰的伪代码同样能让考官看懂你的思路，从而获得高分。
2. Decoding Command Words | 解读指令词
Command words tell you exactly what the examiner expects. Misinterpreting them is one of the most common reasons for lost marks. The table below lists the most frequently used command words in WJEC Pre-U Computer Science papers.
指令词准确告诉你考官的要求。误解指令词是最常见的丢分原因之一。下表列出了 WJEC 预科计算机科学试卷中最常用的指令词。
| Command Word | Meaning | 中文含义 |
|---|---|---|
| Define | Give a precise meaning. | 给出精确定义。 |
| Describe | State the characteristics or main features. | 说明特征或主要特点。 |
| Explain | Give reasons or mechanisms, often requiring ‘how’ or ‘why’. | 解释原因或机理，常需回答”如何”或”为何”。 |
| Compare | Identify similarities and differences. | 指出相似点与不同点。 |
| Evaluate | Weigh up strengths and weaknesses to form a judgement. | 权衡优缺点并作出判断。 |
| State | Give a fact or a short answer without explanation. | 给出事实或简短答案，无需解释。 |
When you see ‘Explain’, always link your answer to a ‘because’ or a cause-and-effect chain. An ‘Evaluate’ question expects you to write a conclusion acknowledging both sides of the argument. Practising with past papers will help you internalise these expectations.
看到”Explain”时，一定要用”因为”或因果链条来连接答案。”Evaluate”题则要求你写一个结论，并承认正反两面观点。多做历年真题有助于内化这些要求。
3. Algorithm and Pseudocode Mastery | 算法与伪代码表达技巧
WJEC expects you to write clear, unambiguous pseudocode or flowcharts. Always use consistent indentation to show control structures. Start with a variable declaration section and then outline steps sequentially.
WJEC 要求你书写清晰、无歧义的伪代码或流程图。务必使用一致的缩进来展示控制结构。先声明变量，然后按顺序列出步骤。
For loops, use keywords like FOR ... ENDFOR, and for decisions, IF ... THEN ... ELSE ... ENDIF. A common mistake is missing the end-of-block markers, which can make the logic ambiguous and cost marks.
对于循环，使用 FOR ... ENDFOR；对于判断，使用 IF ... THEN ... ELSE ... ENDIF。一个常见错误是遗漏块结束标记，这会让逻辑变得含糊不清从而丢分。
When an algorithm requires user input, write INPUT with a clear prompt. For output, use OUTPUT or PRINT. In sorting or searching algorithms, explicitly state the steps and include comments. WJEC marks comments in pseudocode if they clarify intent, but don’t overdo it — crisp comments are best.
当算法需要用户输入时，使用 INPUT 并附上清晰提示；输出时使用 OUTPUT 或 PRINT。在排序或搜索算法中，明确写出每一步并添加注释。WJEC 评分时若注释能阐明意图也会给分，但不要过度——简洁的注释为佳。
4. Programming Problem Solving | 编程题解答策略
Before writing any code, analyse the problem: identify inputs, processes, and outputs. Define test data, including normal, boundary, and erroneous values. This approach not only organises your thoughts but also shows the examiner your systematic thinking.
在编写任何代码之前，先分析问题：明确输入、处理过程和输出。定义测试数据，包括正常值、边界值和错误值。这种方法既能理顺思路，又能向考官展示你的系统性思维。
In programming questions, the mark scheme often allocates marks for: correct syntax, use of appropriate data structures, modular design, and handling of edge cases. For example, if a question asks for a file reading program, ensure you include error handling for file-not-found conditions.
编程题的评分标准通常分配分值给：语法正确性、适当的数据结构使用、模块化设计以及边界情况处理。例如，如果题目要求编写文件读取程序，确保包含文件找不到的错误处理。
Even if you cannot complete the entire code, partial marks are awarded for correct fragments. Therefore, write as much as you can logically — a half-completed WHILE loop with correct condition and structure still earns marks.
即使无法写完整个程序，部分正确的代码片段仍可得分。因此，尽量多写逻辑合理的内容——一个条件正确且结构清晰的半截 WHILE 循环仍然可得部分分数。
5. Data Representation Accuracy | 数据表示与运算的准确性
When converting between binary, denary, and hexadecimal, always show your working. The mark scheme rewards method marks even if the final answer is incorrect due to a slip. For example, binary subtraction using two’s complement should be laid out step by step.
在二进制、十进制和十六进制之间转换时，务必展示计算过程。评分标准既给方法分，即使最终答案因笔误而错。例如，用补码进行二进制减法时，应逐步列明过程。
Floating-point representation questions require careful handling of mantissa and exponent. Write numbers neatly in normalised form, and remember: a positive number must start with 0.1 and a negative number with 1.0 in normalised binary floating point, unless the syllabus states otherwise.
浮点表示题需要仔细处理尾数和阶码。整齐地写出规范化形式，并记住：在规范化二进制浮点中，正数以 0.1 开头，负数以 1.0 开头，除非考纲另有规定。
For character encoding, precise Unicode and ASCII knowledge is expected. You may be asked to calculate storage requirements: Size in bytes = number of characters × bits per character ÷ 8. Always check your units — mixing bits and bytes is a common silly mistake.
对于字符编码，要求准确掌握 Unicode 和 ASCII 知识。你可能需要计算存储空间：字节大小 = 字符数 × 每字符位数 ÷ 8。务必检查单位——混淆位与字节是常见的粗心错误。
6. Systems, Hardware and Networks | 系统、硬件与网络简答题
Short-answer questions on hardware components (such as the fetch-decode-execute cycle or the role of the ALU) require precise terminology. Use technical terms like ‘program counter’, ‘CIR’, ‘MDR’ without abbreviation on first mention.
硬件组件简答题（如取指-解码-执行周期或 ALU 的作用）要求术语精确。首次提及时请使用全称，如 ‘program counter’、’CIR’、’MDR’，不缩写。
For network questions, clearly differentiate between LAN and WAN, and describe transmission media (fibre optic, twisted pair, wireless) with comparisons of bandwidth and latency. When explaining protocols, relate them to the specific layer of the TCP/IP model.
在回答网络题时，明确区分 LAN 和 WAN，描述传输介质（光纤、双绞线、无线）并比较带宽和延迟。解释协议时，要将其联系到 TCP/IP 模型的特定层。
Mark schemes often accept bullet points for short-answer questions, but ensure each bullet is a complete sentence or a standalone fact. Avoid vague statements like ‘it is faster’ without qualification — always say ‘fibre optic provides higher bandwidth and lower electromagnetic interference than copper’.
评分标准通常允许简答题使用列表形式，但每条列表必须是一个完整句子或独立事实。避免模糊表述，如”它更快”——而要说”光纤比铜缆提供更高带宽和更低电磁干扰”。
7. Database and SQL Queries | 数据库与 SQL 查询
SQL questions test your ability to write syntactically correct statements. Remember: SELECT specifies columns, FROM specifies tables, WHERE filters rows. Always end statements with a semicolon ; and use quotation marks for string literals as per the dialect used in papers.
SQL 题考查书写语法正确的语句。记住：SELECT 指定列，FROM 指定表，WHERE 过滤行。语句末尾务必加分号 ;，并根据试卷使用的方言使用引号包裹字符串字面量。
For joins, use INNER JOIN ... ON ... and clearly state the linking fields. When a question involves aggregate functions like COUNT, SUM, or AVG, you must include an appropriate GROUP BY clause. Missing GROUP BY is a notable mark-loser.
对于连接，使用 INNER JOIN ... ON ... 并明确写出连接字段。当题目涉及 COUNT、SUM 或 AVG 等聚合函数时，必须包含相应的 GROUP BY 子句。遗漏 GROUP BY 是显著的失分点。
WJEC mark schemes award marks for correct output column naming using AS aliases and logical operator usage (AND, OR, NOT). In nested queries, ensure the inner query is enclosed in parentheses and returns a single value if used with =.
WJEC 评分标准对使用 AS 别名正确命名输出列以及合理使用逻辑运算符（AND、OR、NOT）均给分。在嵌套查询中，确保内层查询用括号括起来，且当与 = 连用时返回单一值。
8. Ethics, Law and Social Impact | 道德、法律与社会影响题
These questions are assessed using levels-based marking: you need to present a balanced argument with both advantages and disadvantages. Start by stating the relevant UK legislation, such as the Data Protection Act 2018, Computer Misuse Act 1990, and Regulation of Investigatory Powers Act 2000.
这类题目采用层级评分：你需要提出平衡的论点，兼顾优缺点。先阐述相关的英国法律，如《2018 年数据保护法》、《1990 年计算机滥用法》和《2000 年调查权力规制法》。
For ethical issues like facial recognition or AI decision-making, structure your answer with a clear introduction, a paragraph on benefits, a paragraph on concerns, and a concluding judgement. Use connectives such as ‘On the one hand…’, ‘Conversely…’, and ‘In conclusion…’ to achieve coherence.
对于面部识别或 AI 决策等伦理问题，组织答案时要有清晰引言、一段说明好处、一段说明担忧，最后给出结论性判断。使用连接词如”一方面……”、”相反……”和”总之……”来保证连贯性。
Marks are awarded for using concrete examples, so mention real-world cases (e.g., NHS data sharing controversies) to illustrate your points. However, avoid extreme personal opinions — keep the tone academic and evidence-based.
给出具体事例才能得分，因此可提及真实案例（如 NHS 数据共享争议）来论证观点。但避免极端个人观点——保持学术性、基于证据的语气。
9. Time Management and Paper Strategy | 时间管理与试卷策略
At the start of the exam, scan the entire paper to gauge question difficulty and mark allocations. Divide your time proportionally: if a question counts for 10% of the total marks, spend roughly 10% of the total time on it. Leave at least 5 minutes for final review.
考试开始时，浏览全卷以判断题目难度和分值分布。按比例分配时间：若某题占总分 10%，则花大约总时间的 10% 在该题上。至少留出 5 分钟用于最后检查。
Tackle the questions you find easiest first to secure quick marks and build confidence. For programming questions, draft a quick plan on scrap paper before writing the final answer in the answer booklet — this avoids messy corrections.
先做最有把握的题，快速得分并建立信心。对于编程题，先在草稿纸上快速拟定计划，再往答题本上写最终答案——这样可以避免涂改凌乱。
If you get stuck on a part, move on and return to it later. WJEC often separates parts within a question, so later parts may be accessible even if earlier parts are incomplete. Never leave an extended question blank — even a partial response can earn marks for logical structure.
如果某部分卡住了，先跳过，稍后回来做。WJEC 常将题目分拆成独立小题，因此即使前面小问未完成，后面小问也可能照常解答。切勿让论述题空着——即使部分作答，也能因逻辑结构而得分。
10. Common Pitfalls and Final Checks | 常见失分点与最终检查
Spelling and formatting matter. In algorithm questions, missing an ENDIF or ENDFOR can break your logic. In data representation, forgetting to normalise binary floating-point numbers loses easy marks.
拼写和格式很重要。算法题中，漏写 ENDIF 或 ENDFOR 可能破坏逻辑。数据表示题中，忘记规范化二进制浮点数会丢掉容易的分数。
Another common slip is writing SQL keywords like SELECT in lowercase — while not penalised heavily, uppercase makes your answer clearer to the examiner. Also, ensure table and field names are spelled exactly as given in the question.
另一个常见疏忽是用小写书写 SQL 关键字如 SELECT——虽然扣分不重，但大写能让考官更清晰地看懂你的答案。同时，确保表名和字段名与题目给出的拼写完全一致。
Finally, check arithmetic operations: binary addition and hexadecimal conversions should be double-verified. A single bit error can propagate. Use a different method to verify (e.g., convert hex back to denary). Carefully read the question again to ensure you haven’t misinterpreted the command word.
最后，检查算术运算：二进制加法和十六进制转换应复核。一个位错误可能蔓延。用不同方法验证（如将十六进制转回十进制）。再次仔细读题，确保没有误解指令词。
Published by TutorHao | Computer Science Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导Cancel reply