📚 IB vs AQA Computer Science: Marking Criteria Analysis | IB 与 AQA 计算机科学评分标准分析
Understanding how exam boards award marks is essential for success in any high‑stakes qualification. For students tackling Computer Science at the pre‑university level, two rigorous pathways stand out: the International Baccalaureate (IB) Diploma Programme and AQA A‑level. Although both aim to build computational thinking, their assessment philosophies, weightings, and grading rubrics differ markedly. This article dissects the marking criteria of IB and AQA Computer Science – from assessment objectives and internal assessments to paper structures and grade boundaries. By examining what examiners truly value, you can sharpen your revision strategy and maximise your final grade.
理解考试局如何评分对于在任何高利害考试中取得成功都至关重要。对于在大学预科阶段学习计算机科学的学生来说，有两条严格的路径：国际文凭（IB）大学预科项目和AQA A‑level。尽管两者都旨在构建计算思维，但它们的测评理念、权重分配和评分量规却明显不同。本文剖析了IB和AQA计算机科学的评分标准——从评估目标和内部评估到试卷结构与等级分数线。通过审视考官真正看重什么，你可以优化复习策略，最大程度地提高最终成绩。
1. Assessment Objectives at a Glance | 评估目标概览
Both IB and AQA define Assessment Objectives (AOs) that describe the skills students must demonstrate. In IB Computer Science, the four objectives are: (1) Know and understand (2) Apply and use (3) Construct, analyse and evaluate (4) Investigate and construct. AQA, meanwhile, splits its objectives into AO1 (Knowledge and understanding), AO2 (Application) and AO3 (Design, development, evaluation and conclusion). While the phrasing is similar, the way these objectives are weighted across papers is very different, which directly impacts where you should invest your study time.
IB 和 AQA 都界定了描述学生必须展示技能的评估目标（AO）。在 IB 计算机科学中，四个目标是：（1）了解与理解（2）应用与使用（3）构建、分析与评估（4）调查与构建。而 AQA 将其目标划分为 AO1（知识与理解）、AO2（应用）和 AO3（设计、开发、评估与结论）。尽管措辞相似，但这些目标在不同试卷中的权重分配方式差异很大，这直接影响你应在学习时间上投入的方向。
The table below summarises the approximate percentage weight per AO in each qualification:
下表总结了每项资格中每个 AO 的大致百分比权重：
| AO | IB SL/HL | AQA A‑level |
|---|---|---|
| AO1 / Know & understand | ~30% | ~35% |
| AO2 / Apply & use | ~35% | ~20% |
| AO3 / Construct, evaluate & design | ~25% | ~30% |
| AO4 / Investigate & construct (IA only) | ~10% | N/A (included in AO3) |
Notice that IB places a heavier emphasis on application (AO2) through programming and algorithmic thinking, while AQA balances application with design and evaluation. For IB students, the internal assessment also carries a separate AO4 that rewards independent inquiry.
请注意，IB 通过编程和算法思维更加强调应用（AO2），而 AQA 则在应用与设计和评估之间取得了平衡。对于 IB 学生而言，内部评估还包含一个独立的 AO4，奖励独立探究。
2. External Assessment Structure in IB | IB 外部评估结构
The IB Computer Science course (first teaching 2014, with an updated guide for 2025) features two external examination papers for both Standard Level (SL) and Higher Level (HL). Paper 1 tests the core syllabus through multiple‑choice, short‑answer and structured questions. HL students face more complex data‑structure questions and have a longer examination time (150 minutes vs 90 for SL). Paper 2 is an options paper: students study one of four options (e.g. databases, web science) and answer questions on that option. At HL, Paper 2 also includes a compulsory algorithm‑based section that is not present at SL.
IB 计算机科学课程（2014 年开始教学，2025 年更新指南）为标准级别（SL）和高级级别（HL）设置了两份外部考试试卷。试卷 1 通过选择题、简答题和结构化问题测试核心大纲。HL 学生面对更复杂的数据结构问题，考试时间也更长（150 分钟 vs SL 的 90 分钟）。试卷 2 是选修主题试卷：学生从四个选修主题（如数据库、网络科学）中学习一个，并回答有关该主题的问题。在 HL 级别，试卷 2 还包括一个 SL 中没有的必答算法部分。
- Paper 1 marks: SL 70 marks (45%), HL 100 marks (40%)
- Paper 2 marks: SL 45 marks (25%), HL 80 marks (30%)
剩余百分比分配给内部评估（IA）。
3. External Assessment Structure in AQA | AQA 外部评估结构
AQA A‑level Computer Science (7517) is assessed through two written papers and a non‑exam assessment (NEA). Paper 1 is an on‑screen programming exam worth 40% of the A‑level. Students write, test and refine code in a pre‑released skeleton program environment. Paper 2 is a theoretical written paper (40%) covering fundamentals of data representation, computer systems, networking, databases and big data. The examination lasts 2 hours 30 minutes and includes structured questions and extended writing. AS‑level (first year) has its own lighter papers but the A‑level builds directly upon them.
AQA A‑level 计算机科学（7517）通过两份书面试卷和一项非考试评估（NEA）进行评估。试卷 1 是占 A‑level 总成绩 40% 的上机编程考试。学生在预发布的骨架程序环境中编写、测试和优化代码。试卷 2 是理论书面试卷（40%），涵盖数据表示、计算机系统、网络、数据库和大数据等基础知识。考试时长为 2 小时 30 分钟，包含结构化问题和扩展写作题。AS‑level（第一年）有自己的较轻试卷，但 A‑level 直接在其基础上构建。
- Paper 1 (on‑screen): 75 marks, 40%
- Paper 2 (written): 75 marks, 40%
- NEA (coursework): 75 marks, 20%
4. Internal Assessment: IB IA vs AQA NEA | 内部评估：IB IA 与 AQA NEA
The IB internal assessment requires students to develop a computational solution for a real‑world client. SL candidates create a product with 30 hours of work, while HL extends to 40 hours and demands a more sophisticated solution including advanced features such as recursion or dynamic data structures. The IA is marked against five criteria: A (Planning), B (Solution overview), C (Development), D (Functionality) and E (Evaluation), with a total of 34 marks (SL: 30%, HL: 20% after scaling). Each criterion has descriptors for high, medium and low performance; examiners look for clear evidence of iterative design, justified algorithm selection and thorough usability testing.
IB 内部评估要求学生为现实世界客户开发一个计算解决方案。SL 考生用 30 小时的课内时间制作一个产品，而 HL 扩展到 40 小时，要求更复杂的解决方案，包括递归或动态数据结构等高级特性。IA 根据五项标准评分：A（规划）、B（方案概述）、C（开发）、D（功能性）和 E（评估），总计 34 分（SL 占比 30%，HL 经换算后占 20%）。每项标准都有高中低表现的描述符；考官寻找关于迭代设计、算法选择依据以及全面可用性测试的明确证据。
In contrast, AQA's NEA is a practical programming project that accounts for 20% of the final A‑level. Students must produce a documented solution that solves a problem of their choice. The 75‑mark project is assessed across six sections: analysis (10 marks), design (15 marks), technical solution (20 marks), testing (10 marks), evaluation (10 marks) and documentation (10 marks). The emphasis is strongly on technical complexity and evidence of systematic testing. Unlike IB’s client‑focused IA, AQA's NEA can target a hypothetical user as long as the problem is sufficiently complex.
相比之下，AQA 的 NEA 是一个占最终 A‑level 成绩 20% 的实践编程项目。学生必须制作一个解决自选问题的文档化解决方案。这个 75 分的项目按六个部分评估：分析（10 分）、设计（15 分）、技术方案（20 分）、测试（10 分）、评估（10 分）和文档（10 分）。重点在于技术复杂性和系统化测试的证据。与 IB 以客户为中心的 IA 不同，AQA 的 NEA 可以针对假想用户，只要问题足够复杂。
5. Marking Rubric Nuances in IB | IB 评分量规的细微差异
IB's external papers use a mixture of point‑based and levels‑based marking. For structured questions, marks are often awarded for specific keywords, correct syntax and logical reasoning. The options paper (Paper 2) is heavily rubric‑driven: answers must demonstrate depth of knowledge beyond the core. In algorithm questions, you must trace variables correctly and state complexities using Big‑O notation, e.g. O(n log n). No mark is given for vague statements; precision is paramount. The IA, however, is holistically marked using best‑fit approaches. That means a single piece of evidence can lift a criterion descriptor even if other aspects are weaker.
IB 的外部试卷混合使用分点给分和等级给分。对于结构化问题，通常根据特定关键词、正确语法和逻辑推理来给分。选修主题试卷（试卷 2）高度依赖量规：答案必须展示超越核心知识的深度。在算法题中，你必须正确追踪变量，并用大O符号陈述复杂度，如 O(n log n)。模糊的陈述不会给分；精确性至关重要。然而，IA 采用整体最佳匹配方法评分。这意味着即使其他方面较弱，一项单个证据也可以提升某个标准的等级描述。
6. AQA Command Words and Mark Allocation | AQA 指令词与分数分配
AQA's papers are built around command words that signal exactly what the examiner wants. Words like 'State', 'Describe', 'Explain' and 'Evaluate' follow a strict hierarchy of demand. For instance, a 2‑mark 'Describe' question requires identification plus some development; a 4‑mark 'Explain' needs a clear cause‑effect chain. Longer 9‑mark questions often require a reasoned argument or a discussion of trade‑offs. AQA publishes mark schemes that are extremely formulaic: if a candidate does not use the expected terminology, they may lose accessible marks. Programming questions on Paper 1 are machine‑marked for functionality and then checked by senior examiners for code efficiency and style.
AQA 的试卷围绕指令词构建，这些指令词精确地指示考官想要什么。像“陈述”、“描述”、“解释”和“评估”等词遵循严格的需求层级。例如，一个 2 分的“描述”题需要识别加上一些展开；一个 4 分的“解释”题则需要清晰的因果链条。较长的 9 分题通常需要合理的论述或权衡讨论。AQA 公布的评分方案非常公式化：如果考生未使用预期术语，就可能失去可获得的分数。试卷 1 的编程题由机器评分其功能性，然后由资深考官检查代码效率和风格。
7. Comparing Grade Boundaries and Scaling | 比较等级分数线与换算
IB Computer Science grades range from 1 to 7, with 7 being the highest. The final subject mark is a weighted sum of component marks and is scaled to fit a fixed percentage boundary that can shift slightly each session. In May 2023, the grade boundary for a 7 at HL was approximately 79% overall, while an SL 7 was around 72%. These boundaries reflect the difficulty of the papers and the moderation of the IA. AQA, on the other hand, uses A* to E grades for the full A‑level. A* is typically awarded at around 75‑80% of the uniform mark (UMS). Raw mark to UMS conversion changes each year, but an average raw boundary for an A* may be 65‑70% on the written papers. This means that AQA's raw papers are often more challenging to score highly on, so boundaries are correspondingly lower.
IB 计算机科学的等级从 1 到 7，7 为最高。最终科目分数是各组成部分分数的加权和，并经过换算以适应固定的百分比界限，每次考试可能略有浮动。在 2023 年 5 月，HL 获得 7 级的总分大约需要 79%，而 SL 7 级约为 72%。这些界限反映了试卷难度与 IA 的调整。AQA 方面，完整的 A‑level 使用 A* 到 E 的等级。A* 通常在一致分数（UMS）的 75–80% 左右授予。原始分到 UMS 的转换每年变化，但 A* 的平均原始分界限可能在书面试卷的 65–70% 左右。这意味着 AQA 的原始试卷往往更难获得高分，因此界限相应较低。
8. How IA and NEA Moderation Affect Your Marks | IA 与 NEA 审核如何影响你的分数
Both IB and AQA use moderation processes to ensure consistency between schools. IB selects a sample of IA work from each school; if the teacher’s marking is found to be too generous or too harsh, the whole cohort’s marks are adjusted. For AQA, the NEA is marked by the teacher and then externally moderated by AQA. If marks are judged to be generous, all candidates at the centre may have their NEA scores reduced. It is therefore crucial that teachers follow the published exemplar material closely. For students, this means your project must visibly meet every descriptor; borderline evidence often results in lower marks after moderation.
IB 和 AQA 都使用审核流程来确保学校间的一致性。IB 从每所学校抽取部分 IA 作业样本；如果发现教师评分过于宽松或过严，整个班级的分数都会被调整。AQA 的 NEA 由教师评分，然后由 AQA 外部审核。如果认定分数偏松，该中心所有考生的 NEA 成绩都可能被降低。因此，教师必须严格遵循公布的范例材料。对于学生而言，这意味着你的项目必须明显地满足每一条描述符；在审核后，边缘证据常常导致得分降低。
9. Exam Tactics: Using Mark Schemes to Your Advantage | 考试策略：善用评分方案
To maximise marks in IB, practice with official past papers and memorise the mark scheme phrasing for key concepts, such as “layers of abstraction” or “advantages of virtualisation”. In algorithm tracing, always show variable states in a table; a correct table can earn method marks even if the final answer is wrong. In AQA, download the latest command‑word taxonomy and match every question to the required depth. For Paper 1 programming, test your code against all sample data – including edge cases – to secure functionality marks. In both qualifications, time management is vital: allocate roughly one minute per mark, and leave space to return to high‑weight extended‑response questions.
在 IB 中最大化得分，练习官方历年真题，并记住关键概念的评分方案用语，如“抽象层级”或“虚拟化的优势”。在算法追踪中，始终用表格显示变量状态；即使最终答案错误，正确的表格也能获得方法分。在 AQA 中，下载最新的指令词分类，并将每个问题与所需深度匹配。对于试卷 1 编程，用所有样本数据测试你的代码——包括边界情况——以确保功能性得分。在两项资格中，时间管理至关重要：大致分配每分钟 1 分，并留出时间回头处理高分值扩展回答题。
10. Conclusion: Where to Focus Your Effort | 结论：努力的重点方向
IB Computer Science rewards holistic understanding and independent inquiry; its IA demands genuine client engagement and reflective evaluation. AQA A‑level prizes precision, technical depth and systematic problem solving, with the NEA acting as a proving ground for coding craftsmanship. Whichever route you follow, align your revision with the published assessment objectives and mark schemes. If your school offers both, consider your strengths: excellent programmers who love theoretical nuance often shine in AQA, while creative problem solvers who enjoy extended research might prefer the IB.
IB 计算机科学奖励整体理解和独立探究；其 IA 要求真正的客户参与和反思性评估。AQA A‑level 珍视精确、技术深度和系统化问题解决能力，NEA 则作为编码技艺的试验场。无论你选择哪条路，都要将复习与公布的评估标准和评分方案对齐。如果你的学校同时提供二者，请考虑自身优势：热爱理论细微差别的优秀程序员往往在 AQA 中脱颖而出，而喜欢扩展研究的创造性问题解决者可能更偏好 IB。
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