Key Points for Experimental/Practical Assessment in Year 12 CAIE Mathematics | Year 12 CAIE 数学：实验/实践考核要点

📚 Key Points for Experimental/Practical Assessment in Year 12 CAIE Mathematics | Year 12 CAIE 数学：实验/实践考核要点

Year 12 CAIE Mathematics does not have a separate laboratory-based practical exam like the sciences; however, the syllabus is rich with experimental and practical thinking. Across Statistics, Mechanics, and Pure Mathematics, you are constantly asked to apply modelling principles, design sampling strategies, interpret real data, and critique experimental assumptions. These are the ‘practical assessment’ skills that underpin high marks in structured questions. This article collects the essential experimental and practical assessment points that every Year 12 student must master to succeed in the CAIE AS Mathematics exam. We uncover the core ideas behind statistical experiments, mechanical modelling, and the real-world validation of mathematical results, equipping you with the precise language and logic examiners expect.

尽管CAIE Year 12 数学没有像科学科目那样的独立实验操作考试，但教学大纲中充满了实验与实践思维。在统计学、力学和纯数学中，你经常需要应用建模原则、设计抽样策略、解读真实数据以及评判实验假设。这些正是构成高分结构化题目的“实践考核”技能。本文收集了每一位Year 12学生必须在CAIE AS数学考试中掌握的核心实验与实践考核要点。我们将揭示统计实验、力学建模以及数学结果在现实世界中验证背后的关键思想，帮助你掌握考官期望的准确表达和逻辑。

1. Mathematical Modelling and Real-World Problems | 数学建模与实际问题

The act of translating a real-world scenario into a mathematical form is the first practical step. You must identify variables, make simplifying assumptions, and construct a model using functions, equations, or inequalities. For example, optimising the area of a field given a fixed amount of fencing leads to a quadratic model. Examiners will test your ability to interpret the solution in context and discuss its limitations, which is a practical skill that mirrors experimental error analysis.

将现实情景转化为数学形式是第一个实践步骤。你必须识别变量、做出简化假设，并用函数、方程或不等式构建模型。例如，在给定固定围栏长度的情况下优化场地面积，就会得到一个二次模型。考官会考查你结合实际背景解读解的能力，并讨论其局限性，这是一项类似于实验误差分析的实践技能。

• Identify key quantities: determine what is constant and what varies.
• 识别关键量： 确定哪些是常量，哪些是变量。
• Check validity: substitute specific values to see if the model behaves as expected.
• 检验有效性： 代入具体数值，观察模型是否符合预期。

2. Modelling Assumptions in Mechanics | 力学中的建模假设

Mechanics questions are built on a set of practical simplifications that mimic controlled laboratory conditions. Common assumptions include treating an object as a particle (neglecting its size), rods as light (no mass), strings as inextensible, and surfaces as smooth (no friction). You are expected to know when each assumption is appropriate and how it affects the resulting equations. Describing these assumptions accurately is a direct demand of the practical assessment style.

力学问题建立在一系列模拟受控实验条件的实用简化基础之上。常见的假设包括把物体看作质点（忽略其尺寸）、杆视为轻质（无质量）、绳子不可伸长、以及表面光滑（无摩擦）。你需要知道每个假设在什么情况下适用，以及它如何影响推导出的方程。准确描述这些假设是实践考核风格题目的直接要求。

Assumption (假设)	Meaning (含义)	When to use (何时使用)
Particle	Mass at a point, no rotation	Object dimensions negligible
Light string/rod	Mass = 0, tension same throughout	Mass is small compared to others
Inextensible string	Length constant, acceleration same for connected particles	Stretching is insignificant
Smooth surface	No friction force	Friction is negligible or not required

3. Types of Data in Statistical Experiments | 统计实验中的数据类型

Before any statistical experiment, understanding the nature of data is paramount. CAIE Statistics 1 requires you to distinguish between qualitative and quantitative data, and further between discrete and continuous quantitative data. In practical terms, the type of data dictates which diagrams and summary statistics are appropriate. Misclassifying data leads to invalid experimental conclusions, making this a crucial assessment point.

在任何统计实验之前，理解数据的性质至关重要。CAIE 统计学1 要求你区分定性数据和定量数据，并进一步区分离散定量数据和连续定量数据。在实际操作中，数据类型决定了哪些图表和汇总统计量是合适的。错误分类数据会导致无效的实验结论，因此这是一个关键的考核点。

Qualitative data are non-numerical and describe categories or attributes, such as colour or type of material. Quantitative data arise from measurements or counts. Examiners will often present a scenario and ask you to classify variables, a skill directly linked to planning an experiment properly.

定性数据是非数值的，描述类别或属性，如颜色或材料类型。定量数据来自测量或计数。考官常常会给出一个场景，要求你对变量进行分类，这一技能直接关系到正确规划实验。

4. Sampling Techniques as Experimental Design | 抽样技术作为实验设计

In many practical situations, you cannot test an entire population; you must select a sample. The sampling method is an experimental design choice. The CAIE syllabus covers simple random sampling, stratified sampling, systematic sampling, and quota sampling. Each has advantages and biases that affect the reliability of experimental results. Knowing when to use a particular method demonstrates practical understanding.

在许多实际情况下，你无法测试整个总体，必须选择一个样本。抽样方法就是一种实验设计选择。CAIE 大纲涵盖简单随机抽样、分层抽样、系统抽样和配额抽样。每种方法都有其优点和可能带来的偏差，这些会影响实验结果的可靠性。知道何时使用特定方法体现了实践理解能力。

• Simple random sampling: every member equally likely, free from bias but needs a sampling frame.
• 简单随机抽样： 每个成员被选中的概率相同，无偏差，但需要抽样框。
• Stratified sampling: population divided into groups (strata), sample proportional to group size; representative.
• 分层抽样： 总体分成若干组（层），样本大小按层的大小比例分配；具代表性。
• Systematic sampling: choose every k-th element; easy but can introduce pattern bias.
• 系统抽样： 选取每第k个个体；简便，但可能引入周期性偏差。

5. Principles of Experimental Design: Control, Randomisation, and Replication | 实验设计原则：控制、随机化和重复

Although not always labelled explicitly, good experimental design rests on three pillars. Control means keeping other variables constant so that the effect of the factor of interest can be isolated. Randomisation ensures that the allocation of treatments to units is free from deliberate bias. Replication allows you to assess variability and produce more reliable estimates. These concepts appear in hypothesis testing contexts and in critiquing experiments.

尽管并非总是被明确标注，但良好的实验设计建立在三大支柱之上。控制意味着保持其他变量不变，以便分离出所关注因素的影响。随机化确保将处理分配给实验单元时没有主观偏差。重复则可以让你评估变异性并得出更可靠的估计。这些概念出现在假设检验和实验评价的语境中。

For example, when comparing two fertiliser treatments on plant growth, you must control light, water, and soil type. Assign plants to treatment groups randomly, and use many plants per group (replication). In CAIE problems, you might be asked to suggest why an experiment lacks validity; these principles are your diagnostic toolkit.

例如，在比较两种肥料处理对植物生长的影响时，你必须控制光照、水分和土壤类型，将植物随机分配到各组，并且每组使用多株植物（重复）。在CAIE问题中，你可能会被问到为什么某个实验缺乏有效性；这些原则就是你的诊断工具箱。

6. Variables and Confounding in Experiments | 实验中的变量与混杂

Clarity about independent (explanatory) and dependent (response) variables is essential. The independent variable is the one you deliberately change, and the dependent variable is what you measure. A confounding variable is one that influences both the independent and dependent variables, making it impossible to establish a clear cause-effect relationship. In exam questions, identifying a likely confounding variable often earns high marks for ‘evaluation’.

清晰地分辨自变量（解释变量）和因变量（响应变量）很关键。自变量是你有意改变的变量，因变量是你测量的结果。混杂变量是同时影响自变量和因变量的变量，使得无法建立清晰的因果关系。在考试题目中，识别出一个可能的混杂变量常常能在“评估”题上获得高分。

Practical tip: when reading a description of an experiment, underline what was changed and what was measured, then think about any hidden factor that could link them spuriously. This mimics real scientific review.

实践窍门：阅读实验描述时，划线标出什么被改变了，什么被测量了，然后思考任何可能虚假地将它们联系在一起的隐藏因素。这模仿了真实的科学评审过程。

7. Randomised Block and Paired Designs | 随机区组与配对设计

When experimental units have known variability (e.g., different patches of a field, different age groups), a completely randomised design may be inefficient. A randomised block design groups similar units into blocks, and treatments are randomised within each block. A matched pairs design pairs subjects that are similar, then randomly assigns treatments within each pair. These designs control for extraneous variation and increase the sensitivity of the experiment, a key practical assessment concept.

当实验单元存在已知变异时（例如田地不同区块、不同年龄段），完全随机设计可能效率不高。随机区组设计将相似的单元归入区组，然后在每个区组内随机分配处理。配对设计将相似受试者配对，然后在每一对内部随机分配处理。这些设计可以控制额外的变异，提高实验的灵敏度，是一个关键的实践考核概念。

In CAIE Statistics 1, you might be given data from a paired test and asked to perform a hypothesis test for the mean difference. Understanding the experimental reason for pairing (to eliminate subject-to-subject variability) is essential.

在CAIE统计学1中，你可能会遇到配对实验的数据，并被要求对均值差进行假设检验。理解配对的实验原因（消除个体间变异）至关重要。

8. Reliability, Validity, and Bias in Practical Work | 实践工作中的可靠性、有效性与偏差

An experiment or measurement process must be reliable (yielding consistent results under repeated trials) and valid (measuring what it intends to measure). Sources of bias, such as measurement bias (faulty equipment), sampling bias (non-representative sample), and observer bias, can undermine both. Exam questions will ask you to comment on how bias can be reduced, for instance by using randomisation, calibration, or blinding. This directly reflects practical assessment criteria.

实验或测量过程必须是可靠的（在重复试验下产生一致的结果）和有效的（测量其意图测量的东西）。偏差来源，如测量偏差（仪器故障）、抽样偏差（样本不具代表性）和观察者偏差，都可能削弱可靠性和有效性。考题会要求你评论如何减少偏差，例如通过随机化、校准或盲法。这直接反映了实践考核的标准。

Blinding: in a drug trial, single-blind means patients don’t know which treatment they receive; double-blind means neither patients nor clinicians know. This eliminates placebo effects and observer bias, strengthening experimental validity.

盲法：在药物试验中，单盲指患者不知道自己接受哪种治疗；双盲指患者和临床医生都不知道。这消除了安慰剂效应和观察者偏差，增强了实验有效性。

9. The Normal Distribution as a Model for Experimental Data | 正态分布作为实验数据的模型

Many continuous variables measured in experiments (heights, masses, measurement errors) approximately follow a normal distribution. In Year 12, you learn to standardise with Z = (X − μ) / σ, use the standard normal table, and apply the model to real contexts. The practical assessment point is to judge when the normal model is suitable: data should be roughly symmetric and unimodal. Recognizing when the central limit theorem justifies normality for sample means is also valuable.

实验中测量的许多连续变量（身高、质量、测量误差）都近似服从正态分布。在Year 12，你要学会使用 Z = (X − μ) / σ 进行标准化，使用标准正态表，并将模型应用于真实情境。实践考核点在于判断正态模型何时适用：数据应大致对称且单峰。认识到中心极限定理为样本均值的正态性提供依据也很有价值。

If X ~ N(μ, σ²), then Z = (X − μ)/σ ~ N(0,1)

若 X ~ N(μ, σ²)，则 Z = (X − μ)/σ ~ N(0,1)

10. Correlation, Regression, and Causal Inference | 相关性、回归与因果推断

In practical investigations, you often look for relationships between two variables. Scatter diagrams and the product moment correlation coefficient (r) measure linear association. However, the most critical experimental point is: correlation does not imply causation. A high r value might be due to a lurking variable. The regression line y = a + bx can be used for prediction only within the range of observed data. Extrapolation beyond that range is a common practical error.

在实际调查中，你经常需要寻找两个变量之间的关系。散点图和积矩相关系数 (r) 衡量的是线性关联程度。然而，最关键的实验要点是：相关性并不意味着因果关系。一个高r值可能是由潜在的混杂变量导致的。回归直线 y = a + bx 仅可用于观测数据范围内的预测。超出该范围进行外推是一种常见的实践错误。

CAIE questions often give a context (e.g., ice cream sales and drowning incidents) and ask you to explain why a high correlation does not prove causation (hot weather is the lurking variable). This tests your experimental reasoning directly.

CAIE考题经常给出一个情境（例如冰淇淋销量与溺水事件），并要求你解释高相关性为何不能证明因果关系（炎热天气是潜在变量）。这直接测试你的实验推理能力。

11. Data Presentation and Interpretation as Practical Skills | 数据展示与解读作为实践技能

Producing clear, accurate diagrams (box plots, histograms, cumulative frequency curves) is not just a plotting exercise; it is a communication tool for experimental results. You must choose the appropriate diagram for the data type and be able to extract key features: median, quartiles, skewness, and outliers. Interpretation includes commenting on spread and central tendency in the context of the experiment. These are the practical skills that make raw data meaningful.

绘制清晰准确的图（箱线图、直方图、累积频率曲线）不仅仅是描点练习，更是传达实验结果的交流工具。你必须根据数据类型选择合适的图，并能提取关键特征：中位数、四分位数、偏度以及异常值。解读包括结合实验背景评论数据的离散程度和集中趋势。这些正是让原始数据变得有意义的实践技能。

For instance, a box plot comparing two experimental groups can instantly show differences in location and dispersion, supporting conclusions about the treatment effect. Always label axes and provide a title.

例如，比较两个实验组的箱线图可以立即显示位置和分散程度的差异，为处理效果的结论提供支持。务必标注坐标轴并加上标题。

12. Validation of Models and Error Analysis | 模型验证与误差分析

After solving a mathematical model, you must check its reasonableness. Does the answer match real-world expectations? For a projectile, negative time is physically impossible. For a measurement model, percentage error = |(estimated value − true value) / true value| × 100%. In practical assessment, you may be asked to discuss how to improve an experiment or model: reduce measurement error, take more readings, or refine assumptions.

解出数学模型后，你必须检验其合理性。答案是否符合现实预期？对于抛体运动，负时间在物理上是不可能的。对于测量模型，百分比误差 = |(估计值 − 真实值)/真实值| × 100%。在实践考核中，你可能被要求讨论如何改进实验或模型：减少测量误差，获取更多读数，或完善假设。

This reflective step completes the experimental cycle: model, analyse, validate, and refine. It is exactly what examiners mean by ‘evaluate the model’ in high-tariff questions.

这一反思步骤完成了实验循环：建模、分析、验证和细化。这正是在高分值题目中考官所要求的“评价模型”。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）