📚 GCSE OCR Further Maths: Experimental and Practical Assessment Essentials | GCSE OCR 进阶数学：实验与实践考核要点
In OCR GCSE Further Mathematics, there is no separate coursework or practical exam, but 'experimental and practical assessment' refers to the set of skills examined across written papers. These skills involve designing surveys, planning data collection, using probability simulations to model real-world problems, and critically evaluating findings. Mastery of these practical techniques is essential for tackling AO2 (reasoning) and AO3 (problem solving) questions, which often present you with an incomplete investigation and ask you to refine the method or interpret the results. The following guide consolidates all key points you need to demonstrate a strong practical-mathematical mindset.
在OCR GCSE进阶数学中虽然没有独立的课程作业或实践考试，但“实验与实践考核”指贯穿笔试考察的一系列技能。这些技能包括设计调查、规划数据收集、利用概率模拟解决现实问题，以及批判性地评估结论。要攻克AO2（推理）和AO3（问题解决）类题目，这些题目常给出一个未完成的调查，要求你完善方法或解读结果，就必须掌握这些实用技能。以下指南汇总了所有你需要展示的实践数学思维的关键要点。
1. Understanding the Nature of Practical Tasks in the Exam | 理解考试中实践任务的性质
Exam questions may ask you to propose a hypothesis, design a data-collection method, carry out a simulation using random numbers, or critique a given experimental plan. These are 'paper-based practicals', testing whether you can apply statistical and probabilistic concepts as if you were running a real investigation.
考题可能要求你提出假设、设计数据收集方法、使用随机数进行模拟，或批判给定的实验计划。这些都是“纸笔实践”，考察你是否能够像真实调查那样应用统计与概率概念。
The practical elements are embedded in topics such as sampling, probability experiments, and statistical diagrams. Your ability to justify choices—such as why you used simple random sampling instead of stratified—is often more important than performing long calculations.
实践元素嵌入在抽样、概率实验和统计图表等主题中。你为自己的选择提供理由的能力——比如为什么使用简单随机抽样而不是分层抽样——往往比进行冗长的计算更重要。
Marks are awarded for clear communication of method, recognition of bias, and sensible interpretation of numerical results within the context of the problem. Always imagine you are explaining your process to someone who has not seen the data.
得分点在于清晰地表述方法、识别偏差，以及结合问题背景合理阐释数值结果。请始终设想你正在向一个没见过数据的人解释整个流程。
2. Formulating a Testable Hypothesis | 构建可检验的假设
Every practical investigation begins with a hypothesis—a statement that can be tested using data. In GCSE questions, you might be asked to write a suitable hypothesis or identify the null hypothesis for a probability experiment.
任何实践调查都从一个可检验的假设开始——即可用数据验证的陈述。在GCSE题目中，你可能需要写出一个合适的假设，或辨识概率实验中的零假设。
A good hypothesis is specific, measurable, and relates the variables clearly. For instance: 'The probability of obtaining heads when flipping a bent coin is not 0.5' is a valid hypothesis for a simulation experiment.
一个好的假设应该明确、可度量，并且清晰关联变量。例如：“抛掷一枚弯曲硬币得到正面的概率不是0.5”就是模拟实验的一个有效假设。
You may be asked to write the hypothesis in words or using probability notation like P(heads) ≠ 0.5. Avoid vague statements such as 'the coin is biased'; you must quantify what 'biased' means in probability terms.
你可能需要用文字或概率符号如 P(正面) ≠ 0.5 写出假设。避免诸如“硬币有偏差”这样的模糊表述；你必须用概率术语量化“偏差”的含义。
3. Types of Data and Measurement Considerations | 数据类型与测量考量
Before collecting data, you need to identify whether the variables are discrete or continuous. Discrete data can take only particular values (e.g. number of heads in 50 tosses), while continuous data can take any value within a range (e.g. time taken to solve a puzzle).
收集数据前，你需要确定变量是离散还是连续。离散数据只能取特定值（如50次抛掷中正面的次数），而连续数据可以在一个范围内取任意值（如解谜所花时间）。
The measurement instrument and its precision affect the quality of your data. Always comment on the accuracy of a stopwatch (nearest 0.1 s) or a ruler (nearest mm), and consider whether repeated measurements could reduce random errors.
测量仪器及其精度会影响数据质量。务必评论秒表（精确到0.1秒）或尺子（精确到毫米）的准确度，并思考重复测量是否可以减少随机误差。
Data can also be categorical (e.g. colour, make of car) or quantitative. Knowing the data type determines which diagrams and averages are appropriate—for categorical data you can only find the mode, not the mean.
数据还可以是类别型（如颜色、汽车品牌）或数值型。了解数据类型决定哪些图表和平均数适用——对于类别型数据只能求众数，不能求平均数。
4. Choosing and Justifying a Sampling Method | 选择并论证抽样方法
In a practical investigation, you rarely have time to test the entire population. The exam may ask you to choose between random, stratified, systematic, or convenience sampling and explain why your choice is fit for purpose.
在实践调查中，你很少有时间检测整个总体。考题可能要求你在随机抽样、分层抽样、系统抽样或便利抽样中选择，并解释你的选择为何合适。
Simple random sampling ensures every member has an equal chance of being selected, minimising bias. You can generate random numbers using a calculator or a random number table to select items from a numbered list.
简单随机抽样确保每名成员被选中的机会均等，使偏差最小化。你可以用计算器或随机数表生成随机数，从编号列表中选择项目。
e for each stratum using: (stratum size ÷ population size) × total sample size. This guarantees proportional representation.
当总体自然分为不同组（层）时，比如学校中的年级，就用分层抽样。每层样本量的计算为：(层大小 ÷ 总体大小) × 总样本量。这保证了比例代表性。
Stratum sample size = (nₛ ÷ N) × n
层样本量 = (nₛ ÷ N) × n
Always mention potential bias: a convenience sample of your friends is unlikely to be representative, so a conclusion drawn from it may not be valid for the whole population.
一定要提到潜在偏差：朋友的便利样本很可能不具备代表性，因此从中得出的结论对全体可能无效。
5. Designing a Fair Experiment and Controlling Variables | 设计公平实验与控制变量
A 'fair test' means changing only one variable (the independent variable) while keeping all other conditions the same. For instance, if testing how the number of index cards affects the strength of a bridge model, the type of cards, bridge length, and loading method must be constant.
“公平实验”意味着只改变一个变量（自变量），而保持其他所有条件不变。例如，检验卡片张数如何影响模型桥的强度时，卡片类型、桥的长度和加载方式必须保持不变。
The dependent variable is what you measure (e.g. mass held before collapse). Clearly defining these variables in your description will earn method marks, even if your design is not fully implemented.
因变量是你测量的变量（如倒塌前承受的质量）。在描述中明确定义这些变量可以获得方法分，即使你的设计没有被完全实施。
Control groups and repeated trials are part of good experimental design. If testing a new teaching method, a control group taught conventionally provides a baseline for comparison.
控制组和重复试验是良好实验设计的一部分。如果测试一种新教学法，一个接受传统教学的控制组就提供了比较的基线。
6. Running Probability Simulations and Using Random Numbers | 进行概率模拟与使用随机数
Probability simulations are used when a real experiment is expensive, time‑consuming, or impractical. The exam may ask you to design a simulation using dice, coins, or random numbers to model a scenario like 'the chance of rain on 3 out of 5 days'.
当真实实验昂贵、耗时或不可行时，就使用概率模拟。考试可能要求你设计一个使用骰子、硬币或随机数的模拟，来建模比如“5天中有3天下雨”的情景。
To simulate a probability of 0.3, you can allocate digits 0–2 as 'success' and 3–9 as 'failure' in a random number table. You must state your allocation rule clearly, e.g. '00–29 represents rain, 30–99 represents no rain'.
要模拟0.3的概率，你可以在随机数表中将数字0–2定为“成功”，3–9定为“失败”。你必须清晰地声明分配规则，例如“00–29代表下雨，30–99代表无雨”。
When using dice, you can define outcomes proportionally: a six-sided dice gives each face probability 1/6. To simulate an event with probability 1/3, you can say 'throwing a 1 or 2 means success'.
使用骰子时，你可以按比例定义结果：六面骰子每个面的概率是1/6。要模拟概率为1/3的事件，可以规定“投出1或2表示成功”。
| Simulation Method | Example Allocation | Context |
| Coin | Head = pass, Tail = fail | Probability 0.5 events |
| Standard dice | 1,2 → success; 3–6 → fail | Probability 1/3 |
| Random number table | 00–24 → success (p=0.25) | Any probability expressed as decimal |
模拟方法及分配示例
You should run a large number of trials (at least 100) and calculate the relative frequency, which should converge to the theoretical probability as the number of trials increases. State this link clearly.
你应该运行大量试验（至少100次）并计算相对频率；随着试验次数增加，相对频率应收敛到理论概率。请明确陈述这种联系。
7. Collecting and Recording Data Systematically | 系统地收集与记录数据
A clear data collection table is essential for organising your observations. The exam may provide an incomplete table and ask you to fill in missing headings, units, or tallies. Always include a 'tally' column when counting frequencies, and a 'frequency' column for the total.
清晰的数据收集表对于整理观察结果至关重要。考题可能提供一个不完整的表格，要求你补全缺失的标题、单位或频数标记。在统计频数时，务必包含“画记”栏和“频数”栏。
r data to a graph. Columns should be labelled with the variable name and unit, for example 'Time, t (s)' ous data into intervals (e.g. 10 ≤ t < 20) can simplify presentation but must not overlap.
设计表格时应便于将数据转到图表上。各列应标注变量名和单位，例如“时间，t (秒)”。将连续数据分组到区间（如10 ≤ t < 20）可以简化展示，但区间必须互不重叠。
Never record raw data using only a calculator's memory; the examiners want to see your working. Even in a paper‑based task, show how you order the data, construct stem‑and‑leaf diagrams, or compile grouped frequency tables.
不要只用计算器存储器记录原始数据；考官希望看到你的演算过程。即使在纸笔任务中，也要展示如何对数据排序、绘制茎叶图或编制分组频数表。
8. Analysing Data from an Experiment | 分析实验数据
Once data is collected, you must calculate appropriate statistics: mean, median, mode, range, and possibly interquartile range. The mean is given by:
收集数据后，你必须计算合适的统计量：平均数、中位数、众数、极差，有时还有四分位距。平均数公式为：
Mean = (Σx) ÷ n or Mean = (Σfx) ÷ Σf for grouped data
平均数 = (Σx) ÷ n 或分组数据用 平均数 = (Σfx) ÷ Σf
The median is the middle value; for a frequency distribution use cumulative frequency. Deciding which average is most appropriate depends on the presence of outliers and the data type—use a median for skewed data.
中位数是中间值；对于频数分布使用累积频数。哪个平均数最合适取决于是否存在异常值以及数据类型——对于偏态数据应使用中位数。
Graphical representation is a key skill. You may be asked to produce or interpret bar charts (for discrete/categorical data), histograms (for continuous grouped data with frequency density), line graphs, scatter graphs, or cumulative frequency curves.
图形表示是一项关键技能。你可能需要绘制或解读条形图（用于离散/类别数据）、直方图（用于连续分组数据，注意频数密度）、折线图、散点图或累积频数曲线。
For a scatter graph, describe correlation (positive, negative, none) and draw a line of best fit by eye. The line of best fit can be used to estimate unknown values (interpolation) or predict outside the data range (extrapolation, which is less reliable).
对于散点图，描述相关性（正、负、无），并用目测画出最佳拟合线。最佳拟合线可以用来估算未知值（内插），或预测数据范围外的值（外推，不太可靠）。
9. Interpreting Results and Drawing Conclusions | 解读结果与得出结论
Conclusions must refer back to the original hypothesis. If your relative frequency after 200 trials is 0.32 and your hypothesised probability was 0.3, you should state that the experiment supports the model, but more trials would improve reliability.
结论必须回扣原始假设。如果200次试验后的相对频率是0.32，而你假设的概率是0.3，你应该说该实验支持该模型，但增加试验次数会提高可靠性。
Never claim that a small set of trials 'proves' a probability. Instead, use language such as 'the evidence suggests' or 'the results are consistent with the theoretical value'. The distinction between proof and evidence is explicitly tested.
千万不要声称少量试验就“证明”了某个概率。相反，应使用“证据表明”或“结果与理论值一致”这类表述。证明与证据的区别会被明确考察。
Comment on outliers: an anomalous point may indicate a recording error or genuine variability. You might suggest repeating the experiment or revising the measurement technique.
对异常值加以评论：异常点可能表明记录错误或真实的变动。你可以建议重复实验或改进测量技术。
10. Evaluating the Experimental Method | 评估实验方法
Every practical investigation has limitations, and the exam expects you to identify them. Typical points include sample size too small, lack of control over confounding variables, measurement imprecision, and potential observer bias.
任何实践调查都有局限性，考试期望你能指出来。典型要点包括：样本量太小、未能控制混杂变量、测量不精确，以及潜在的观察者偏差。
Suggest specific improvements: use a larger and more representative sample, repeat measurements and take an average, automate data logging to reduce human reaction‑time errors, or employ double‑blind procedures in subjective studies.
提出具体改进方法：使用更大且更具代表性的样本；重复测量并取平均值；采用自动数据记录以减少人为反应时间误差；或在主观研究中采用双盲程序。
An evaluation of a simulation must address the validity of the model. If you used a dice to simulate bus arrival times, discuss whether the real‑world distribution is truly uniform—perhaps buses are more likely to arrive during peak times.
对模拟的评估必须讨论模型效度。如果你用骰子模拟公交到达时间，请讨论现实世界的分布是否真的均匀——或许公交车更可能在高峰期到达。
Reliability relates to the reproducibility of results. You can comment that if another student repeated the experiment and got a similar outcome, the findings would be more reliable.
信度涉及结果的可重复性。你可以说，如果另一位学生重复该实验并得到相似结果，那么发现就更可信。
11. Common Mistakes and How to Avoid Them | 常见错误与避免方法
lity = number of successful outcomes ÷ total trials. It may vary between experiments; theoretical probability is fixed by the model.
一个常见错误是混淆实验概率（相对频率）与理论概率。记住：实验概率 = 成功结果数 ÷ 总试验数。它可能随实验而异；理论概率则是由模型固定的。
s failing to use appropriate class intervals when grouping continuous data. Classes like 0–5, 5–10 will cause ambiguity at the boundary (5 is counted in which class?). Use 0 ≤ x < 5, 5 ≤ x < 10 instead.
另一个错误是对连续数据分组时用了不合适的组距。像0–5、5–10这样的分组会在边界处产生歧义（5 计入哪一组？）。应使用 0 ≤ x < 5, 5 ≤ x < 10。
When drawing a line of best fit, do not join the first and last point. The line should reflect the trend of all points. Also, avoid extending the line through the origin unless there is a valid reason (no zero‑error in the context).
在画最佳拟合线时，不要直接将首尾两点相连。直线应反映所有点的趋势。同时，除非有正当理由（背景中无零点误差），避免将直线延伸通过原点。
Finally, always check that your conclusions are consistent with the data you were given. Inventing a dramatic claim that is not supported by the figures will lose marks.
最后，务必检查结论与你所给的数据是否一致。编造一个数据上并不支持的夸张说法会失分。
12. Applying Practical Skills to Unfamiliar Contexts | 将实践技能应用于陌生情境
OCR examination questions often present a novel scenario, such as a scientist testing a new fertiliser or a quality‑control manager inspecting light bulbs. You must transfer the generic experimental framework to these settings without being distracted by the narrative.
OCR考题经常给出一个新颖情境，比如科学家测试新肥料，或者质量控制经理检验灯泡。你必须将通用的实验框架迁移到这些场景中，而不被叙述分散注意力。
Identify the independent, dependent, and control variables from the text. Write a brief plan: 'I would divide the plants into two groups, apply fertiliser A to one group and no fertiliser to the other, keep light, water, and soil identical, and measure the height after 4 weeks.'
从文本中识别自变量、因变量和控制变量。写出简要计划：“我将植物分为两组，一组施肥料A，另一组不施肥，保持光照、水分和土壤相同，4周后测量高度。”
If the question asks for a simulation, adapt the allocation rule to the given probability. For example, if 12% of items are defective, use two‑digit random numbers where 00–11 represents defective and 12–99 represents non‑defective.
如果题目要求模拟，调整分配规则以匹配给定的概率。例如，若12%的产品有缺陷，就用两位随机数，00–11代表缺陷，12–99代表无缺陷。
Always end with a recommendation: based on the investigation, what should the company do next? This shows you can apply the mathematics to decision‑making, a key AO3 skill.
最后总要给出建议：基于本次调查，公司接下来该怎么做？这表明你能将数学应用于决策，这是关键的AO3技能。
Published by TutorHao | Further Maths 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