📚 Year 11 Cambridge Statistics: Interdisciplinary Integrated Question Practice | 跨学科综合题型训练
Statistics is not an isolated subject; it is a universal language used to make informed decisions across science, business, social research and beyond. In Cambridge IGCSE Statistics, cross-curricular questions test your ability to interpret real-world data, apply probability models in unfamiliar contexts and justify your conclusions using statistical reasoning. This article provides a structured revision workout, blending topics from biology, economics, psychology and environmental science with key statistical techniques you need to master.
统计学不是孤立的学科，它是一门通用语言，广泛应用于科学、商业、社会研究等领域，帮助我们做出明智的决策。在剑桥 IGCSE 统计考试中，跨学科题型要求你解读真实数据、在不熟悉的情境中应用概率模型，并用统计推理来证明你的结论。本文提供了一份系统的复习训练，把生物学、经济学、心理学和环境科学中的话题与你必须掌握的重要统计方法结合起来。
1. Interdisciplinary Data Interpretation | 跨学科数据解读
Many exam questions begin with a table or graph from a subject like geography or health science. You must extract information, identify misleading representations and suggest improvements.
很多考题从地理或健康科学等学科中提供一张表格或图表。你必须提取信息，识别误导性的呈现方式并提出改进建议。
For example, a biologist records the body mass (kg) of 40 adult penguins from two islands: Island A and Island B. The data are summarised in a back-to-back stem-and-leaf diagram. You might be asked to compare the distribution of masses, stating which island has a higher median and greater interquartile range.
例如，一位生物学家记录了两个岛屿（A岛和B岛）40只成年企鹅的体重（千克）。数据用背靠背茎叶图汇总。你可能会被要求比较体重的分布，说明哪个岛屿的中位数更高、四分位距更大。
Always link your statistical comment to the real-world context: 'Penguins on Island B are typically heavier, which could be due to richer food sources,' rather than just stating numbers.
始终将统计评论与现实背景联系起来：“B岛的企鹅通常更重，这可能是因为食物来源更丰富”，而不仅仅是列出数字。
2. Probability Models in Scientific Experiments | 科学实验中的概率模型
In chemistry or physics, experimental outcomes often follow a known probability distribution. Understanding when to apply the binomial or normal model is a key skill.
在化学或物理实验中，实验结果常遵循已知的概率分布。懂得何时应用二项分布或正态模型是一项关键技能。
A laboratory test for a disease has a 95% ly 8 tests give the correct result. This is a binomial situation with n = 10 and p = 0.95.
一项疾病实验室检测的准确率为95%。如果检测10个随机样本，计算恰好有8个检测结果正确的概率。这是一个二项分布问题，n = 10，p = 0.95。
P(X = 8) = ₁₀C₈ × (0.95)⁸ × (0.05)² ≈ 0.0746
Not only do you perform the calculation, you also need to explain: 'There is about a 7.5% chance that exactly 8 out of 10 tests are correct, which is fairly low despite the high accuracy rate.'
你不仅要计算，还需解释：“尽管准确率很高，恰好8次正确的结果也只有大约7.5%的概率，这是相当低的。”这种联系实验背景的解释是得分的关键。
3. Statistical Charts and Business Decisions | 统计图表与商业决策
Business studies questions often supply sales data over time. You need to construct, interpret and critically evaluate charts such as compound bar charts, percentage bar charts, pie charts and line graphs.
商业研究类题目常提供随时间变化的销售数据。你需要绘制、解读并批判性地评估复合条形图、百分比条形图、饼图和折线图等。
A store records its quarterly revenue for two years in four product categories. A component bar chart can show how total revenue changes and how the product mix contributes. When asked 'In which quarter did electronics account for the largest proportion of revenue?' you should calculate percentages, not rely on the absolute height of the segment.
一家商店记录了两年来四个产品类别的季度收入。复合条形图可以显示总收入的变化及各产品类别的贡献。当被问到“电子产品在哪一季度占收入比例最大？”时，你应该计算百分比，而不是仅凭分段的绝对高度判断。
Be ready to discuss why a particular chart is misleading. For example, a truncated vertical axis can exaggerate small differences; a 3D pie chart can distort proportions. Business decisions based on such graphs could lead to incorrect resource allocation.
准备好讨论为什么某个图表具有误导性。例如，截断的纵轴会夸大微小差异；3D饼图会扭曲比例。基于这类图表的商业决策可能导致资源分配错误。
4. Correlation and Causation in Social Sciences | 社会科学中的相关与因果
Sociology or psychology investigations often look for relationships between variables, such as hours of screen time and well-being score. Calculating Pearson's product-moment correlation coefficient r, or Spearman's rank correlation coefficient, is common. But the most important skill is distinguishing correlation from causation.
社会学或心理学调查常寻找变量间的关系，例如屏幕使用时间与幸福感得分。计算皮尔逊积矩相关系数 r 或斯皮尔曼等级相关系数是常见的。但最重要的技能是区分相关与因果关系。
al media and self-reported concentration level. You would conclude a strong negative correlation. However, you cannot say 'social media causes poor concentration' because there could be confounding variables, such as sleep quality or stress.
例如，12名学生的数据显示社交媒体使用时间与自报注意力水平之间的 r = -0.81。你会得出强负相关的结论，但不能说“社交媒体导致注意力下降”，因为可能存在混杂变量，如睡眠质量或压力。
Always write, 'There is a significant linear relationship' or 'As A increases, B tends to decrease,' but avoid causal language unless the experiment was controlled.
始终要写“存在显著的线性关系”或“随着A增加，B趋于下降”，但除非实验是受控的，避免使用因果性语言。这是跨学科题中常考的推理严谨性。
5. Normal Distribution in Psychological Measurement | 正态分布在心理学测量中的应用
Many psychological traits, such as IQ scores, reaction times or anxiety levels, are modelled by a normal distribution. You must be able to standardise, use normal tables, and find probabilities and percentiles.
许多心理特征，如智商分数、反应时间或焦虑水平，都可用正态分布建模。你必须能够标准化，使用正态分布表，并求出概率和百分位数。
An IQ test has a mean of 100 and a standard deviation of 15. Find the percentage of the population with an IQ between 85 and 115. Using z-scores:
z₁ = (85 - 100) ÷ 15 = -1.00, z₂ = (115 - 100) ÷ 15 = 1.00
From tables, P(-1 < Z < 1) = 0.6826, so about 68.3% of people fall in this range. You could then be asked: 'A psychologist wants to identify the top 5% of creative problem-solvers. What IQ score is the threshold?' This is the 95th percentile, approximately 100 + 1.645×15 ≈ 124.7.
从表中，P(-1 < Z < 1) = 0.6826，因此约68.3%的人落在此范围内。你可能会接着被问到：“一位心理学家想识别出创造力解题能力排名前5%的人，IQ分数的阈值是多少？”这就是第95百分位数，约100 + 1.645×15 ≈ 124.7。这类题目要求你灵活应用标准化技术，并结合心理学情境解释结果。
6. Hypothesis Testing in Medical Trials | 假设检验在医学试验中的应用
A pharmaceutical company claims its new drug lowers blood pressure by at least 8 mmHg on average. You carry out a hypothesis test at the 5% significance level using a sample of 30 patients. This typical medical statistics scenario requires a structured approach: state hypotheses, choose test (one-sample t-test or z-test if population standard deviation is known), calculate test statistic, compare with critical value, and conclude in context.
一家制药公司声称其新药平均可将血压降低至少8 mmHg。你在5%的显著性水平下，用30名患者样本进行假设检验。这个典型的医学统计情境需要结构化方法：陈述假设，选择检验（若总体标准差已知用单样本z检验，否则用t检验），计算检验统计量，与临界值比较，并在情境下得出结论。
ation 2.1 mmHg. Test H₀: μ ≥ 8 against H₁: μ < 8. The test statistic t = (7.2 - 8) / (2.1÷√30) ≈ -2.09. For 29 degrees of freedom, the critical t-value (one-tailed, 5% . There is sufficient evidence that the mean reduction is less than 8 mmHg, so the company's claim is not supported.
假设样本平均降压7.2 mmHg，样本标准差2.1 mmHg。检验 H₀: μ ≥ 8 对 H₁: μ < 8。检验统计量 t = (7.2 - 8) / (2.1÷√30) ≈ -2.09。自由度为29，单侧5%临界t值约为-1.699。因为-2.09 < -1.699，拒绝H₀。有充分证据表明平均降压小于8 mmHg，因此不支持公司的声明。解释时务必使用医学情境：“该药物可能无法达到宣称的降压效果，需进一步试验。”
7. Sampling Methods in Market Research | 抽样方法与市场研究
Business and sociology exam questions frequently present a market research scenario, asking you to critique the sampling method and suggest improvements. Whether it is quota, stratified, systematic, or simple random sampling, you need to describe the process and evaluate bias and representativeness.
商业和社会学考题常给出一个市场研究情景，要求你评论抽样方法并提出改进建议。无论是配额抽样、分层抽样、系统抽样还是简单随机抽样，你需要描述过程并评估偏差和代表性。
For instance: 'A company wants to survey customer satisfaction among its three store types: city centre, suburban mall, and online. It interviews the first 50 customers leaving each location on a Monday morning.' This is quota sampling (by location) with convenience elements. Criticise: the time of day and day of week might bias the sample; not all customer types are represented. Suggest: stratified random sampling with proportional allocation from a full customer list, ensuring each subgroup is randomly selected.
例如：“一家公司想调查其三种门店类型（市中心、郊区商场和线上）的顾客满意度。它在周一上午采访了每个地点最早离开的50名顾客。”这是带有便利性元素的配额抽样（按地点）。批评：时段和星期几可能造成样本偏差；并非所有顾客类型都得到代表。建议：采用按顾客名册的分层随机抽样，按比例分配，并确保每个子组随机选择。这类题目考查你对抽样原理的深层理解。
8. Time Series Forecasting and Economic Indicators | 时间序列预测与经济指标
Economics relies heavily on time series data, such as monthly unemployment rates, inflation indices or GDP growth. You should be able to plot moving averages, identify seasonal variation and use a trend line to make predictions. A four-point moving average is commonly used for quarterly data.
经济学高度依赖时间序列数据，如月度失业率、通胀指数或GDP增长。你应该能够绘制移动平均线，识别季节性变化，并用趋势线进行预测。对于季度数据，通常使用四点移动平均。
Given quarterly sales figures: Q1: 120, Q2: 200, Q3: 150, Q4: 90 (in £ . The next one drops Q1 and adds next Q1. Using centred moving averages, you can then calculate seasonal effects: e.g., Q2 actual minus moving average. Forecasting involves projecting the trend and adding back the average seasonal component.
给定季度销售额：Q1: 120, Q2: 200, Q3: 150, Q4: 90（单位：千英镑），第一个四点移动平均为(120+200+150+90)/4 = 140。下一个移动平均去掉Q1加上下一个Q1。使用中心化移动平均后，可计算季节效应：例如Q2实际值减去移动平均值。预测包括延伸趋势并加上平均季节成分。
Interpret forecasts cautiously: 'Assuming the current trend and seasonal pattern continue, Q2 next year is forecast to be £215,000, but external economic shocks could affect this.'
谨慎解读预测值：“假设当前趋势和季节模式持续，明年Q2的预测值为215,000英镑，但外部经济冲击可能影响此结果。”这种限定性表述展示了高级的统计思维。
9. Probability Tree Diagrams in Genetics | 遗传学中的概率树图
Genetics is a classic interdisciplinary field for probability. Using Punnett squares and tree diagrams, you can determine the probability of certain genotypes and phenotypes. A straightforward example: two heterozygous parents (Bb) for a recessive condition. Draw a probability tree to show the possible genotypes of their offspring and calculate the probability that a child is a carrier but not affected.
遗传学是概率的经典跨学科领域。使用庞纳特方格和树图，可以确定特定基因型和表现型的概率。一个简单例子：两个杂合子父母（Bb）携带一个隐性遗传病。画出概率树图显示后代可能的基因型，并计算孩子是携带者但不患病的概率。
bB (0.25), bb (0.25). A carrier not affected would be Bb or bB, which is 0.25+0.25 = 0.5.
树图：母亲的等位基因（B或b，各0.5），然后父亲的（B或b，各0.5）。结果：BB (0.25), Bb (0.25), bB (0.25), bb (0.25)。携带者但不患病的是 Bb 或 bB，概率为0.25+0.25=0.5。你可以进一步计算条件概率：“已知孩子不患病，他是携带者的概率是多少？”（答案：0.5 / 0.75 = 2/3）。这类问题将概率基本法则与生物学背景无缝结合。
10. Integrated Case Practice: Regression Analysis in Environmental Science | 综合案例练习：环境科学中的回归分析
e particulate matter PM₂.₅ concentration (y) in the air. Data from 10 parks produce a regression line y = 35.2 - 1.85x, with a correlation coefficient r = -0.93.
作为最后一道训练，考虑一项环境研究：调查城市公园中的树木数量 (x) 与空气中平均细颗粒物 PM₂.₅ 浓度 (y) 之间的关系。来自10个公园的数据产生回归线 y = 35.2 - 1.85x，相关系数 r = -0.93。
Interpret the slope: each additional tree is associated with a decrease of 1.85 µ sing the equation, predict the PM₂.₅ level for a park with 12 trees: 35.2 - 1.85×12 = 13.0 µ possible. This shows the model is only valid within the range of the data (interpolation).
解释斜率：每增加一棵树，PM₂.₅ 平均下降1.85 µg/m³。使用该方程，预测一个拥有12棵树的公园的PM₂.₅水平：35.2 - 1.85×12 = 13.0 µg/m³。但也要讨论外推的危险：如果一个公园有25棵树，预测值将为负，这是不可能的。这表明模型仅在数据范围内有效（内插）。
Evaluate the reliability: r² = 0.8649, so about 86.5% ection may also affect pollution. A city council should not simply plant more trees to solve air quality issues without considering these variables. This holistic, cross-curricular evaluation is exactly what top-grade answers require.
评估可靠性：r² = 0.8649，因此 PM₂.₅ 中约86.5%的变异由树木数量解释。然而，相关不等于因果；车流量和风向等其他因素也可能影响污染。市议会不应仅仅通过种植更多树木来解决空气质量问题，还需考虑这些变量。这种整体的、跨学科的评价正是高分答案所要求的。
11. Exam Technique for Cross-Curricular Questions | 跨学科题的考试技巧
To excel in Integrated questions, always read the context carefully. Underline the statistical task hidden in the scenario. Is it asking for a measure of central tendency, a probability, a test decision, or an evaluation of a sampling method? Then structure your answer:
要在综合题中取得优异成绩，一定要仔细阅读情境。标出隐藏于情境中的统计任务。它是要求计算集中趋势量数、概率、检验决策，还是评价抽样方法？然后构建你的答案：
- Step 1: Identify – Name the statistical technique needed. / 步骤一：识别 – 说出所需的统计方法名称。
- Step 2: Calculate/Graph – Show all workings, label axes, use proper notation. / 步骤二：计算或绘图 – 展示所有过程，标注坐标轴，使用正确的符号。
- Step 3: Interpret in context – 'This means that...' using units and the subject's language. / 步骤三：在情境中解释 – “这意味着...” 使用单位和学科语言。
- Step 4: Evaluate/Limitations – Mention assumptions, potential bias, or confounding variables. / 步骤四：评估/局限性 – 提及假设、潜在偏差或混杂变量。
Remember, Cambridge examiners value the quality of statistical communication as much as numerical accuracy. Practice writing concise, context-rich sentences. / 请记住，剑桥考官重视统计交流的质量，不亚于数值的准确性。练习写出简洁且富含情境的句子。
Published by TutorHao | Statistics 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