📚 Interdisciplinary Mixed Question Training for IGCSE CAIE Statistics | IGCSE CAIE 统计：跨学科综合题型训练
IGCSE CAIE Statistics questions increasingly place statistical techniques inside real-world contexts from biology, economics, geography, physics and business. To score well, you must not only calculate correctly but also interpret results in the language of the subject. This article trains you through ten key areas of mixed-question practice, using paired English and Chinese explanations, worked examples and exam-style tips.
IGCSE CAIE 统计题目越来越多地把统计方法放进生物、经济、地理、物理和商业等真实情境中。想要拿高分，你不仅要会计算，还要能用学科语言解释结果。本文通过十大关键领域的综合题型训练，提供中英对照讲解、例题和考试技巧。
1. Interdisciplinary Data Collection | 跨学科数据收集
Statistical work begins with data. In CAIE IGCSE Statistics, data can be categorical, discrete or continuous. A biology investigation may record leaf lengths as continuous data, while a business survey may count the number of items sold per day as discrete data. A geography questionnaire might collect favourite transport types as categorical data. Identifying the data type is often the first mark in a mixed question.
统计工作从数据开始。在 CAIE IGCSE 统计中，数据可以是分类数据、离散数据或连续数据。生物调查可能记录叶片长度作为连续数据，而商业调查可能统计每天售出的商品数量作为离散数据。地理问卷可能收集最喜欢的交通方式作为分类数据。识别数据类型往往是综合题的第一分。
You should also decide whether data is primary or secondary. Primary data is collected by you, such as measuring reaction times in a physics experiment. Secondary data is obtained from published sources, for example using government rainfall records for a geography project. Mixed questions may ask you to compare reliability, cost and time for these two sources.
你还需要判断数据是一手数据还是二手数据。一手数据是你自己收集的，例如在物理实验中测量反应时间。二手数据来自已发布的资料，例如在地理项目中使用政府降雨记录。综合题可能要求你比较这两种来源的可靠性、成本和时间。
- Biology: continuous leaf length, discrete number of petals, categorical flower colour.
- Economics: discrete number of workers, continuous wage, categorical industry type.
- Geography: continuous river flow rate, discrete number of floods, categorical rock type.
- Physics: continuous temperature, discrete count of radioactive decays, categorical material type.
生物：叶片长度是连续数据，花瓣数是离散数据，花色是分类数据。经济：工人数是离散数据，工资是连续数据，行业类型是分类数据。地理：河流流量是连续数据，洪水次数是离散数据，岩石类型是分类数据。物理：温度是连续数据，放射性衰变计数是离散数据，材料类型是分类数据。
2. Sampling Strategies in Real Contexts | 真实情境中的抽样策略
In real investigations, you usually cannot test every individual in a population. Sampling saves time and money, but the sample must be representative. Random sampling gives every member an equal chance of selection, reducing bias. For example, a medical trial might randomly select patients from a hospital list to test a new treatment.
在真实调查中，你通常无法测试总体中的每一个个体。抽样可以节省时间和成本，但样本必须具有代表性。随机抽样让每个成员都有相等的被选机会，从而减少偏差。例如，医学试验可能会从医院名单中随机选择患者来测试新疗法。
Stratified sampling divides the population into groups, or strata, and takes a proportional sample from each group. A school survey on homework time might stratify by year group so that Year 11 students are represented fairly. Systematic sampling selects every kth item, for instance testing every 50th light bulb on a production line. Quota sampling is often used in market research, where interviewers select a fixed number of people from different age bands.
分层抽样将总体分成若干组，即层，并从每一层中按比例抽取样本。一项关于作业时间的学校调查可以按年级分层，以确保 11 年级学生被公平地代表。系统抽样选取每第 k 个个体，例如在生产线上每 50 个灯泡检测一个。配额抽样常用于市场调查，调查员从不同年龄段中选取固定数量的人。
| Sampling method | Cross-curricular example | Advantage |
|---|---|---|
| Random | Selecting patients for a drug trial | Less selection bias |
| Stratified | School survey by year group | Represents all groups |
| Systematic | Factory quality control of every 50th bulb | Simple and quick |
| Quota | Market research by age band | Easy without a full list |
抽样方法：随机抽样 - 选择药物试验的患者 - 减少选择偏差；分层抽样 - 按年级进行学校调查 - 代表所有群体；系统抽样 - 工厂每 50 个灯泡质检 - 简单快捷；配额抽样 - 按年龄段进行市场调查 - 无需完整名单。
3. Charts and Diagrams Across Subjects | 跨学科图表与图示
Presenting data clearly is essential in mixed questions. Bar charts are for categorical or discrete data, while histograms are for continuous data. A business studies question might use a bar chart to show sales by product category, whereas a biology question might use a histogram to display the continuous distribution of heights of plants.
在综合题中，清晰地展示数据至关重要。条形图用于分类数据或离散数据，而直方图用于连续数据。商业研究题可能用条形图显示各产品类别的销售额，而生物题可能用直方图展示植物高度的连续分布。
Pie charts show proportions of a whole, useful in economics for budget shares. Cumulative frequency curves help find medians and quartiles, often used in geography to analyse river lengths. Box plots compare distributions quickly, such as comparing reaction times before and after training in a sports science experiment.
饼图显示整体中各部分的比例，在经济学中常用于预算份额。累计频率曲线有助于求中位数和四分位数，常用于地理中分析河流长度。箱线图可以快速比较分布，例如在运动科学实验中比较训练前后的反应时间。
| Chart type | Data type | Example context |
|---|---|---|
| Bar chart | Categorical or discrete | Economics: sales by product |
| Histogram | Continuous | Biology: plant heights |
| Pie chart | Categorical proportion | Economics: household budget |
| Cumulative frequency curve | Continuous frequency | Geography: river lengths |
| Box plot | Continuous or discrete | Sports science: reaction times |
图表类型：条形图 - 分类或离散数据 - 经济：产品销售额；直方图 - 连续数据 - 生物：植株高度；饼图 - 分类比例 - 经济：家庭预算；累计频率曲线 - 连续频率 - 地理：河流长度；箱线图 - 连续或离散数据 - 运动科学：反应时间。
4. Averages and Measures of Spread | 平均数与离散程度
The mean, median and mode are three measures of central tendency. The mean uses all values but is affected by outliers. The median is robust to extreme values, which makes it useful in geography when comparing river discharges that include rare floods. The mode is used for categorical data, such as the most common blood type in a biology class.
平均数、中位数和众数是三种集中趋势的度量。平均数使用所有数值，但容易受异常值影响。中位数对极端值稳健，这在地理中比较包含罕见洪水的河流流量时很有用。众数用于分类数据，例如生物课上最常见的血型。
Spread can be measured by range, interquartile range (IQR) or standard deviation. The range is quick but sensitive to outliers. The IQR covers the middle 50% of data and is often paired with the median. Standard deviation measures how far values are from the mean, making it ideal for repeated measurements in physics experiments where precision matters.
离散程度可以用极差、四分位距（IQR）或标准差来衡量。极差计算快，但对异常值敏感。四分位距覆盖中间 50% 的数据，通常与中位数一起使用。标准差衡量数值偏离平均数的程度，非常适合物理实验中重视精度的重复测量。
Mean: x̄ = Σx ÷ n | SD: σ = √[ Σ(x − x̄)² ÷ n ]
公式：平均数 x̄ = Σx ÷ n，标准差 σ = √[ Σ(x − x̄)² ÷ n ]。
In a mixed biology and statistics question, you may be given the heights of ten plants and asked which average best represents the data. If one plant is unusually tall, the median is better than the mean. If you are later asked to compare consistency, use the standard deviation or IQR rather than range alone.
在一道生物与统计的综合题中，你可能会得到十株植物的高度，并被问及哪个平均数最能代表数据。如果有一株植物异常高，中位数比平均数更好。如果之后要求比较一致性，应使用标准差或四分位距，而不是只使用极差。
5. Probability in Scientific Trials | 科学试验中的概率
Probability is used to model uncertainty in science and economics. The probability of an event is a number between 0 and 1. Experimental probability is found from relative frequency, such as the proportion of seeds that germinate in a biology trial. Theoretical probability uses equally likely outcomes, such as rolling a fair die in a mathematics investigation.
概率用于模拟科学和经济学中的不确定性。事件的概率是一个介于 0 到 1 之间的数。实验概率通过相对频率求得，例如生物试验中种子发芽的比例。理论概率基于等可能结果，例如数学调查中掷一枚公平骰子。
ou must multiply along the branches and add probabilities for mutually exclusive outcomes.
树状图有助于处理独立事件。如果一道遗传学题说豌豆植株产生黄色种子的概率为 0.75，产生绿色种子的概率为 0.25，那么两个独立后代都产生黄色种子的概率是 0.75 × 0.75 = 0.5625。你必须沿分支相乘，并把互斥结果的概率相加。
In medical testing, probability is used to interpret false positives and false negatives. A test may be 95% accurate, but the probability that a positive result is correct also depends on how common the disease is. CAIE questions often ask you to use a tree diagram or a two-way table to calculate such conditional probabilities.
在医学检测中，概率被用来解释假阳性和假阴性。一项检测可能有 95% 的准确率，但阳性结果正确的概率还取决于疾病的普遍程度。CAIE 题目经常要求你用树状图或双向表来计算这类条件概率。
6. Probability Distributions in Context | 情境中的概率分布
The binomial distribution models the number of successes in a fixed number of independent trials, each with the same probability of success. It applies to situations such as the number of defective items in a batch of ten factory products, where each item has the same chance of being defective.
二项分布模拟在固定次数的独立试验中成功的次数，每次试验的成功概率相同。它适用于例如一批十件工厂产品中次品数量的情形，其中每件产品有相同的次品概率。
P(X = r) = C(n, r) × pʳ × qⁿ⁻ʳ
公式：P(X = r) = C(n, r) × pʳ × qⁿ⁻ʳ，其中 p 是成功概率，q = 1 − p。
The normal distribution describes continuous data that clusters symmetrically around a mean, such as heights of students, masses of apples or measurement errors in a physics lab. It is bell-shaped and fully determined by its mean and standard deviation. In mixed questions, you may be given the mean and standard deviation and asked to compare two distributions, for example lengths of two species of fish in a biology study.
正态分布描述围绕平均数对称聚集的连续数据，例如学生身高、苹果质量或物理实验室中的测量误差。它呈钟形，并由均值和标准差完全确定。在综合题中，你可能会得到均值和标准差，并被要求比较两个分布，例如生物研究中两种鱼的长度。
A typical business question might say that the lifetime of a brand of batteries is normally distributed with mean 50 hours and standard deviation 5 hours. You can use the symmetry of the distribution to state that about 68% of batteries last between 45 and 55 hours. Remember that normal distribution calculations at IGCSE level usually rely on symmetry and standard deviation bands.
一道典型的商业题可能会说某品牌电池的寿命服从正态分布，平均值为 50 小时，标准差为 5 小时。你可以利用分布的对称性得出约 68% 的电池寿命在 45 到 55 小时之间。记住，IGCSE 层面的正态分布计算通常依赖对称性和标准差区间。
7. Bivariate Data and Correlation | 双变量数据与相关
Bivariate data involves two variables measured on the same individuals. A scatter diagram shows the relationship between them. If the points rise from left to right, the correlation is positive; if they fall, it is negative. A business example is advertising expenditure and sales revenue, which usually have positive correlation.
双变量数据涉及同一组个体上测量的两个变量。散点图显示它们之间的关系。如果点从左到右上升，则相关性为正；如果下降，则为负。商业中广告支出和销售收入通常呈正相关。
Correlation strength is measured by the product moment correlation coefficient, r. It ranges from −1 to 1. A geography example could show a negative correlation between altitude and air temperature: as altitude increases, temperature decreases. A value near 0 indicates no linear relationship.
相关强度用积矩相关系数 r 来衡量，取值范围为 −1 到 1。地理例子可能显示海拔与气温之间呈负相关：海拔升高，气温降低。接近 0 的值表示没有线性关系。
r = Sxy ÷ √Sxx × Syy
公式：r = Sxy ÷ √Sxx × Syy，其中 Sxy、Sxx、Syy 是离差平方和。
Correlation does not imply causation. If ice cream sales and drowning incidents both rise in summer, they are correlated because a third variable, temperature, affects both. In exam mixed questions, you must state clearly that a high correlation alone does not prove one variable causes the other.
相关不意味着因果。如果冰淇淋销量和溺水事件在夏季都上升，它们相关是因为第三个变量——气温——同时影响两者。在考试综合题中，你必须清楚地说明高相关本身并不能证明一个变量导致另一个变量。
8. Regression and Prediction | 回归与预测
+ bx, where b is the gradient and a is the y-intercept. The least squares regression line minimises the sum of the squares of the vertical distances from the points to the line.
回归线穿过散点图，用来模拟两个变量之间的线性关系。方程通常写作 y = a + bx，其中 b 是斜率，a 是 y 截距。最小二乘回归线使各点到直线的垂直距离平方和最小。
b = Sxy ÷ Sxx | a = ȳ − b x̄
公式：b = Sxy ÷ Sxx，a = ȳ − b x̄。
A physics experiment might measure the extension of a spring for different masses. If you plot extension against mass and find the regression line, you can predict the extension for a new mass inside the data range. This is called interpolation and is generally reliable. Predicting outside the range is extrapolation, which is risky because the linear trend may not continue.
物理实验可能测量不同质量下弹簧的伸长量。如果你绘制伸长量对质量的散点图并求出回归线，就可以预测数据范围内一个新质量对应的伸长量。这叫做内插，通常比较可靠。在数据范围之外进行预测叫做外推，这有风险，因为线性趋势可能不会延续。
Published by TutorHao | IGCSE 统计 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