📚 GCSE Cambridge Statistics: Core Knowledge Summary | GCSE剑桥统计：核心知识点梳理
This article brings together the essential topics covered in the Cambridge GCSE Statistics syllabus, providing a clear, bilingual recap of key concepts, techniques, and formulae. From data types and sampling to probability distributions and regression, each section is designed to reinforce understanding and support exam preparation.
本文梳理了剑桥 GCSE 统计学考纲中的核心内容，以清晰的中英对照方式归纳关键概念、方法和公式。从数据类型与抽样到概率分布与回归分析，每一节都旨在巩固理解，助力备考。
1. Types of Data and Collection Methods | 数据类型与收集方法
Data is usually classified as categorical (qualitative) or numerical (quantitative). Categorical data describe qualities or labels, while numerical data represent measurable quantities.
数据通常分为分类（定性）数据和数值（定量）数据。分类数据描述属性或标签，数值数据表示可测量的数量。
Numerical data can be discrete, taking only specific values (e.g. number of students), or continuous, taking any value within a range (e.g. height, time).
数值数据可以是离散的，仅取特定值（例如学生人数），也可以是连续的，可在某个范围内取任意值（例如身高、时间）。
Data can be collected through primary methods such as experiments, surveys, or observations, or through secondary sources like published reports and databases.
数据可以通过一手方法收集，如实验、调查或观察，也可以通过二手来源收集，如已发布的报告和数据库。
When designing a questionnaire, it is important to avoid leading questions, ensure clarity, and pilot the survey to reduce response bias.
设计问卷时，必须避免诱导性问题、确保表述清晰，并进行试调查以减少回答偏差。
2. Sampling Techniques | 抽样技术
Random sampling gives every member of the population an equal chance of being selected, reducing selection bias. Simple random sampling is the most basic form.
随机抽样使总体中每个成员被抽中的机会相等，从而减少选择偏差。简单随机抽样是最基本的形式。
Stratified sampling divides the population into distinct groups (strata) and selects a random sample from each, proportional to its size. This ensures representation of key subgroups.
分层抽样先将总体划分为不同的组（层），然后从每一层按比例随机抽取样本，保证关键子群的代表性。
Systematic sampling selects every k-th member from a list after a random start, while cluster sampling randomly selects entire groups when a full list of individuals is unavailable.
系统抽样在随机起始点后名单上每隔 k 个抽取一个成员；整群抽样则在无法获得个体完整名单时随机抽取整群。
Convenience sampling and quota sampling are non-probability methods that are quicker but may introduce significant bias and should be evaluated critically.
便利抽样和配额抽样是非概率方法，速度较快但可能引入明显偏差，需要审慎评估。
3. Tabulation and Frequency Distributions | 表格与频数分布
Raw data is often organised into frequency tables, which list each value or class interval alongside its frequency. These tables are the foundation for further calculations.
原始数据常被整理成频数表，列出每个值或组距及其频数。这些表格是后续计算的基础。
For grouped data, classes should be of equal width where possible and clearly defined with boundaries to avoid ambiguity. Cumulative frequency adds frequencies cumulatively, helping to find medians and percentiles.
对于分组数据，组距应尽可能相等，并用明确的界限定义以避免歧义。累积频数将频数累计相加，有助于求中位数和百分位数。
Two-way tables or contingency tables display relationships between two categorical variables and are used in probability and the chi-squared test at more advanced levels.
双向表或称列联表显示两个分类变量之间的关系，在概率以及更高层次的卡方检验中都会用到。
4. Graphical Representation of Data | 数据的图形表示
Bar charts represent categorical or discrete data with equal-width bars separated by gaps; the height shows frequency or relative frequency.
条形图用等宽且有间隔的长条表示分类或离散数据，长条的高度代表频数或相对频率。
Pie charts display proportions of a whole, with each sector’s angle proportional to the frequency. A full circle represents 360° and 100% of the data.
饼图展示整体中各部分的比例，每个扇形的圆心角与频数成正比。整个圆代表 360° 和数据的 100%。
Histograms are used for continuous grouped data, where the area of each bar is proportional to frequency; frequency density = frequency / class width must be used when class widths vary.
直方图用于连续的分组数据，每个矩形的面积与频数成正比；当组距不等时必须使用频数密度 = 频数 / 组距。
Cumulative frequency curves (ogives) and box-and-whisker plots summarise the median, quartiles, and range, making it easy to compare distributions visually.
累积频数曲线（肩形图）和箱线图可概括中位数、四分位数和极差，便于直观地比较分布。
5. Measures of Central Tendency | 集中趋势的度量
The mean is the arithmetic average. For raw data, mean = Σx / n. For grouped data, mean ≈ Σfx / Σf, where x is the class midpoint.
均值是算术平均数。对于原始数据，均值 = Σx / n；对于分组数据，均值 ≈ Σfx / Σf，其中 x 为组中值。
Mean (grouped) = Σfx / Σf
The median is the middle value when data is ordered. For n values, the position is (n+1)/2. In grouped data, linear interpolation using cumulative frequency is required.
中位数是排序后位于中间的值。对于 n 个数值，位置为 (n+1)/2。在分组数据中，需要使用累积频数进行线性插值。
The mode is the most frequently occurring value. A dataset can have one mode (unimodal), two modes (bimodal), or more, and for grouped data it is estimated from the modal class.
众数是出现次数最多的值。数据集可以有一个众数（单峰）、两个众数（双峰）或更多；对于分组数据，众数由众数组估计得出。
In skewed distributions, the mean, median, and mode differ: in a positively skewed distribution, mean > median > mode; in a negative skew, the reverse holds.
在偏态分布中，均值、中位数和众数不同：正偏态时均值最大，负偏态时相反。
6. Measures of Spread: Range, IQR, Variance, Standard Deviation | 离散程度的度量
The range is simply the difference between the largest and smallest values. It is easily affected by outliers.
极差是最大值与最小值之差，易受异常值影响。
The interquartile range (IQR) = upper quartile (Q3) – lower quartile (Q1), measuring the spread of the middle 50% and is more robust against outliers.
四分位距 (IQR) = 上四分位数 (Q3) – 下四分位数 (Q1)，衡量中间 50% 数据的离散程度，对异常值更具稳健性。
Variance and standard deviation are the most important measures of spread. For a population, variance σ² = Σ(x – μ)² / N; standard deviation σ = √variance.
方差和标准差是最重要的离散程度度量。对于总体，方差 σ² = Σ(x – μ)² / N；标准差 σ = √方差。
σ² = Σ(x – μ)² / N
When dealing with a sample, the divisor (n – 1) is used to give an unbiased estimate. Spread measures are essential for comparing data reliability.
处理样本时，分母用 (n – 1) 以得到无偏估计。离散度量对于比较数据的可靠性至关重要。
7. Probability Fundamentals and Tree Diagrams | 概率基础与树形图
Probability measures the likelihood of an event, with P(event) = number of favourable outcomes / total number of equally likely outcomes, provided outcomes are equally probable.
概率衡量事件发生的可能性，P(事件) = 有利结果数 / 所有等可能结果的总数，前提是各结果等可能。
The complement rule states that P(not A) = 1 – P(A). Two events are mutually exclusive if they cannot occur together; then P(A or B) = P(A) + P(B).
互补规则指出 P(非A) = 1 – P(A)。若两事件不能同时发生，则互斥，此时 P(A 或 B) = P(A) + P(B)。
For independent events, P(A and B) = P(A) × P(B). Conditional probability is given by P(A|B) = P(A and B) / P(B) and is essential for dependent events.
对于独立事件，P(A 且 B) = P(A) × P(B)。条件概率表示为 P(A|B) = P(A 且 B) / P(B)，对于非独立事件非常重要。
Tree diagrams help organise probabilities for multi-stage events by multiplying along branches and adding across outcomes; they show all possible sequences clearly.
树形图通过沿着分支相乘、各结果相加来组织多阶段事件的概率，清晰展示所有可能的序列。
8. The Binomial Distribution | 二项分布
The binomial distribution models the number of successes in a fixed number of independent trials, each with the same probability of success p.
二项分布描述在固定次数的独立试验中成功的次数，每次试验成功的概率 p 相同。
If X ~ B(n, p), the probability of exactly r successes is given by the formula:
若 X ~ B(n, p)，恰好有 r 次成功的概率由以下公式给出：
P(X = r) = nCr pr (1 – p)n – r
The mean (expected value) of a binomial distribution is np, and the variance is np(1 – p). The shape is roughly symmetric when p ≈ 0.5 and skewed otherwise.
二项分布的均值（期望值）为 np，方差为 np(1 – p)。当 p ≈ 0.5 时分布大致对称，否则呈偏态。
Binomial probabilities can be calculated using the formula, Pascal’s triangle for small n, or cumulative distribution tables provided in examinations.
二项概率可利用公式、小 n 值的帕斯卡三角形或考试提供的累积分布表来计算。
9. The Normal Distribution and Standardisation | 正态分布与标准化
The normal distribution is a bell-shaped, continuous probability distribution characterised by its mean μ and standard deviation σ.
正态分布是一种钟形的连续概率分布，由均值 μ 和标准差 σ 决定其位置与形状。
About 68% of values lie within one standard deviation of the mean, 95% within two, and 99.7% within three. This empirical rule is crucial for estimation.
大约 68% 的数据落在均值上下一个标准差内，95% 落在两个标准差内，99.7% 落在三个标准差内。这一经验法则对估计很重要。
To compare scores from different normal distributions, we standardise using the z-score:
为了比较不同正态分布的得分，我们使用 z 分数进行标准化：
z = (x – μ) / σ
Standard normal distribution tables (with μ = 0, σ = 1) then give probabilities for any range. The total area under the curve is 1.
标准正态分布表（μ = 0, σ = 1）可提供任意区间的概率。曲线下的总面积为 1。
10. Scatter Diagrams and Correlation | 散点图与相关性
A scatter diagram plots bivariate data to visualise the relationship between two numerical variables. Positive correlation means both variables increase together; negative correlation means one increases while the other decreases.
散点图绘制二元数据，以观察两个数值变量之间的关系。正相关表示两者同时增加；负相关表示一个增加而另一个减少。
Correlation is not causation. A strong correlation may be due to a hidden third factor, so conclusions must be drawn carefully.
相关关系不等于因果关系。强相关可能源于隐藏的第三因素，因此下结论时须谨慎。
Pearson’s product-moment correlation coefficient r measures the strength and direction of a linear relationship, ranging from –1 (perfect negative) to +1 (perfect positive). r ≈ 0 indicates no linear correlation.
皮尔逊积矩相关系数 r 度量线性关系的强度和方向，取值范围为 –1（完全负相关）到 +1（完全正相关）。r ≈ 0 表示无线性相关。
The line of best fit can be drawn ‘by eye’ through the middle of the scatter points, or calculated using regression analysis for more precision.
最佳拟合线可以通过观察散点图“凭眼力”绘制，也可以通过回归分析精确计算。
11. Linear Regression | 线性回归
Linear regression finds the equation of the straight line that best models the relationship between an independent variable x and a dependent variable y.
线性回归得出最能描述自变量 x 与因变量 y 之间关系的直线方程。
The regression line is usually written as y = a + bx, where b is the gradient and a the y-intercept. Using the least squares method:
回归线通常写作 y = a + bx，其中 b 是斜率，a 是截距。采用最小二乘法：
b = Σ(x – x̄)(y – ȳ) / Σ(x – x̄)²
a = ȳ – b x̄
Once the equation is determined, it can be used for interpolation (within the data range) but extrapolation beyond the data can be unreliable.
一旦确定方程，便可用于插值（在数据范围内），但超出数据范围的外推可能不可靠。
The regression line passes through the mean point (x̄, ȳ). Residuals, the differences between observed and predicted y-values, help evaluate the model’s fit.
回归线经过均值点 (x̄, ȳ)。残差（观测值与预测值之差）有助于评估模型的拟合优度。
12. Time Series Analysis and Moving Averages | 时间序列分析与移动平均
A time series is a set of observations recorded in time order, such as monthly sales figures or daily temperatures. It often shows trend, seasonal variation, and irregular fluctuations.
时间序列是按时间顺序记录的一组观测值，如月度销售额或每日气温，通常呈现出趋势、季节变动和不规则波动。
Moving averages smooth out short-term irregularities to reveal the underlying trend. A k-point moving average replaces each value with the mean of k consecutive terms.
移动平均可平滑短期不规则波动，从而揭示基本趋势。k 点移动平均用连续 k 项的均值替换每个原始值。
To deseasonalise data, seasonal components are estimated and removed, making it easier to compare figures from different time periods and to forecast future values.
为了剔除季节性，估计并移除季节成分，这样更容易比较不同时期的数值并进行预测。
Plotting trends and seasonal components on the same graph helps in analysing the overall pattern and in making business or economic decisions based on historical data.
将趋势和季节成分绘制在同一张图上，有助于分析整体模式，并根据历史数据做出商业或经济决策。
Published by TutorHao | Statistics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导