📚 Maths Stats Key Concepts Explained | 数学统计知识点精讲
Statistics is a branch of mathematics that deals with collecting, analysing, interpreting, and presenting data. In A‑Level Mathematics, the statistics component builds foundational skills for data handling, probability modelling, and making informed decisions under uncertainty. This article walks you through all major statistical topics, pairing every English explanation with its Chinese counterpart so you can master both the concepts and the bilingual terminology.
统计是数学的一个分支，涉及数据的收集、分析、解释和呈现。在 A‑Level 数学中，统计部分培养数据处理、概率建模及在不确定性下做出合理决策的基础能力。本文带你梳理所有主要统计主题，每段英文说明都配有对应的中文解释，助你同时掌握概念和双语术语。
1. Introduction to Statistics | 统计导论
Statistics provides tools to turn raw data into meaningful information. It is divided into descriptive statistics (summarising data) and inferential statistics (drawing conclusions from samples). Understanding the difference between a population and a sample is the first step to designing any statistical study.
统计学提供将原始数据转化为有意义信息的工具。它分为描述性统计（概括数据）和推断性统计（从样本得出结论）。理解总体与样本的区别是设计任何统计研究的第一步。
A population includes every member of the group we study, while a sample is a subset selected from it. Census studies the whole population; a sample survey studies only part of it. Random sampling methods such as simple random sampling, stratified sampling, and cluster sampling help avoid bias.
总体包含研究对象的每一个成员，而样本是从中选出的一个子集。普查研究整个总体；抽样调查只研究其中一部分。简单随机抽样、分层抽样、整群抽样等随机抽样方法有助于避免偏差。
Parameters are numerical summaries of a population (e.g. population mean μ), while statistics are numerical summaries of a sample (e.g. sample mean x̄). The aim of inferential statistics is to estimate parameters using sample statistics and to quantify the uncertainty involved.
参数是总体的数字概括（如总体均值 μ），而统计量是样本的数字概括（如样本均值 x̄）。推断统计的目标是利用样本统计量估计参数，并量化所涉及的不确定性。
2. Types of Data | 数据类型
Data can be classified as qualitative (categorical) or quantitative (numerical). Qualitative data describe attributes and can be nominal (no natural order, e.g. colours) or ordinal (ordered categories, e.g. satisfaction ratings). Quantitative data are numbers and can be discrete (countable, e.g. number of students) or continuous (measurable, e.g. height).
数据可分为定性（分类）数据或定量（数值）数据。定性数据描述属性，可以是名义的（无自然顺序，如颜色）或有序的（有序类别，如满意度评分）。定量数据是数字，可以是离散的（可数的，如学生人数）或连续的（可测量的，如身高）。
Another important distinction is between primary and secondary data. Primary data are collected first-hand by the researcher; secondary data already exist and are reused. Both have strengths and limitations in terms of cost, time, and reliability.
另一个重要区分是原始数据与二手数据。原始数据由研究者亲自收集；二手数据已经存在并被再利用。两者在成本、时间和可靠性方面各有优缺点。
Grouped data arise when raw values are placed into intervals (class intervals). When working with grouped data, we use midpoints of intervals to estimate summary statistics. The choice of class width can affect the shape of a histogram.
当原始数值被归入区间（组距）时就产生了分组数据。处理分组数据时，我们使用区间的中点来估计汇总统计量。组距宽度的选择会影响直方图的形状。
3. Measures of Central Tendency | 集中趋势的度量
The three main measures of central tendency are the mean, median, and mode. The mean (x̄) is the arithmetic average: sum of all values divided by the number of values. For grouped data, we use Σfx/Σf where f is frequency and x is the midpoint. The median is the middle value when data are ordered; for n observations, it is the (n+1)/2 th value. The mode is the most frequently occurring value.
集中趋势的三个主要度量是均值、中位数和众数。均值（x̄）是算术平均数：所有值的总和除以值的个数。对于分组数据，我们用 Σfx/Σf 计算，其中 f 是频数，x 是组中点。中位数是数据按顺序排列时的中间值；对于 n 个观测值，它是第 (n+1)/2 个值。众数是出现最频繁的值。
Each measure has pros and cons. The mean uses all data but is sensitive to outliers. The median is robust to outliers but ignores the actual values of most data. The mode is useful for categorical data but may not be unique. In a symmetrical distribution, mean = median = mode; in a skewed distribution, they differ.
每种度量都有优缺点。均值利用所有数据但对离群值敏感。中位数对离群值稳健但忽略大多数数据的实际值。众数对分类数据有用但可能不唯一。在对称分布中，均值 = 中位数 = 众数；在偏态分布中，三者不同。
For grouped data, the modal class is the class interval with the highest frequency, which we estimate using interpolation to find the mode more precisely.
对于分组数据，众数类是频数最高的组距区间，我们通常用插值法更精确地估计众数。
4. Measures of Dispersion | 离散程度的度量
Measures of dispersion describe the spread of data. The range is the simplest: maximum – minimum. The interquartile range (IQR) = Q3 – Q1, which covers the middle 50% of data and is resistant to outliers. Quartiles are found by splitting the ordered data into four equal parts.
离散程度度量描述数据的分散情况。极差是最简单的：最大值 − 最小值。四分位距 (IQR) = Q3 − Q1，涵盖中间 50% 的数据且对离群值稳健。四分位数是通过将有序数据分成四等份得到的。
The variance and standard deviation are the most important measures. For a sample: variance s² = Σ(x − x̄)² / (n − 1); standard deviation s = √s². For a population, we divide by n instead. Variance is measured in squared units, so standard deviation restores the original units and is easier to interpret.
方差和标准差是最重要的度量。对于样本：方差 s² = Σ(x − x̄)² / (n − 1)；标准差 s = √s²。对于总体，分母用 n。方差以平方单位度量，因此标准差恢复原始单位且更容易解释。
When comparing datasets, coefficient of variation (CV = s / x̄ × 100%) allows comparisons of relative variability irrespective of units. Outliers can be identified using methods such as the 1.5×IQR rule or by considering values more than 2 standard deviations from the mean.
比较数据集时，变异系数（CV = s / x̄ × 100%）可以在不考虑单位的情况下比较相对变异性。使用 1.5×IQR 规则或考虑距离均值超过 2 个标准差的值可以识别离群值。
5. Probability Basics | 概率基础
Probability quantifies the likelihood of an event, ranging from 0 (impossible) to 1 (certain). For a finite sample space with equally likely outcomes, P(A) = number of favourable outcomes / total number of outcomes. The complement rule states P(A’) = 1 − P(A). Addition rule: P(A ∪ B) = P(A) + P(B) − P(A ∩ B). For mutually exclusive events, P(A ∩ B) = 0.
概率量化事件发生的可能性，范围从 0（不可能）到 1（必然）。对于等可能结果的有限样本空间，P(A) = 有利结果数 / 总结果数。补集规则：P(A’) = 1 − P(A)。加法规则：P(A ∪ B) = P(A) + P(B) − P(A ∩ B)。对于互斥事件，P(A ∩ B) = 0。
Conditional probability P(A|B) = P(A ∩ B) / P(B) is the probability of A given B has occurred. Events A and B are independent if P(A ∩ B) = P(A) × P(B) or equivalently P(A|B) = P(A). Tree diagrams and Venn diagrams are powerful tools for visualising and solving multi‑stage probability problems.
条件概率 P(A|B) = P(A ∩ B) / P(B) 是在 B 已发生的条件下 A 发生的概率。如果 P(A ∩ B) = P(A) × P(B) 或等价地 P(A|B) = P(A)，则事件 A 和 B 独立。树状图和文氏图是可视化和解决多阶段概率问题的有力工具。
The law of total probability states that if B₁, B₂,… are mutually exclusive and exhaustive, then P(A) = Σ P(A|Bᵢ)P(Bᵢ). Bayes’ theorem extends this to reverse conditional probabilities: P(B|A) = [P(A|B)P(B)] / P(A).
全概率公式指出，若 B₁, B₂,… 互斥且穷尽，则 P(A) = Σ P(A|Bᵢ)P(Bᵢ)。贝叶斯定理将其扩展到逆向条件概率：P(B|A) = [P(A|B)P(B)] / P(A)。
6. Discrete Random Variables | 离散随机变量
A discrete random variable X takes a countable number of distinct values. Its probability distribution lists each possible value x and its probability P(X = x). The sum of all probabilities must equal 1. The expected value E(X) = Σ x·P(X = x) represents the long‑run average. The variance Var(X) = E(X²) − [E(X)]², where E(X²) = Σ x²·P(X = x).
离散随机变量 X 取可数个不同值。其概率分布列出每个可能值 x 及其概率 P(X = x)。所有概率之和必须等于 1。期望值 E(X) = Σ x·P(X = x) 代表长期平均值。方差 Var(X) = E(X²) − [E(X)]²，其中 E(X²) = Σ x²·P(X = x)。
Linear transformations follow simple rules: E(aX + b) = aE(X) + b and Var(aX + b) = a²Var(X). The standard deviation is σ = √Var(X). A cumulative distribution function F(x) = P(X ≤ x) is sometimes required.
线性变换遵循简单规则：E(aX + b) = aE(X) + b，Var(aX + b) = a²Var(X)。标准差为 σ = √Var(X)。有时需要累积分布函数 F(x) = P(X ≤ x)。
The discrete uniform distribution is the simplest: if X takes values 1,2,…,n with equal probability, then P(X = x) = 1/n, E(X) = (n+1)/2. Understanding these basics prepares you for the binomial and other standard distributions.
离散均匀分布是最简单的：若 X 取值 1,2,…,n 且等概率，则 P(X = x) = 1/n，E(X) = (n+1)/2。理解这些基础有助于你学习二项分布等标准分布。
7. 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 success probability p. If X ~ B(n, p), then P(X = r) = ⁿCᵣ p^r (1−p)^(n−r), where ⁿCᵣ = n! / [r!(n−r)!]. The conditions are: fixed n, independent trials, two possible outcomes each trial, constant p.
二项分布描述固定次数的独立试验中，每次试验具有相同成功概率 p 时的成功次数。若 X ~ B(n, p)，则 P(X = r) = ⁿCᵣ p^r (1−p)^(n−r)，其中 ⁿCᵣ = n! / [r!(n−r)!]。条件为：固定 n、试验独立、每次试验两种结果、p 恒定。
The mean and variance are E(X) = np, Var(X) = np(1−p). The distribution is symmetrical when p = 0.5, skewed right when p < 0.5, and skewed left when p > 0.5. Calculations of more than one probability (e.g. P(X ≥ 2)) often require summing individual probabilities or using cumulative tables.
均值和方差为 E(X) = np，Var(X) = np(1−p)。分布当 p = 0.5 时对称，p < 0.5 时右偏，p > 0.5 时左偏。计算多个概率（如 P(X ≥ 2)）通常需对各个概率求和或使用累积表。
As n grows large, the binomial distribution can be approximated by a normal distribution if np and n(1−p) are both greater than 5 (or 10, depending on the required accuracy). A continuity correction must be applied when using the normal approximation.
当 n 很大且 np 和 n(1−p) 都大于 5（或 10，视精度要求而定）时，二项分布可用正态分布近似。使用正态近似时必须进行连续性校正。
8. Normal Distribution | 正态分布
The normal distribution is a continuous probability distribution with a bell‑shaped curve. If X ~ N(μ, σ²), the probability density function is symmetric about the mean μ, with points of inflection at μ ± σ. The total area under the curve is 1, and probabilities correspond to areas.
正态分布是一种连续概率分布，具有钟形曲线。若 X ~ N(μ, σ²)，其概率密度函数关于均值 μ 对称，拐点在 μ ± σ 处。曲线下总面积为 1，概率对应面积。
Standardisation converts any normal variable to Z ~ N(0, 1) using Z = (X − μ) / σ. Standard normal tables give Φ(z) = P(Z ≤ z). For non‑standard problems, find the z‑value, use the table, and interpret. Inverse normal calculations find x given a probability: first find z from the table, then x = μ + zσ.
标准化通过 Z = (X − μ) / σ 将任意正态变量转换为 Z ~ N(0, 1)。标准正态表给出 Φ(z) = P(Z ≤ z)。对于非标准问题，找出 z 值，查表并解释。逆向正态计算在给定概率下求 x：先从表中找出 z，然后 x = μ + zσ。
Many natural phenomena are approximately normally distributed, and the distribution is fundamental to statistical inference. The central limit theorem tells us that the sample mean of a large sample will be approximately normally distributed regardless of the population distribution.
许多自然现象近似正态分布，该分布也是统计推断的基础。中心极限定理告诉我们，无论总体分布如何，大样本的样本均值将近似服从正态分布。
9. Sampling & Estimation | 抽样与估计
A statistic is a function of sample data, e.g. x̄. Its sampling distribution describes how the statistic varies from sample to sample. The standard deviation of a statistic is called its standard error. For the sample mean, if X ~ N(μ, σ²), then x̄ ~ N(μ, σ²/n). More generally, the central limit theorem ensures that for large n, x̄ is approximately normal.
统计量是样本数据的函数，如 x̄。其抽样分布描述了统计量在不同样本间如何变化。统计量的标准差称为标准误差。对于样本均值，若 X ~ N(μ, σ²)，则 x̄ ~ N(μ, σ²/n)。更一般地，中心极限定理保证在大 n 下 x̄ 近似正态。
A point estimate gives a single value for a parameter. An interval estimate (confidence interval) provides a range of plausible values. A 95% confidence interval for μ with known σ is x̄ ± 1.96 × σ/√n. When σ is unknown, we use the sample standard deviation s and the t‑distribution: x̄ ± t × s/√n.
点估计为参数给出单一值。区间估计（置信区间）给出一个合理范围。当 σ 已知时，μ 的 95% 置信区间为 x̄ ± 1.96 × σ/√n。当 σ 未知时，使用样本标准差 s 和 t 分布：x̄ ± t × s/√n。
The width of a confidence interval depends on confidence level, sample size, and variability. Larger samples or lower confidence levels produce narrower intervals. Correct interpretation: if we repeated the sampling many times, 95% of the constructed intervals would capture μ.
置信区间的宽度取决于置信水平、样本量和变异性。更大的样本量或更低的置信水平会产生更窄的区间。正确解释：如果多次重复抽样，构造的区间中将有 95% 包含 μ。
10. Hypothesis Testing | 假设检验
Hypothesis testing is a formal procedure to assess whether sample data provide enough evidence to support a claim about a population parameter. The null hypothesis H₀ represents a position of no effect or status quo (often H₀: μ = μ₀). The alternative hypothesis H₁ is what we aim to support (e.g., μ > μ₀, μ < μ₀, or μ ≠ μ₀).
假设检验是用来评估样本数据是否为关于总体参数的某个论断提供了充分证据的正式程序。原假设 H₀ 代表无效应或现状的立场（通常 H₀: μ = μ₀）。备择假设 H₁ 是我们试图支持的论断（如 μ > μ₀，μ < μ₀ 或 μ ≠ μ₀）。
The test statistic measures how far the sample statistic is from H₀. For the mean: z = (x̄ − μ₀) / (σ/√n) [if σ known] or t = (x̄ − μ₀) / (s/√n) [if σ unknown]. The p‑value is the probability of obtaining a test statistic at least as extreme as the observed, assuming H₀ is true. We reject H₀ if p < α (significance level, often 0.05).
检验统计量衡量样本统计量与 H₀ 的距离。对于均值：z = (x̄ − μ₀) / (σ/√n) [σ 已知] 或 t = (x̄ − μ₀) / (s/√n) [σ 未知]。p 值是在 H₀ 为真的前提下得到至少与观测值一样极端的检验统计量的概率。若 p < α（显著性水平，通常 0.05），则拒绝 H₀。
Two types of errors can occur: Type I error is rejecting a true H₀ (probability = α); Type II error is failing to reject a false H₀. The power of a test is 1 − β, where β is the probability of a Type II error. Always write conclusions in context, not just ‘reject H₀’.
可能出现两类错误：第一类错误是拒绝真实的 H₀（概率 = α）；第二类错误是未能拒绝错误的 H₀。检验的功效为 1 − β，其中 β 是第二类错误的概率。请始终结合上下文撰写结论，而不仅仅写“拒绝 H₀”。
11. Correlation and Regression | 相关与回归
Bivariate data involve two variables measured on the same individuals. A scatter plot reveals the relationship. Correlation measures the strength and direction of a linear association. The product‑moment correlation coefficient (Pearson’s r) is given by:
r = Σ(x − x̄)(y − ȳ) / √[Σ(x − x̄)² Σ(y − ȳ)²]
r lies between −1 and +1. r close to ±1 indicates strong linear correlation; r ≈ 0 suggests no linear correlation. Note that correlation does not imply causation.
双变量数据涉及同一样本个体上测量的两个变量。散点图能揭示关系。相关衡量线性关联的强度和方向。积矩相关系数（皮尔逊 r）由上述公式给出。r 介于 −1 和 +1 之间。r 接近 ±1 表示强线性相关；r ≈ 0 表明无线性相关。注意相关并不意味因果。
Linear regression finds the line of best fit: y = a + bx. The least squares method minimises Σ(y − ŷ)². The slope b = S_xy / S_xx, where S_xy = Σ(x − x̄)(y − ȳ) and S_xx = Σ(x − x̄)². The intercept a = ȳ − bx̄. This regression line of y on x is used to predict y for given x, but care must be taken with extrapolation.
线性回归寻找最佳拟合线：y = a + bx。最小二乘法最小化 Σ(y − ŷ)²。斜率 b = S_xy / S_xx，其中 S_xy = Σ(x − x̄)(y − ȳ)，S_xx = Σ(x − x̄)²。截距 a = ȳ − bx̄。这条 y 对 x 的回归线用于给定 x 预测 y，但外推时需谨慎。
Residuals (e = y − ŷ) should be randomly scattered around zero. Patterns may indicate a curved relationship or non‑constant variance. The coefficient of determination r² represents the proportion of variation in y explained by x.
残差（e = y − ŷ）应随机分布在零附近。出现模式可能暗示曲线关系或方差非常数。决定系数 r² 表示 y 中由 x 解释的变异比例。
12. Summary & Exam Tips | 总结与备考建议
Mastering A‑Level statistics requires a blend of conceptual understanding and procedural fluency. Always define variables, state formulas clearly, and show working step by step. Use correct statistical notation and ensure your calculator is in the right mode (degree/radian, statistical mode).
掌握 A‑Level 统计需要概念理解与流程熟练相结合。始终定义变量、清晰地写出公式并逐步展示过程。使用正确的统计符号，并确保计算器处于正确的模式（度/弧度、统计模式）。
When interpreting results, always relate your answer to the context of the problem. For hypothesis tests, write a full conclusion: compare p‑value to significance level, state whether H₀ is rejected, and interpret what that means in words. Watch out for common pitfalls like confusing one‑tailed and two‑tailed tests, forgetting continuity corrections, or misreading normal tables.
解释结果时，始终将答案与问题上下文关联。对于假设检验，写出完整结论：比较 p 值与显著性水平，说明是否拒绝 H₀，并用文字解释其含义。注意常见陷阱，如混淆单尾与双尾检验、忘记连续性校正或误读正态分布表。
Practise past papers regularly, time yourself, and review mark schemes to understand what examiners expect. Build your bilingual vocabulary of statistical terms — it will serve you well whether you are writing exams or communicating mathematics internationally.
定期练习历年真题、计时并查看评分标准，以了解考官的期望。掌握统计术语的双语词汇——无论参加考试还是国际数学交流，都会对你大有裨益。
Published by TutorHao | Maths Stats 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Cancel reply