📚 AS Edexcel Statistics: Core Concepts Review | AS Edexcel 统计：核心知识点梳理
Mastering AS Edexcel Statistics begins with a solid understanding of data types, representation, measures of centre and spread, probability, key distributions, and hypothesis testing. This article distils the essential concepts you will encounter across the specification, offering clear descriptions, key formulas, and exam-orientated advice. Whether you are building your revision notes or checking for gaps, this resource will guide you through every major topic.
掌握 AS Edexcel 统计，首先要牢固理解数据类型、数据表示、集中趋势与离散程度的度量、概率、重要分布以及假设检验。本文梳理了考纲中必学的核心知识，提供清晰的讲解、关键公式和应试要点。无论你是在整理复习笔记还是查漏补缺，这篇内容都能帮助你系统把握每一块重要主题。
1. Data Types and Sampling | 数据类型与抽样
Data can be classified as qualitative (categorical) or quantitative (numerical). Quantitative data is further divided into discrete data, which take exact numerical values (e.g. number of students), and continuous data, which can take any value within a range (e.g. height, time). Understanding these distinctions is fundamental to choosing appropriate diagrams and calculations.
数据可以分为定性（分类）数据和定量（数值）数据。定量数据又可细分为离散数据（取精确数值，如学生人数）和连续数据（在某个区间内可取任意值，如身高、时间）。理解这些区别是选择恰当图表与计算方法的基础。
Sampling methods enable us to draw conclusions about a population using a smaller subset. In AS Edexcel you need to recognise random sampling (simple, systematic, stratified), which allows probability-based inference, and non-random sampling (quota, opportunity/convenience), which may introduce bias but can be quicker or cheaper. The sampling frame—a list of all population members—is crucial for random methods.
抽样方法让我们能够通过较小的子集对总体下结论。AS Edexcel 要求你识别随机抽样（简单、系统、分层），它允许基于概率的推断；以及非随机抽样（配额、机会/便利），它可能引入偏差但有时更快或更经济。抽样框——列出所有总体成员的名单——对随机方法至关重要。
2. Graphical Representations | 图形表示
Box plots (box‑and‑whisker diagrams) display the minimum, lower quartile (Q₁), median (Q₂), upper quartile (Q₃) and maximum. They are excellent for comparing distributions and highlighting skewness. The interquartile range (IQR = Q₃ – Q₁) measures spread and is unaffected by outliers.
箱线图（盒须图）展示最小值、下四分位数（Q₁）、中位数（Q₂）、上四分位数（Q₃）与最大值。它非常适合比较分布并突出偏态。四分位距（IQR = Q₃ – Q₁）度量离散程度且不受异常值影响。
Histograms group continuous data into classes. The key rule is that frequency is proportional to area, so the vertical axis is frequency density = frequency ÷ class width. Outliers can be identified using the rule: values below Q₁ – 1.5×IQR or above Q₃ + 1.5×IQR. Cumulative frequency curves aid in estimating medians and percentiles.
直方图将连续数据分组，核心规则是频数与面积成正比，因此纵轴为频数密度 = 频数 ÷ 组距。异常值可通过规则识别：小于 Q₁ – 1.5×IQR 或大于 Q₃ + 1.5×IQR 的值。累积频率曲线有助于估算中位数和百分位数。
3. Measures of Central Tendency and Dispersion | 集中趋势与离散度量
The three common averages are the mean, median and mode. The mean (x̄ or μ) uses all data values but is sensitive to outliers. The median is the middle value and is robust to outliers, making it preferable for skewed distributions. The mode is the most frequent value and can be used with categorical data.
三种常见的平均值是平均数、中位数和众数。平均数（x̄ 或 μ）用上所有数据值但对异常值敏感。中位数是中间值，对异常值稳健，因此在偏态分布中更受青睐。众数是出现频率最高的值，可用于分类数据。
Measures of dispersion include the range, interquartile range (IQR), variance and standard deviation. The variance for a population is σ² = Σ(x – μ)²/N, while for a sample it is s² = Σ(x – x̄)²/(n – 1). The standard deviation is the square root of the variance. Linear coding y = a + bx affects the mean as E(Y) = a + bE(X) and the standard deviation as σy = |b|σx.
离散程度的度量包括极差、四分位距（IQR）、方差和标准差。总体方差为 σ² = Σ(x – μ)²/N，样本方差为 s² = Σ(x – x̄)²/(n – 1)。标准差是方差的平方根。线性变换 y = a + bx 影响平均数：E(Y) = a + bE(X)，标准差为 σy = |b|σx。
4. Bivariate Data: Correlation and Regression | 双变量数据：相关与回归
Correlation measures the strength and direction of a linear relationship between two variables. The product moment correlation coefficient (PMCC) r is given by r = Sxy / √(Sxx Syy), where Sxy = Σ(x – x̄)(y – ȳ) and similarly for Sxx, Syy. Values of r near +1 indicate strong positive correlation, near –1 strong negative, and near 0 weak or no linear correlation.
相关度量两个变量间线性关系的强度与方向。积矩相关系数 (PMCC) r 公式为 r = Sxy / √(Sxx Syy)，其中 Sxy = Σ(x – x̄)(y – ȳ)，Sxx、Syy 同理。r 接近 +1 表示强正相关，接近 –1 表示强负相关，接近 0 表示弱或无线性相关。
The least squares regression line of y on x is y = a + bx, where b = Sxy/Sxx and a = ȳ – bx̄. This line can be used for prediction within the range of observed x-values (interpolation) but should not be trusted for x-values far outside the original range (extrapolation). Always interpret slope and intercept in context.
y 对 x 的最小二乘回归直线方程为 y = a + bx，其中斜率 b = Sxy/Sxx，截距 a = ȳ – bx̄。该直线可用于观测 x 值范围内的预测（内插法），但不应用于远离原始范围的 x 值（外推法）。始终要结合实际意义解释斜率和截距。
5. Probability Basics | 概率基础
Probability of an event A, denoted P(A), lies between 0 and 1. For mutually exclusive events, P(A ∪ B) = P(A) + P(B). For any two events, the addition rule is P(A ∪ B) = P(A) + P(B) – P(A ∩ B). Independent events satisfy P(A ∩ B) = P(A)P(B), and conditional probability is defined as P(A|B) = P(A ∩ B)/P(B).
事件 A 的概率 P(A) 取值在 0 到 1 之间。对互斥事件，P(A ∪ B) = P(A) + P(B)。对任意两事件，加法法则是 P(A ∪ B) = P(A) + P(B) – P(A ∩ B)。独立事件满足 P(A ∩ B) = P(A)P(B)，条件概率定义为 P(A|B) = P(A ∩ B)/P(B)。
Tree diagrams help visualise successive independent events or conditional trials, multiplying along branches. Venn diagrams and sample space diagrams are also essential tools. A common exam task is to complete a probability distribution from given information, ensuring that all probabilities sum to 1.
树状图有助于可视化连续的独立事件或条件试验，沿分支相乘。维恩图和样本空间图也是重要工具。考试常见任务是利用给定信息补全概率分布，并确保所有概率之和为 1。
6. Discrete Random Variables | 离散随机变量
A discrete random variable X takes a countable set of values with probabilities P(X = x). The probability distribution must satisfy Σ P(X = x) = 1. The expected value E(X) = Σ x·P(X = x) gives the long‑term average, and the variance Var(X) = E(X²) – [E(X)]² = Σ (x – μ)² P(X = x).
离散随机变量 X 取可数个值，对应概率为 P(X = x)。概率分布必须满足 Σ P(X = x) = 1。期望值 E(X) = Σ x·P(X = x) 给出长期平均值，方差 Var(X) = E(X²) – [E(X)]² = Σ (x – μ)² P(X = x)。
For a linear transformation Y = a + bX, the expected value becomes E(Y) = a + bE(X) and the variance becomes Var(Y) = b² Var(X). This is particularly useful when dealing with coded data or combining random variables.
对于线性变换 Y = a + bX，期望值变为 E(Y) = a + bE(X)，方差变为 Var(Y) = b² Var(X)。这在处理编码数据或组合随机变量时特别有用。
7. Binomial Distribution | 二项分布
A binomial distribution arises from a fixed number n of independent trials, each with two outcomes (success/failure) and a constant probability of success p. We write X ~ B(n, p). The probability of exactly r successes is P(X = r) = C(n, r) pr (1 – p)n–r, where C(n, r) = n! / [r!(n – r)!].
二项分布基于固定的试验次数 n，每次试验相互独立，结果只有成功或失败，且成功的概率 p 恒定。记作 X ~ B(n, p)。恰好 r 次成功的概率为 P(X = r) = C(n, r) pr (1 – p)n–r，其中 C(n, r) = n! / [r!(n – r)!]。
The mean and variance of a binomial distribution are μ = np and σ² = np(1 – p). Tables or calculators can be used to find cumulative probabilities P(X ≤ x) or individual probabilities. You will need to interpret word problems, identify n and p, and decide whether ‘greater than’, ‘at least’ or ‘at most’ phrasing is used.
二项分布的均值和方差分别为 μ = np 和 σ² = np(1 – p)。可使用表格或计算器求累积概率 P(X ≤ x) 或单个概率。你需要解读文字题，识别 n 和 p，并判断题目是要求“大于”“至少”还是“至多”的概率。
8. Normal Distribution | 正态分布
The normal distribution is a continuous symmetric distribution defined by its mean μ and variance σ², written as X ~ N(μ, σ²). The total area under the curve is 1. To find probabilities, we standardise to the standard normal Z ~ N(0, 1²) using z = (x – μ)/σ. Then P(X < x) = P(Z < z) can be read from tables.
正态分布是一种连续对称分布，由其均值 μ 和方差 σ² 定义，记作 X ~ N(μ, σ²)。曲线下的总面积为 1。求概率时，我们用 z = (x – μ)/σ 标准化为标准正态 Z ~ N(0, 1²)，然后 P(X < x) = P(Z < z) 可通过表格查得。
The inverse normal function is used when given a probability and needing the corresponding x-value. Key percentages from the empirical rule: roughly 68% of data lies within μ ± σ, 95% within μ ± 1.96σ, and 99.7% within μ ± 3σ. Always sketch a bell curve to visualise the area you need.
当已知概率需要求对应 x 值时，使用逆正态函数。根据经验法则的关键百分比：约 68% 的数据落在 μ ± σ 内，95% 落在 μ ± 1.96σ 内，99.7% 落在 μ ± 3σ 内。一定要画出钟形曲线来可视化所需的面积。
9. Hypothesis Testing | 假设检验
A hypothesis test assesses evidence for a claim about a population parameter. The null hypothesis H₀ assumes no effect or status quo; the alternative H₁ is what we seek evidence for. Tests can be one‑tailed (H₁: p < ... or p > …) or two‑tailed (H₁: p ≠ …).
假设检验用来评估关于总体参数的某个主张是否有证据支持。原假设 H₀ 假定无效应或维持现状；备择假设 H₁ 是我们寻求证据支持的主张。检验可以是单尾（H₁: p < … 或 p > …）或双尾（H₁: p ≠ …）。
For a binomial test of proportion, the test statistic is the observed number of successes. The critical region is determined by the significance level α (often 5% or 1%). We reject H₀ if the test statistic falls in the critical region, or if the p‑value is less than α. The conclusion must be written in context and avoid claiming ‘prove’—we say ‘there is sufficient evidence to reject H₀’ or ‘insufficient evidence to reject H₀’.
对于二项分布的比例检验，检验统计量是观测到的成功次数。临界区由显著性水平 α（通常为 5% 或 1%）决定。如果检验统计量落在临界区内，或 p 值小于 α，则拒绝 H₀。必须结合实际写出结论，避免说“证明”——我们应说“有充分证据拒绝 H₀”或“证据不足以拒绝 H₀”。
When testing a mean with a known population variance, you use the normal distribution. The test statistic is z = (x̄ – μ₀)/(σ/√n). The same principles of critical regions and p‑values apply. Always check the assumptions before carrying out the test.
当已知总体方差时检验均值，使用正态分布。检验统计量为 z = (x̄ – μ₀)/(σ/√n)。临界区和 p 值的原理相同。进行检验前务必检查前提假设。
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Cancel reply