Pre-U Edexcel Statistics: Core Concepts Overview | Pre-U Edexcel 统计：核心知识点梳理

📚 Pre-U Edexcel Statistics: Core Concepts Overview | Pre-U Edexcel 统计：核心知识点梳理

Mastering Edexcel Pre-U Statistics requires a solid grasp of data representation, probability models, statistical inference, and the ability to apply these concepts to real-world problems. This guide distills the essential topics into a clear, structured overview, pairing every English explanation with a corresponding Chinese translation to support bilingual learners. From sampling methods to hypothesis testing, each section builds your confidence for the final examination.

掌握 Edexcel Pre-U 统计学需要牢固掌握数据表示、概率模型、统计推断以及将这些概念应用于实际问题的能力。本指南将核心知识点浓缩成一个清晰、有条理的概览，每一条英文解释都配有对应的中文翻译，以支持双语学习者。从抽样方法到假设检验，每一节都将帮助你建立应对最终考试的信心。

---

1. Types of Data and Sampling Methods | 数据类型与抽样方法

Data can be classified as qualitative (categorical, e.g. colours or brands) or quantitative (numerical). Quantitative data may be discrete, taking only distinct values such as counts, or continuous, taking any value within a range such as height. Understanding the type of data is the first step in choosing an appropriate statistical analysis.

数据可以分为定性（类别，例如颜色或品牌）和定量（数值）。定量数据可以是离散的，仅取计数值等不连续的值，也可以是连续的，取一定范围内的任意值，如身高。理解数据类型是选择合适统计分析方法的第一步。

Simple random sampling gives every member of the population an equal chance of selection, minimising bias. Stratified sampling divides the population into distinct groups and draws a proportionate sample from each stratum, ensuring representation of key subgroups. Other methods include systematic, quota, and opportunity sampling, each with their own advantages and limitations.

简单随机抽样使总体中每个成员被选中的机会相等，从而将偏差降至最低。分层抽样将总体划分为不同的组，并从每个层中按比例抽取样本，确保关键子群体的代表性。其他方法包括系统抽样、配额抽样和机会抽样，每种方法都有其优点和局限性。

---

2. Measures of Central Tendency | 集中趋势的度量

The three principal measures are the mean, median, and mode. For a set of n observations x₁, x₂, …, xₙ, the arithmetic mean is x̄ = (∑xᵢ)/n. The mean incorporates every data point and is suitable for symmetric distributions, but it is sensitive to outliers.

三个主要度量是平均数、中位数和众数。对于n个观测值 x₁, x₂, …, xₙ，算术平均数为 x̄ = (∑xᵢ)/n。平均数利用了每一个数据点，适用于对称分布，但对异常值敏感。

The median is the middle value when data are ordered; it is robust to extreme values and is preferred for skewed distributions. The mode is the most frequently occurring value and is particularly useful for categorical data. In a symmetric distribution, the mean, median, and mode coincide.

中位数是数据排序后位于中间的值；它对极端值具有稳健性，适合偏态分布。众数是出现频率最高的值，特别适用于分类数据。在对称分布中，平均数、中位数和众数重合。

---

3. Measures of Dispersion and Spread | 离散程度与数据散布的度量

Range and interquartile range (IQR) give simple measures of spread. The range is the difference between the maximum and minimum values. The IQR, calculated as Q₃ – Q₁, covers the middle 50% of the data and is resistant to outliers, making it useful for comparing box plots.

极差和四分位距 (IQR) 给出了简单的离散度量。极差是最大值与最小值之差。IQR 计算为 Q₃ – Q₁，覆盖中间50%的数据，对异常值不敏感，因此非常适合用于箱线图的比较。

Variance and standard deviation quantify spread around the mean. For a population, variance is σ² = ∑(xᵢ – μ)² / N; for a sample, we use s² = ∑(xᵢ – x̄)² / (n – 1). The standard deviation is the square root of variance, σ or s, and shares the same units as the original data. A smaller standard deviation indicates data closely clustered around the mean.

方差和标准差量化了数据围绕平均数的离散程度。对于总体，方差为 σ² = ∑(xᵢ – μ)² / N；对于样本，我们使用 s² = ∑(xᵢ – x̄)² / (n – 1)。标准差是方差的平方根 σ 或 s，与原始数据具有相同的单位。较小的标准差表示数据紧密地聚集在平均数周围。

---

4. Fundamentals of Probability | 概率基础

The probability of an event A, denoted P(A), lies between 0 and 1. The addition rule for mutually exclusive events states P(A ∪ B) = P(A) + P(B). When events are not mutually exclusive, the general addition rule applies: P(A ∪ B) = P(A) + P(B) – P(A ∩ B).

事件 A 的概率，记作 P(A)，在0到1之间。互斥事件的加法规则为 P(A ∪ B) = P(A) + P(B)。当事件不互斥时，应用一般加法规则：P(A ∪ B) = P(A) + P(B) – P(A ∩ B)。

Conditional probability is defined as P(A|B) = P(A ∩ B) / P(B), provided P(B) > 0. Two events are independent if and only if P(A ∩ B) = P(A) × P(B), which also implies P(A|B) = P(A). Tree diagrams and Venn diagrams are powerful tools for visualising compound events and solving multi-stage probability problems.

条件概率定义为 P(A|B) = P(A ∩ B) / P(B)，前提是 P(B) > 0。当且仅当 P(A ∩ B) = P(A) × P(B) 时，两个事件独立，这也意味着 P(A|B) = P(A)。树状图和韦恩图是可视化复合事件并求解多阶段概率问题的强大工具。

---

5. Discrete Random Variables and Expectation | 离散随机变量与期望

A discrete random variable X takes a countable set of values, each with an associated probability P(X = x) = p(x). The sum of all probabilities must equal 1. The probability mass function (PMF) describes this distribution completely.

离散随机变量 X 取可数个值，每个值都有相应的概率 P(X = x) = p(x)。所有概率之和必须等于1。概率质量函数 (PMF) 完整地描述了这一分布。

The expected value E(X) = ∑ x·p(x) gives the theoretical long-run average. The variance Var(X) = E[(X – μ)²] = E(X²) – [E(X)]², measuring the spread of the distribution. For discrete uniform, binomial, and Poisson models, these moments take specific formulas that streamline calculations.

期望值 E(X) = ∑ x·p(x) 给出了理论上的长期平均值。方差 Var(X) = E[(X – μ)²] = E(X²) – [E(X)]²，衡量分布的离散程度。对于离散均匀分布、二项分布和泊松分布，这些矩采用特定的公式，简化了计算过程。

---

6. Binomial and Poisson Distributions | 二项分布与泊松分布

A binomial distribution, X ~ B(n, p), models the number of successes in n independent trials, each with a constant probability of success p. Its probability function is P(X = k) = ⁿCₖ pᵏ (1 – p)ⁿ⁻ᵏ, with mean μ = np and variance σ² = np(1 – p). The distribution is symmetric when p = 0.5 and becomes skewed as p deviates from 0.5.

二项分布 X ~ B(n, p) 用于建模在 n 次独立试验中成功的次数，每次试验成功的概率恒为 p。其概率函数为 P(X = k) = ⁿCₖ pᵏ (1 – p)ⁿ⁻ᵏ，均值为 μ = np，方差为 σ² = np(1 – p)。当 p = 0.5 时分布对称，当 p 偏离 0.5 时呈偏态。

The Poisson distribution, X ~ Po(λ), models the number of events occurring in a fixed interval of time or space, assuming events happen independently at a constant mean rate λ. Its formula is P(X = k) = e⁻λ λᵏ / k!, with mean and variance both equal to λ. When n is large and p is small, the binomial distribution can be approximated by a Poisson with λ = np.

泊松分布 X ~ Po(λ) 用于建模在固定时间或空间间隔内发生的事件数，假设事件以恒定平均速率 λ 独立发生。其公式为 P(X = k) = e⁻λ λᵏ / k!，均值和方差都等于 λ。当 n 较大而 p 较小时，二项分布可用 λ = np 的泊松分布近似。

---

7. The Normal Distribution and Standardisation | 正态分布与标准化

The normal distribution, N(μ, σ²), is a continuous, bell-shaped distribution that is completely defined by its mean μ and variance σ². The curve is symmetric about μ, and approximately 68% of observations fall within one standard deviation of the mean, 95% within two, and 99.7% within three.

正态分布 N(μ, σ²) 是一种连续钟形分布，完全由其均值 μ 和方差 σ² 定义。曲线关于 μ 对称，大约68%的观测值落在均值的一个标准差内，95%落在两个标准差内，99.7%落在三个标准差内。

Any normal variable can be standardised using Z = (X – μ) / σ, which follows the standard normal distribution N(0,1). Tables of the cumulative standard normal function Φ(z) allow probability calculations, and inverse normal tables help find critical values for confidence intervals and hypothesis tests. The normal also serves as an approximation for binomial and Poisson distributions under suitable conditions, with continuity correction applied when converting discrete to continuous scales.

任何正态变量都可以通过 Z = (X – μ) / σ 标准化，服从标准正态分布 N(0,1)。标准正态累积函数 Φ(z) 的表可用于概率计算，逆正态表有助于查找置信区间和假设检验的临界值。在适当条件下，正态分布也可作为二项分布和泊松分布的近似，此时从离散到连续尺度需应用连续性校正。

---

8. Correlation and Linear Regression | 相关与线性回归

The Pearson product-moment correlation coefficient r measures the strength and direction of a linear relationship between two variables. It is given by r = Sxy / √(Sxx · Syy), where Sxy = ∑(xᵢ – x̄)(yᵢ – ȳ), Sxx = ∑(xᵢ – x̄)², and Syy = ∑(yᵢ – ȳ)². Values of r close to 1 or –1 indicate a strong linear association, while values near 0 suggest little or no linear trend.

皮尔逊积差相关系数 r 衡量两个变量之间线性关系的强度和方向。计算公式为 r = Sxy / √(Sxx · Syy)，其中 Sxy = ∑(xᵢ – x̄)(yᵢ – ȳ)，Sxx = ∑(xᵢ – x̄)²，Syy = ∑(yᵢ – ȳ)²。r 值接近 1 或 –1 表示强线性关联，接近 0 提示线性趋势微弱或不存在。

The least squares regression line of y on x has equation y = a + bx, where b = Sxy / Sxx and a = ȳ – b·x̄. This line minimises the sum of squared vertical residuals. The regression coefficient b represents the estimated change in y for a one-unit increase in x. It is essential to distinguish between the response variable (dependent) and the explanatory variable (independent) when applying regression.

y 对 x 的最小二乘回归线方程为 y = a + bx，其中 b = Sxy / Sxx，a = ȳ – b·x̄。这条直线使垂直残差平方和最小。回归系数 b 表示 x 每增加一个单位时 y 的估计变化量。应用回归时，必须区分响应变量（因变量）和解释变量（自变量）。

---

9. Introduction to Hypothesis Testing | 假设检验入门

Hypothesis testing is a formal procedure for evaluating claims about a population parameter. It begins with stating a null hypothesis H₀ and an alternative hypothesis H₁. The significance level α (often 0.05) defines the threshold for rejecting H₀. A test statistic is computed from the sample, and its probability under H₀ (the p-value) is compared with α.

假设检验是一种评估关于总体参数的主张的正式程序。它从提出零假设 H₀ 和备择假设 H₁ 开始。显著性水平 α（常取0.05）定义了拒绝 H₀ 的门槛。根据样本计算检验统计量，并将其在 H₀ 下的概率（p 值）与 α 进行比较。

For a binomial test of a proportion p, H₀: p = p₀, the p-value is the probability of obtaining a result at least as extreme as the observed value, using the binomial distribution. If p-value ≤ α, H₀ is rejected in favour of H₁. Critical regions and critical values provide an equivalent approach: if the test statistic falls into the critical region, H₀ is rejected. One-tailed tests address directional alternatives (e.g. p > p₀), while two-tailed tests address non-directional alternatives (p ≠ p₀).

对于比例 p 的二项检验，H₀: p = p₀，p 值是使用二项分布计算得到的、与观测值至少同样极端的结果出现的概率。若 p 值 ≤ α，则拒绝 H₀，接受 H₁。临界区域和临界值提供了等价的处理方法：若检验统计量落入临界区域，则拒绝 H₀。单尾检验针对有方向性的备择（如 p > p₀），而双尾检验针对无方向性的备择（p ≠ p₀）。

---

10. Further Testing: t‑tests and Chi‑squared Tests | 进阶检验：t 检验与卡方检验

When the population variance σ² is unknown, a one-sample t‑test uses the sample standard deviation s and the t distribution with n – 1 degrees of freedom. The test statistic is t = (x̄ – μ₀) / (s/√n). Paired t‑tests compare two dependent samples by analysing differences, while two-sample t‑tests compare independent means, assuming equal or unequal variances depending on the setting.

当总体方差 σ² 未知时，单样本 t 检验使用样本标准差 s 和自由度为 n – 1 的 t 分布。检验统计量为 t = (x̄ – μ₀) / (s/√n)。配对 t 检验通过分析差值比较两个依赖样本，而双样本 t 检验比较独立均值，根据情况假定方差相等或不相等。

Chi‑squared (χ²) tests assess the fit between observed and expected frequencies. For a goodness-of-fit test, the statistic is χ² = ∑ (Oᵢ – Eᵢ)² / Eᵢ, with degrees of freedom ν = number of categories – 1 – (number of estimated parameters). For a contingency table testing independence between two categorical variables, expected frequencies are calculated from row and column totals, and ν = (rows – 1)(columns – 1). Large χ² values indicate a poor fit or an association between variables.

卡方 (χ²) 检验评估观测频率与期望频率之间的拟合度。对于拟合优度检验，统计量为 χ² = ∑ (Oᵢ – Eᵢ)² / Eᵢ，自由度 ν = 类别数 – 1 –（估计参数个数）。对于检验两个分类变量独立性的列联表，期望频率由行列总和计算得出，且 ν = (行数 – 1)(列数 – 1)。较大的 χ² 值表示拟合不佳或变量之间存在关联。

---

11. Sampling Distributions and the Central Limit Theorem | 抽样分布与中心极限定理

A sampling distribution describes the probability distribution of a statistic (e.g. the sample mean) calculated from all possible samples of a fixed size drawn from a population. Understanding this distribution is fundamental to inference because it links sample estimates to population parameters.

抽样分布描述了从总体中抽取的所有固定容量样本所计算的统计量（例如样本均值）的概率分布。理解这一分布是推断的基础，因为它将样本估计量与总体参数联系起来。

The Central Limit Theorem (CLT) states that, for a sufficiently large sample size n (typically n ≥ 30), the sampling distribution of the sample mean is approximately normal with mean μ and variance σ²/n, regardless of the shape of the population distribution. This justifies the use of z‑procedures and t‑procedures in hypothesis testing and confidence intervals for means when dealing with large samples or normally distributed populations.

中心极限定理 (CLT) 指出，对于足够大的样本容量 n（通常 n ≥ 30），样本均值的抽样分布近似服从均值为 μ、方差为 σ²/n 的正态分布，无论总体分布形状如何。这为在处理大样本或正态总体时，在均值的假设检验和置信区间中使用 z 程序和 t 程序提供了理论依据。

---

12. Confidence Intervals for Means and Proportions | 均值与比例的置信区间

A confidence interval provides a range of plausible values for an unknown population parameter, with a specified level of confidence (e.g. 95%). For a population mean with known σ, the interval is x̄ ± z* · σ/√n, where z* is the critical value from N(0,1) corresponding to the chosen confidence level. When σ is unknown, the t‑distribution replaces the normal, giving x̄ ± t* · s/√n with n – 1 degrees of freedom.

置信区间为未知总体参数提供一个合理值的范围，并指定置信水平（如95%）。对于已知 σ 的总体均值，区间为 x̄ ± z* · σ/√n，其中 z* 是对应选定置信水平的 N(0,1) 临界值。当 σ 未知时，用 t 分布替代正态分布，得到 x̄ ± t* · s/√n，自由度为 n – 1。

For a population proportion p, the large-sample confidence interval uses the normal approximation: p̂ ± z* · √(p̂(1 – p̂)/n), where p̂ is the sample proportion. The margin of error reflects variability due to sampling, and the width of the interval decreases as the sample size increases. Correct interpretation stresses that the interval captures the parameter in the long run across repeated samples.

对于总体比例 p，大样本置信区间采用正态近似：p̂ ± z* · √(p̂(1 – p̂)/n)，其中 p̂ 为样本比例。误差边际反映了因抽样引起的变异，且区间宽度随样本量增加而减小。正确的解读强调，在重复抽样下，该区间在长期中能捕捉住参数。

---

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）