AS AQA Further Mathematics 9665 Statistics Revision Guide | AS AQA 进阶数学 9665 统计学复习指南

📚 AS AQA Further Mathematics 9665 Statistics Revision Guide | AS AQA 进阶数学 9665 统计学复习指南

This comprehensive revision guide covers the key topics for the Statistics component of the OxfordAQA International AS Level Further Mathematics 9665 specification. Each section focuses on the essential concepts, formulas, and exam techniques you need to secure top marks.

本复习指南涵盖牛津国际AQA AS进阶数学（9665）统计学部分的核心考点。每个章节聚焦于必考概念、公式和考试技巧，助你在考试中稳拿高分。

---

1. Probability Fundamentals | 概率基础

Probability measures the likelihood of an event occurring, always lying between 0 and 1. For any two mutually exclusive events A and B, the addition rule states that P(A ∪ B) = P(A) + P(B). For independent events, the multiplication rule gives P(A ∩ B) = P(A) × P(B).

概率衡量事件发生的可能性，其值总在0到1之间。对于任意两个互斥事件A和B，加法法则给出 P(A ∪ B) = P(A) + P(B)。对于独立事件，乘法法则给出 P(A ∩ B) = P(A) × P(B)。

You must be able to distinguish between mutually exclusive events (which cannot occur simultaneously) and independent events (where one event’s occurrence does not affect the other’s probability). Venn diagrams and tree diagrams are powerful tools for visualising these relationships.

你必须能够区分互斥事件（不能同时发生）与独立事件（一个事件的发生不影响另一事件的概率）。韦恩图和树状图是可视化这些关系的强大工具。

P(A ∪ B) = P(A) + P(B) − P(A ∩ B)

The general addition rule applies to all events—subtracting the intersection prevents double-counting. When dealing with complement events, remember P(A′) = 1 − P(A).

通用加法法则适用于所有事件——减去交集部分可避免重复计算。处理补事件时，记住 P(A′) = 1 − P(A)。

---

2. Conditional Probability and Bayes’ Theorem | 条件概率与贝叶斯定理

Conditional probability, denoted P(A|B), represents the probability of event A occurring given that B has already occurred. The fundamental formula is P(A|B) = P(A ∩ B) / P(B), provided P(B) > 0.

条件概率，记作 P(A|B)，表示在事件B已发生的条件下事件A发生的概率。基本公式为 P(A|B) = P(A ∩ B) / P(B)，前提是 P(B) > 0。

Bayes’ Theorem extends this concept to reverse conditional probabilities. This is particularly useful when a problem provides P(B|A) but asks for P(A|B).

贝叶斯定理将这一概念推广至反转条件概率。当题目给出 P(B|A) 却要求 P(A|B) 时尤为有用。

P(A|B) = [P(B|A) × P(A)] / P(B)

In the 9665 syllabus, the denominator P(B) is frequently found using the law of total probability, which requires partitioning the sample space into mutually exclusive events. A two-way table or tree diagram with probabilities labelled on branches is often the most efficient approach in exam questions.

在9665大纲中，分母 P(B) 常通过全概率公式求得，即需要将样本空间划分为互斥事件。在考试题中，使用双向表格或在树状图分支上标注概率往往是最有效的方法。

• Always check that your conditional probability is between 0 and 1.
• Draw a tree diagram when the problem involves sequential events.
• Use P(A ∩ B) = P(A|B) × P(B) to move between joint and conditional probabilities.

• 务必检查条件概率在0到1之间。
• 当问题涉及相继事件时，画出树状图。
• 利用 P(A ∩ B) = P(A|B) × P(B) 在联合概率与条件概率之间转换。

---

3. Discrete Random Variables | 离散随机变量

A discrete random variable X takes distinct values with associated probabilities. The probability distribution function (PDF) must satisfy two conditions: each P(X = xᵢ) ≥ 0, and the sum of all probabilities equals 1.

离散随机变量X取不同的值，每个值对应一个概率。其概率分布函数（PDF）须满足两个条件：每个 P(X = xᵢ) ≥ 0，且所有概率之和等于1。

The expected value (mean) E(X) = μ is the weighted average of the outcomes, calculated as Σ xᵢ P(X = xᵢ). The variance Var(X) = σ² measures the spread of the distribution.

期望值（均值）E(X) = μ 是结果的加权平均值，计算公式为 Σ xᵢ P(X = xᵢ)。方差 Var(X) = σ² 衡量分布的离散程度。

Var(X) = E(X²) − [E(X)]² = Σ xᵢ² P(X = xᵢ) − μ²

For a linear transformation Y = aX + b, the key results are E(Y) = aE(X) + b and Var(Y) = a²Var(X). Note carefully that the standard deviation σₓ = √Var(X) scales by the factor |a|, not a², because it is the square root of the variance.

对于线性变换 Y = aX + b，关键结论是 E(Y) = aE(X) + b 以及 Var(Y) = a²Var(X)。请特别注意，标准差 σₓ = √Var(X) 按 |a| 因子缩放而非 a²，因为标准差是方差的平方根。

---

4. Binomial Distribution | 二项分布

The binomial distribution models the number of successes X in n independent trials, each with constant success probability p. We write X ~ B(n, p). The probability of exactly r successes is given by:

二项分布用于建模 n 次独立试验中的成功次数 X，每次试验的成功概率 p 恒定。记作 X ~ B(n, p)。恰好 r 次成功的概率为：

P(X = r) = ⁿCᵣ pʳ(1 − p)ⁿ⁻ʳ

The mean and variance of a binomial distribution are E(X) = np and Var(X) = np(1 − p). Three conditions must hold for a binomial model: fixed number of trials, independence between trials, and constant probability of success.

二项分布的均值与方差分别为 E(X) = np 和 Var(X) = np(1 − p)。使用二项模型须满足三个条件：试验次数固定、各次试验相互独立、成功概率恒定。

When the question wording includes phrases like “at least 2” or “no more than 3”, you must sum individual probabilities. Using complementary events, such as P(X ≥ 2) = 1 − P(X = 0) − P(X = 1), often simplifies calculations significantly.

当题目出现”至少2次”或”不多于3次”等表述时，你需要将各个概率相加。利用互补事件，如 P(X ≥ 2) = 1 − P(X = 0) − P(X = 1)，通常能大幅简化计算。

Condition | 条件	Required check | 检查要点
Fixed n | n 固定	The number of trials is predetermined | 试验次数预先确定
Independence | 独立性	Outcome of one trial does not affect another | 一次试验结果不影响其他试验
Constant p | p 恒定	Probability of success is the same each trial | 每次试验成功概率相同

---

5. Poisson Distribution | 泊松分布

The Poisson distribution models the number of events occurring in a fixed interval of time or space when events occur independently at a constant average rate λ. We write X ~ Po(λ), and the probability of exactly r events is:

泊松分布用于建模在固定时间或空间区间内发生的事件次数，要求事件独立发生且平均速率 λ 恒定。记作 X ~ Po(λ)，恰好发生 r 次事件的概率为：

P(X = r) = e⁻λ λʳ / r!

For a Poisson distribution, E(X) = λ and Var(X) = λ—the equality of mean and variance is a defining feature. A key extension is that if X ~ Po(λ), then the number of events in an interval of length t follows Po(λt).

泊松分布的 E(X) = λ 且 Var(X) = λ——均值等于方差是其标志性特征。一个重要推广是：若 X ~ Po(λ)，则长度为 t 的区间内事件数服从 Po(λt)。

The Poisson distribution is also used as an approximation to the binomial distribution B(n, p) when n is large (n ≥ 50) and p is small (np ≤ 5), using λ = np. You should state this approximation condition clearly in your exam answer when applying it.

泊松分布也可作为二项分布 B(n, p) 的近似，当 n 较大（n ≥ 50）且 p 较小（np ≤ 5）时，取 λ = np。在考卷中应用该近似时，应清楚说明这一条件。

Another useful property is the sum of independent Poisson variables: if X ~ Po(λ₁) and Y ~ Po(λ₂), then X + Y ~ Po(λ₁ + λ₂). This is frequently tested in two-part exam questions involving combined intervals.

另一个有用的性质是独立泊松变量的和：若 X ~ Po(λ₁) 且 Y ~ Po(λ₂)，则 X + Y ~ Po(λ₁ + λ₂)。该性质常出现在涉及合并区间两问式考题中。

---

6. Normal Distribution | 正态分布

The normal distribution is a continuous probability distribution characterised by its bell-shaped curve and two parameters: the mean μ and the standard deviation σ. We write X ~ N(μ, σ²). The curve is symmetric about the mean, and the total area under the curve equals 1.

正态分布是一种连续概率分布，以其钟形曲线和两个参数为特征：均值 μ 与标准差 σ。记作 X ~ N(μ, σ²)。曲线关于均值对称，曲线下总面积为1。

To find probabilities, you must first standardise to the standard normal variable Z = (X − μ) / σ, which follows N(0, 1). You then use the standard normal distribution table provided in the formula booklet.

求概率时，必须先标准化为标准正态变量 Z = (X − μ) / σ，它服从 N(0, 1)。然后使用公式手册中提供的标准正态分布表。

Z = (X − μ) / σ

Two common contexts in the 9665 Statistics paper are inverse normal problems (finding the value of x given a probability) and the normal approximation to the binomial distribution. In the latter case, for X ~ B(n, p) with np ≥ 5 and n(1 − p) ≥ 5, we approximate X ~ N(np, np(1 − p)).

9665统计学试卷中两类常见题型是逆正态问题（已知概率求 x 值）以及正态近似二项分布。对于后者，当 X ~ B(n, p) 且 np ≥ 5、n(1 − p) ≥ 5 时，可用 X ~ N(np, np(1 − p)) 近似。

A critical technical point: when applying the normal approximation to a discrete binomial distribution, you must use a continuity correction. For example, P(X ≤ 6) becomes P(X < 6.5) in the normal approximation, while P(X ≥ 4) becomes P(X > 3.5).

一个关键的技术要点：将正态近似应用于离散的二项分布时，必须使用连续性修正。例如，P(X ≤ 6) 在正态近似中变为 P(X < 6.5)，而 P(X ≥ 4) 变为 P(X > 3.5)。

---

7. Hypothesis Testing | 假设检验

Hypothesis testing is a formal procedure for making statistical decisions using sample data. You begin by stating the null hypothesis H₀ (the assumption to be tested, usually representing “no change” or “no effect”) and the alternative hypothesis H₁.

假设检验是利用样本数据做出统计决策的规范程序。首先需要陈述原假设 H₀（被检验的假设，通常代表”无变化”或”无效应”）和备择假设 H₁。

For a one-tailed test of a binomial parameter, we test H₀: p = p₀ against H₁: p > p₀ (or p < p₀). For a two-tailed test, H₁: p ≠ p₀. The test statistic is the observed number of successes X, which follows B(n, p₀) under H₀.

对于二项参数的单尾检验，我们检验 H₀: p = p₀ 对 H₁: p > p₀（或 p < p₀）。对于双尾检验，则 H₁: p ≠ p₀。检验统计量为观察到的成功次数 X，在 H₀ 下服从 B(n, p₀)。

The critical region is the set of values of the test statistic that lead to rejection of H₀. Its total probability equals the significance level α (typically 5% or 1%). To determine the critical region for a one-tailed test at significance level α, find the smallest value of c such that P(X ≥ c) ≤ α for an upper-tail test.

拒绝域是使检验统计量落入即拒绝 H₀ 的取值集合，其总概率等于显著性水平 α（通常为5%或1%）。对于单尾检验，确定拒绝域的方法为：在上尾检验中，找到满足 P(X ≥ c) ≤ α 的最小值 c。

You should also be able to calculate the p-value, defined as the probability of obtaining a test statistic at least as extreme as the observed value, assuming H₀ is true. If the p-value < significance level, we reject H₀; otherwise, there is insufficient evidence to reject H₀.

你还应能计算 p 值，其定义为在 H₀ 为真的假设下，得到比观察值至少同样极端的检验统计量的概率。若 p 值小于显著性水平，则拒绝 H₀；否则，没有足够证据拒绝 H₀。

When the sample size is large, the normal approximation may be used for the test statistic. In the 9665 syllabus, you may also perform hypothesis tests for the Poisson mean λ, using the distribution Po(λ₀) under H₀.

当样本量较大时，检验统计量可使用正态近似。在9665大纲中，你还可以对泊松均值 λ 进行假设检验，在 H₀ 下使用 Po(λ₀) 分布。

• Always state H₀ and H₁ in terms of the parameter, not the sample statistic.
• State the distribution of the test statistic under H₀.
• Compare the p-value or test statistic with the significance level explicitly in your conclusion.
• Use the phrase “significant evidence” or “insufficient evidence”—never say “prove” or “accept H₀”.

• 始终用参数而非样本统计量来表述 H₀ 和 H₁。
• 说明检验统计量在 H₀ 下的分布。
• 在结论中明确将 p 值或检验统计量与显著性水平比较。
• 使用”有显著证据”或”证据不足”——绝不要说”证明”或”接受H₀”。

---

8. Correlation and Regression | 相关与回归

Correlation quantifies the strength and direction of a linear relationship between two variables. The product-moment correlation coefficient (PMCC), denoted r, ranges from −1 to +1, where r = +1 indicates perfect positive linear correlation, r = −1 indicates perfect negative correlation, and r = 0 indicates no linear correlation.

相关度量两个变量之间线性关系的强度与方向。积矩相关系数（PMCC），记作 r，取值在 −1 到 +1 之间，其中 r = +1 表示完全正线性相关，r = −1 表示完全负相关，r = 0 表示无线性相关。

Spearman’s rank correlation coefficient rₛ is used when data are ranked or when the relationship is monotonic but not necessarily linear. It is calculated by first ranking both data sets, then applying the PMCC formula to the ranks.

斯皮尔曼等级相关系数 rₛ 适用于数据以等级表示或关系为单调但不一定线性的情形。其计算方法是先将两组数据分别排名，再对排名应用PMCC公式。

rₛ = 1 − (6Σd²) / (n(n² − 1))

where d is the difference between the ranks of each paired observation and n is the number of pairs. This shortcut formula is valid only when all ranks are distinct with no ties.

其中 d 为每对观测的等级差，n 为数据对数。该快捷公式仅在无并列排名时有效。

Regression analysis finds the equation of a line of best fit. The least-squares regression line of y on x is written as y = a + bx, where the slope b = Sxy / Sxx and the intercept a = ȳ − bx̄. Key formulas are:

回归分析用于求最佳拟合直线方程。y 对 x 的最小二乘回归线记作 y = a + bx，其中斜率 b = Sxy / Sxx，截距 a = ȳ − bx̄。关键公式为：

Sxx = Σx² − (Σx)²/n, Sxy = Σxy − (Σx)(Σy)/n

You should understand that the regression line is used for prediction, and that extrapolation (predicting outside the observed range of x) is unreliable. Also note that the regression line of y on x is not the same as the regression line of x on y.

你应理解回归线用于预测，且外推（在观测到的 x 范围之外进行预测）是不可靠的。还需注意，y 对 x 的回归线与 x 对 y 的回归线并不相同。

---

9. Hypothesis Testing for Correlation | 相关系数的假设检验

You may be required to test whether a sample correlation coefficient provides significant evidence of correlation in the population. The null hypothesis is H₀: ρ = 0 (no correlation in the population), and the alternative is H₁: ρ ≠ 0 for a two-tailed test.

你可能需要检验样本相关系数是否提供了总体存在相关的显著证据。原假设为 H₀: ρ = 0（总体无相关），双尾检验的备择假设为 H₁: ρ ≠ 0。

Rather than computing a test statistic from scratch, the 9665 syllabus typically provides a table of critical values for the PMCC and Spearman’s coefficient. You compare the absolute value of your calculated coefficient |r| with the critical value at the given significance level and sample size n.

9665大纲通常直接提供PMCC和斯皮尔曼系数的临界值表，而非要求从头计算检验统计量。你需要将计算出的系数绝对值 |r| 与给定显著性水平和样本量 n 下的临界值进行比较。

If |r| > critical value, the result is significant and we reject H₀, concluding that there is evidence of correlation in the population. If |r| ≤ critical value, the result is not significant and there is insufficient evidence of correlation.

若 |r| 大于临界值，则结果显著，拒绝 H₀，得出结论：有证据表明总体中存在相关。若 |r| ≤ 临界值，则结果不显著，没有足够证据证明存在相关。

---

10. Exam Technique and Common Pitfalls | 考试技巧与常见误区

In the OxfordAQA 9665 Statistics paper, marks are awarded not only for correct calculations but also for clear reasoning and correct statistical language. Always define your random variable explicitly, for example “Let X be the number of defective items in a sample of 20.”

在牛津AQA 9665统计学试卷中，得分不仅来自正确的计算，还来自清晰的推理和规范的统计语言。务必明确定义随机变量，例如”设 X 为20件样本中的次品数”。

When using the normal distribution table, be careful with symmetry properties: P(Z < −z) = 1 − P(Z < z). For finding P(a < Z < b), compute P(Z < b) − P(Z < a). When the table gives P(Z < z), always draw a small sketch of the standard normal curve and shade the relevant area.

使用正态分布表时，注意对称性质：P(Z < −z) = 1 − P(Z < z)。求 P(a < Z < b) 时，计算 P(Z < b) − P(Z < a)。当表格给出 P(Z < z) 时，一定要画出标准正态曲线的小草图并标出相关区域。

Common pitfalls to avoid:

需要避免的常见误区：

• Forgetting the continuity correction when using a normal approximation to a discrete distribution.
• Confusing Var(X) with E(X²)—always subtract [E(X)]².
• Stating H₁ after the test has been performed instead of before.
• Using the Poisson approximation to the binomial without checking np ≤ 5 and n ≥ 50.
• Ranking data incorrectly when ties exist in Spearman’s coefficient.

• 对离散分布使用正态近似时忘记连续性修正。
• 混淆 Var(X) 与 E(X²)——务必减去 [E(X)]²。
• 在检验完成之后才写 H₁ 而非事先陈述。
• 未检查 np ≤ 5 和 n ≥ 50 就使用泊松近似二项分布。
• 斯皮尔曼系数计算中处理并列排名时出错。

Finally, manage your time wisely. In a typical AS paper, allocate approximately one and a half minutes per mark. Write down the distribution you are using, the parameters, and the key formula before performing calculations—method marks are often available even if arithmetic goes wrong.

最后，合理分配时间。在典型的AS试卷中，每分大约分配一分半钟。在计算之前写出所用分布、参数和关键公式——即使计算有误，方法分通常仍然可得。

---

Final Checklist | 考前清单

Before entering the examination, ensure you can recall from memory the mean and variance of the binomial and Poisson distributions, the standardisation formula for the normal distribution, the least-squares regression formulas, and the structure of a hypothesis test from hypotheses through to conclusion.

进入考场前，请确保你能凭记忆写出二项分布和泊松分布的均值与方差、正态分布标准化公式、最小二乘回归公式，以及从假设到结论的完整假设检验流程。

Practise past-paper questions under timed conditions and mark them strictly. The Statistics paper rewards precision in both calculation and communication—use the correct notation, show your working, and always answer in the context of the question.

在限时条件下练习历年真题并严格评分。统计学试卷既奖励计算的精确性，也奖励表述的规范性——使用正确的符号、展示计算过程，并始终结合题目背景作答。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）