Hypothesis Testing for IB & AQA Maths | IB与AQA数学：假设检验考点精讲

📚 Hypothesis Testing for IB & AQA Maths | IB与AQA数学：假设检验考点精讲

Hypothesis testing is one of the most important topics in both IB Mathematics and AQA International A-Level Maths. It provides a formal framework for making decisions based on sample data. Whether you are dealing with binomial probabilities or normal distributions, understanding the logic of null and alternative hypotheses, significance levels, p-values, and types of errors is essential for scoring high marks in your statistics paper. This revision guide covers every key concept you need, with step-by-step explanations and side-by-side English‑Chinese content to help bilingual learners master the material.

假设检验是 IB 数学和 AQA 国际 A-Level 数学中最重要的主题之一。它提供了一个基于样本数据作出决策的正式框架。无论你处理的是二项分布概率还是正态分布，理解零假设与备择假设、显著性水平、p 值以及两类错误的逻辑，对于在统计考试中取得高分至关重要。本考点精讲涵盖每一个关键概念，配有逐步讲解和中英对照内容，帮助双语学习者完全掌握。

1. What is Hypothesis Testing? | 什么是假设检验？

A hypothesis test is a statistical method used to decide whether there is enough evidence in a sample to reject a stated claim about a population parameter. In the context of IB and AQA maths, you will typically test a proportion p in a binomial setting or a population mean μ when the population variance is known. The process always starts by assuming the null hypothesis is true, then examines how extreme the observed sample statistic is under that assumption.

假设检验是一种统计方法，用于判断样本中是否有足够证据拒绝关于总体参数的某个声明。在 IB 和 AQA 数学中，你通常会检验二项分布中的比例 p，或在已知总体方差时检验总体均值 μ。该过程总是先假设零假设成立，然后考察在该假设下观测到的样本统计量有多极端。

Every hypothesis test follows a structured sequence: state the hypotheses, choose a significance level, calculate the test statistic or find the p-value, and draw a conclusion in the context of the problem. No marks are earned by jumping straight to the numbers – you must show the reasoning clearly.

每个假设检验都遵循结构化的步骤：陈述假设、选择显著性水平、计算检验统计量或求 p 值，并针对题目背景写出结论。直接跳到数字是不会得分的——必须清晰地展示推理过程。

2. Null and Alternative Hypotheses | 零假设与备择假设

The null hypothesis, denoted H₀, is the statement that there is no effect, no difference, or that the population parameter equals a specific value. The alternative hypothesis, H₁ (or Hₐ), is what you believe might be true instead. The test aims to see if the data provide sufficient evidence to reject H₀ in favour of H₁.

零假设，记为 H₀，是指没有效应、没有差异或总体参数等于某一特定值的陈述。备择假设 H₁（或 Hₐ）则是你认为可能成立的另一种陈述。检验的目的是考察数据是否提供足够证据，拒绝 H₀ 而支持 H₁。

For a binomial test concerning a population proportion p, a typical set of hypotheses might be:

H₀: p = 0.3
H₁: p > 0.3 (one‑tailed)

对于关于总体比例 p 的二项检验，一组典型的假设可能是：

H₀: p = 0.3
H₁: p > 0.3 （单尾）

Always define p clearly before writing the hypotheses. For a normal distribution test on the mean μ, you might write:

在写假设之前，务必明确定义 p。对于正态分布均值的检验，可以这样写：

H₀: μ = 50
H₁: μ ≠ 50 (two‑tailed)

3. Significance Level and Rejection Region | 显著性水平与拒绝域

The significance level, α, is the probability of rejecting a true null hypothesis – a Type I error. Common choices in IB and AQA exams are α = 0.05, 0.01 or 0.10. The rejection region (also called the critical region) is the set of values of the test statistic that would lead to rejecting H₀.

显著性水平 α 是拒绝一个真实的零假设的概率，即第一类错误。IB 和 AQA 考试中常见的选择是 α = 0.05, 0.01 或 0.10。拒绝域（也称临界域）是导致拒绝 H₀ 的检验统计量的取值集合。

For a one‑tailed test with α = 0.05 in a binomial setting, the rejection region might be all values X ≥ c, where c is the smallest integer such that P(X ≥ c) ≤ 0.05 under H₀. The boundary value c is called the critical value.

对于二项分布且 α = 0.05 的单尾检验，拒绝域可能是所有满足 X ≥ c 的值，其中 c 是在 H₀ 下满足 P(X ≥ c) ≤ 0.05 的最小整数。边界值 c 被称为临界值。

4. One‑Tailed vs Two‑Tailed Tests | 单尾检验与双尾检验

A one‑tailed test is used when the alternative hypothesis specifies a direction (greater than or less than). A two‑tailed test is used when you are only interested in whether the parameter differs from the hypothesised value, regardless of direction. The choice affects the location of the rejection region and how the significance level is split.

当备择假设指定了方向（大于或小于）时，使用单尾检验。当你只关心参数是否与假设值不同而不论方向时，使用双尾检验。这一选择会影响拒绝域的位置以及显著性水平的分配方式。

In a two‑tailed test at α = 0.05, the total rejection region is split equally between the two tails, so each tail carries an area of 0.025. In a one‑tailed test, the entire 0.05 is placed in the single tail of interest. Misidentifying the type of test is a common mistake that can cost you all the marks in a hypothesis‑testing question.

在 α = 0.05 的双尾检验中，整个拒绝域被均分到两个尾部，因此每个尾部面积为 0.025。在单尾检验中，整个 0.05 被放在所关注的单一尾部。错误识别检验类型是一个常见错误，可能让你在假设检验题目中失去全部分数。

5. p‑Value: Definition and Interpretation | p 值：定义与解释

The p‑value is the probability of obtaining a test statistic at least as extreme as the one observed, assuming that H₀ is true. A small p‑value (e.g., p < 0.05) indicates strong evidence against the null hypothesis, leading to its rejection. A large p‑value suggests insufficient evidence to reject H₀.

p 值是在 H₀ 成立的条件下，得到至少与观测值一样极端的检验统计量的概率。较小的 p 值（例如 p < 0.05）表明有足够强的证据反对零假设，从而拒绝它。较大的 p 值则表明没有足够证据拒绝 H₀。

For a one‑tailed binomial test where X ~ B(20, 0.3) and the observed value is x = 10, the p‑value is P(X ≥ 10) under H₀. This is calculated using binomial cumulative probabilities. In AQA exams, you are expected to use tables or statistical functions on your calculator.

对于 X ~ B(20, 0.3) 且观测值为 x = 10 的单尾二项检验，p 值为在 H₀ 下的 P(X ≥ 10)。这需要使用二项累积概率计算。在 AQA 考试中，你应该使用表格或计算器上的统计功能来求值。

6. Critical Value Approach | 临界值法

Instead of computing a p‑value, you may find the critical region directly. Determine the critical value(s) such that the probability of falling into the critical region is exactly α (or as close as possible without exceeding it). If the test statistic falls inside the critical region, reject H₀.

除了计算 p 值，你也可以直接找出临界域。确定一个或多个临界值，使得落入临界域的概率恰好为 α（或尽可能接近而不超过）。若检验统计量落入临界域，则拒绝 H₀。

Suppose you have H₀: p = 0.25, H₁: p < 0.25, n = 20, α = 0.05. Find the largest integer r such that P(X ≤ r) ≤ 0.05 under H₀. If the observed reading is less than or equal to r, H₀ is rejected. This method is especially straightforward with binomial distribution tables.

假设 H₀: p = 0.25, H₁: p < 0.25, n = 20, α = 0.05。找出最大整数 r 使得 H₀ 下 P(X ≤ r) ≤ 0.05。如果观测值小于或等于 r，则拒绝 H₀。这种方法在使用二项分布表时特别直接。

7. Hypothesis Testing for Binomial Distributions | 二项分布的假设检验

Binomial hypothesis tests are the backbone of AQA and IB statistics when dealing with proportions. Here X ~ B(n, p₀) under H₀. The sample provides a count of successes. Both one‑tailed and two‑tailed tests appear frequently. You must check that the data satisfy the binomial conditions: fixed number of trials, two possible outcomes, constant probability, and independent trials.

二项假设检验是 AQA 和 IB 统计中处理比例时的核心内容。在 H₀ 下，X ~ B(n, p₀)。样本给出成功次数。单尾和双尾检验都经常出现。你必须确认数据满足二项条件：试验次数固定、两种可能结果、概率恒定以及试验独立。

When performing a two‑tailed binomial test, you need to compare the total probability of the two tails. For example, if you observe x successes, you might look at the smaller tail by symmetry or simply double the appropriate one‑tailed p‑value, provided the distribution is not too skewed. The conclusion must be stated in words, relating back to the original claim.

在进行双尾二项检验时，你需要考虑两个尾部的总概率。例如，如果观测到 x 次成功，你可以根据对称性考虑较小的一边，或直接将相应的单尾 p 值乘以 2，只要分布不太偏斜。结论必须用文字陈述，并与原始声明联系起来。

8. Hypothesis Testing for Normal Distributions (z‑test) | 正态分布的假设检验（z 检验）

When the population is normally distributed or the sample size is large, and the population variance σ² is known, you use a z‑test for the mean. The test statistic is:

Z = (x̄ − μ₀) / (σ/√n)

当总体服从正态分布或样本量较大，且总体方差 σ² 已知时，你使用均值 z 检验。检验统计量为：

Z = (x̄ − μ₀) / (σ/√n)

Compare this Z value against critical values from the standard normal distribution (e.g., ±1.96 for a two‑tailed test at α = 0.05). Alternatively, calculate the p‑value using Φ(|Z|). Always check that conditions are met: random sample, known variance, and approximate normality justified by the Central Limit Theorem if n ≥ 30.

将该 Z 值与标准正态分布的临界值进行比较（例如 α = 0.05 双尾检验的临界值为 ±1.96）。或者利用正态分布表计算 p 值。务必检查条件是否满足：随机样本、已知方差，以及若 n ≥ 30 可由中心极限定理保证近似正态性。

9. Type I and Type II Errors | 第一类错误与第二类错误

A Type I error occurs when the null hypothesis is true but the test rejects it. The probability of this error is exactly α, the significance level. A Type II error happens when the null hypothesis is false but we fail to reject it; its probability is denoted by β. The two errors are inversely related for a fixed sample size.

第一类错误发生在零假设为真但检验却拒绝了它的情况下。这类错误的概率恰好是显著性水平 α。第二类错误发生在零假设不真但我们未能拒绝它的情况下，其概率记为 β。对于固定的样本量，这两类错误是此消彼长的关系。

Reducing α makes it harder to reject H₀, which lowers the chance of a Type I error but increases the risk of a Type II error. Exam questions often ask you to interpret the meaning of these errors in context, so always link the error description to the specific situation in the problem.

降低 α 会增加拒绝 H₀ 的难度，从而减少第一类错误的概率，但会增加第二类错误的风险。考试题目经常要求你在具体情境中解释这些错误的含义，因此务必将错误描述与题目中的具体情况联系起来。

10. Power of a Test | 检验的功效

The power of a hypothesis test is the probability of correctly rejecting a false null hypothesis. It equals 1 − β. A powerful test is more likely to detect a real effect when it exists. Power depends on the sample size, the significance level, and the true value of the parameter under H₁.

假设检验的功效是指正确拒绝一个不真的零假设的概率，它等于 1 − β。功效高的检验更有可能在真实效应存在时检测出它。功效取决于样本量、显著性水平以及 H₁ 下的参数真实值。

In AQA and IB contexts, you may be asked to calculate the power of a test for a specific alternative value of p or μ. For a binomial test, you find P(X in critical region | alternative value). Larger sample sizes and larger effect sizes increase power. Understanding power helps you design better experiments.

在 AQA 和 IB 的考试中，你可能会被要求针对某个具体的备择参数值（如 p 或 μ）计算检验的功效。对于二项检验，计算 P(X 落入拒绝域 | 备择参数值)。更大的样本量和更大的效应量会提高功效。理解功效有助于设计更好的实验。

11. Writing Conclusions and Common Pitfalls | 写出结论与常见陷阱

A strong conclusion always states whether there is sufficient evidence to reject H₀, references the significance level used, and is phrased in the context of the original problem. Do not say “accept H₀”; instead, say “there is insufficient evidence to reject H₀” because hypothesis testing never proves a null hypothesis true.

一个严谨的结论总是说明是否有足够证据拒绝 H₀，提及所使用的显著性水平，并用原题背景来表述。不要说“接受 H₀”，而应说“没有足够证据拒绝 H₀”，因为假设检验永远不能证明零假设为真。

Common mistakes include: misreading one‑tailed vs two‑tailed, using the wrong p‑value calculation, confusing the sample statistic with the population parameter, and forgetting to define the test statistic. Practice extensively with past paper questions from both IB and AQA to internalise the flow of a hypothesis test.

常见错误包括：误判单尾与双尾、使用错误的 p 值计算、混淆样本统计量与总体参数，以及忘记定义检验统计量。大量练习 IB 和 AQA 过往真题，将假设检验的流程内化于心，是避免这些错误的关键。

Published by TutorHao | Maths Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Hypothesis Testing for IB & AQA Maths | IB与AQA数学：假设检验考点精讲