Hypothesis Testing for IB CCEA Mathematics: Key Revision Points | IB CCEA 数学假设检验考点精讲

📚 Hypothesis Testing for IB CCEA Mathematics: Key Revision Points | IB CCEA 数学假设检验考点精讲

Hypothesis testing is a core topic in both IB Mathematics and CCEA A-level Mathematics, appearing regularly in Statistics components. Whether you are preparing for IB Analysis & Approaches, Applications & Interpretation, or CCEA’s pure and applied modules, mastering the logic of hypothesis testing, recognising which test to use, and interpreting p-values and critical regions are essential skills. This revision guide walks you through the fundamental concepts, step-by-step procedures, and common pitfalls, with clear bilingual explanations to strengthen your understanding and exam technique.

假设检验是 IB 数学与 CCEA A-level 数学统计部分的核心主题，经常出现在试卷中。无论你准备的是 IB 分析与方法、应用与解释，还是 CCEA 的纯数与应用模块，掌握假设检验的逻辑、选择正确的检验方法、解读 p 值与拒绝域都是必须掌握的技能。本精讲用中英双语带你梳理基本概念、分步步骤与常见陷阱，帮你夯实理解并提升应试技巧。

1. Introduction to Hypothesis Testing | 假设检验简介

Hypothesis testing is a statistical method for making decisions about a population parameter based on sample data. The process begins by assuming a null hypothesis is true, and then we examine whether the observed sample provides sufficient evidence to reject it. In IB and CCEA syllabi, this technique is applied to proportions, means, and associations, using distributions such as the binomial, normal, t, and chi-squared.

假设检验是一种基于样本数据对总体参数做出判断的统计方法。整个过程从假设原假设为真开始，然后检验观测到的样本是否提供了足够的证据来拒绝它。在 IB 和 CCEA 的课程中，这一方法被用于比例、均值和关联性检验，涉及二项分布、正态分布、t 分布和卡方分布。

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

The null hypothesis, denoted H₀, is a statement of no effect or no difference. The alternative hypothesis, H₁, represents the claim we seek evidence for. For example, in testing whether a coin is fair, H₀: p = 0.5 and H₁: p ≠ 0.5. H₁ can be one-sided (p < 0.5) or two-sided. The conclusion of a hypothesis test is always phrased in terms of H₀: either we reject H₀ or we do not reject H₀. We never say we “accept” H₀.

原假设，记为 H₀，是一种陈述“无效应”或“无差异”的命题。备择假设 H₁ 则代表我们试图寻找证据支持的命题。例如，检验一枚硬币是否均匀时，H₀: p = 0.5，H₁: p ≠ 0.5。H₁ 可以是单侧的（p < 0.5）或双侧的。假设检验的结论总是围绕 H₀ 表述：我们拒绝 H₀，或者不拒绝 H₀。我们绝不说“接受”H₀。

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

The significance level α (alpha) is the probability of rejecting H₀ when it is actually true. Common values are 0.01, 0.05, or 0.10. The rejection (or critical) region consists of all values of the test statistic that lead to rejecting H₀. If the test statistic falls inside the rejection region, we reject H₀. The boundaries of this region are called critical values, which depend on α and the distribution of the test statistic.

显著性水平 α 是当 H₀ 实际为真时错误拒绝它的概率。常用取值有 0.01、0.05 或 0.10。拒绝域（临界域）包含所有导致拒绝 H₀ 的检验统计量取值。如果检验统计量落入了拒绝域，我们就拒绝 H₀。拒绝域的边界值称为临界值，它取决于 α 以及检验统计量的分布。

4. Test Statistic and p-value | 检验统计量与 p 值

A test statistic summarises the sample data in a single value, used to decide whether to reject H₀. In IB and CCEA questions you will encounter statistics such as Z, t, χ², or simply the number of successes in a binomial test. The p-value is the probability of obtaining a result at least as extreme as the one observed, assuming H₀ is true. If p-value ≤ α, we reject H₀. The p-value method is often preferred because it provides a measure of the strength of evidence against H₀.

检验统计量用一个数值概括样本数据，用于决定是否拒绝 H₀。在 IB 和 CCEA 试题中你会遇到 Z 值、t 值、χ² 值，或者二项检验中直接使用成功次数。p 值是在 H₀ 为真的前提下，获得至少与观测结果一样极端的结果的概率。如果 p 值 ≤ α，我们拒绝 H₀。p 值法往往更受青睐，因为它能衡量反对 H₀ 的证据强度。

5. One-tailed and Two-tailed Tests | 单尾与双尾检验

A one-tailed test is used when H₁ specifies a direction, for example H₁: μ > 20. The entire rejection region lies in one tail of the sampling distribution. A two-tailed test is used when H₁ does not specify a direction, such as H₁: μ ≠ 20; the rejection region is split equally between the two tails. Choosing the correct tail is crucial because it affects the critical value and the p-value calculation. Exam questions often require you to state whether the test is one- or two-tailed and to justify your choice.

当 H₁ 指明方向时，例如 H₁: μ > 20，我们使用单尾检验，整个拒绝域落在抽样分布的一个尾部。当 H₁ 没有指明方向时，例如 H₁: μ ≠ 20，我们使用双尾检验，拒绝域被均等地分到两个尾部。选择正确的尾型至关重要，因为它影响临界值和 p 值的计算。考试题经常要求你说明检验是单尾还是双尾并解释理由。

6. Type I and Type II Errors | 第一类与第二类错误

A Type I error occurs when H₀ is true but we reject it. The probability of a Type I error is exactly α. A Type II error occurs when H₀ is false but we fail to reject it; its probability is denoted β. The power of a test is 1 – β, the probability of correctly rejecting a false H₀. While IB and CCEA rarely ask for detailed calculations of β, you must understand these concepts and be able to identify the type of error in a given context.

当 H₀ 为真但我们拒绝了它，就犯了第一类错误，其概率正是 α。当 H₀ 为假但我们没有拒绝它，则犯了第二类错误，其概率记为 β。检验的功效是 1 – β，即正确拒绝错误 H₀ 的概率。虽然 IB 和 CCEA 很少要求详细计算 β，但你必须理解这些概念并能在具体情境中识别错误类型。

Decision	H₀ True	H₀ False
Reject H₀	Type I error (α)	Correct (Power)
Do not reject H₀	Correct	Type II error (β)

表格：假设检验中的决策与错误类型。横排为“决策”，竖排为“H₀ 真/假”。

---

7. Binomial Hypothesis Testing | 二项假设检验

Binomial tests appear frequently in IB and CCEA papers when a population proportion is of interest. Assume X ~ B(n, p). We test H₀: p = p₀ by calculating the probability of obtaining the observed number of successes, or a more extreme value, under H₀. For a one-tailed test we find P(X ≤ x) or P(X ≥ x). For a two-tailed test we find the sum of probabilities in both tails that are as extreme as the observed. If this probability (p-value) ≤ α, reject H₀. You may also use critical regions: find the largest r such that P(X ≤ r) ≤ α/2, etc., and compare the observed count.

二项检验在 IB 和 CCEA 试卷中经常出现，用于总体比例的检验。假设 X ~ B(n, p)。检验 H₀: p = p₀ 的方法是计算在 H₀ 下得到观测成功次数或更极端值的概率。对于单尾检验，求 P(X ≤ x) 或 P(X ≥ x)。对于双尾检验，求两个尾部中与观测值一样极端的所有概率之和。如果该概率（p 值）≤ α，则拒绝 H₀。你也可以使用临界域：找出最大的 r 使得 P(X ≤ r) ≤ α/2 等，再与观测次数比较。

8. Normal Hypothesis Testing | 正态假设检验

When the sample size is large or the population is normally distributed, we test a population mean using the Z-statistic. The test statistic is Z = (x̄ – μ₀) / (σ/√n). For a known σ, compare Z with critical values from the standard normal distribution, e.g. ±1.96 for a two-tailed test at α = 0.05. If σ is unknown and the sample size is small, a t-test should be used, but CCEA sometimes uses Z when σ is given. Check your exam formula book for exact procedures. The p-value is found using normal tables.

当样本容量较大或者总体服从正态分布时，我们使用 Z 统计量检验总体均值。检验统计量为 Z = (x̄ – μ₀) / (σ/√n)。若 σ 已知，将 Z 与标准正态分布的临界值比较，例如 α = 0.05 的双尾检验临界值为 ±1.96。若 σ 未知且样本量较小，应使用 t 检验，但 CCEA 有时会在给定 σ 时用 Z。查阅你的考试公式表以获得精确步骤。p 值通过查正态分布表得出。

Z = (x̄ – μ₀) / (σ / √n)

---

9. t-test for Population Mean | 总体均值的 t 检验

When the population standard deviation σ is unknown and we estimate it with the sample standard deviation s, the test statistic follows a t-distribution with n – 1 degrees of freedom. This is standard in IB Applications & Interpretation and some CCEA applied units. The test statistic is t = (x̄ – μ₀) / (s/√n). You compare the calculated t with critical values from the t-table for the given df and α, or find the p-value using technology. Remember that as n increases, the t-distribution approaches the normal distribution.

当总体标准差 σ 未知，我们用样本标准差 s 进行估计时，检验统计量服从自由度为 n – 1 的 t 分布。这在 IB 应用与解释以及 CCEA 某些应用单元中是标准做法。检验统计量为 t = (x̄ – μ₀) / (s/√n)。将计算出的 t 值与给定自由度和 α 的 t 分布临界值进行比较，或利用技术工具计算 p 值。记住，当 n 增大时，t 分布趋近于正态分布。

t = (x̄ – μ₀) / (s / √n), df = n – 1

---

10. Chi-squared Test for Independence | 独立性卡方检验

The chi-squared (χ²) test for independence examines whether two categorical variables are associated. The observed frequencies are compared with expected frequencies calculated under the assumption of independence. The test statistic is χ² = Σ (O – E)² / E. It follows a χ² distribution with (r – 1)(c – 1) degrees of freedom, where r and c are the number of rows and columns in the contingency table. A large χ² value indicates a discrepancy between observed and expected counts, leading to rejection of H₀: the variables are independent.

独立性卡方检验用于考察两个分类变量是否有关联。观测频数与其在独立假设下的期望频数进行比较。检验统计量为 χ² = Σ (O – E)² / E。它服从自由度为 (r – 1)(c – 1) 的 χ² 分布，其中 r 和 c 为列联表的行数和列数。χ² 值很大说明观测值与期望值之间存在较大差异，从而导致拒绝 H₀（变量相互独立）。

χ² = Σ (O – E)² / E

---

11. Common Mistakes and Exam Tips | 常见错误与应试技巧

Typical errors include confusing one-tailed and two-tailed tests, misinterpreting the p-value as the probability that H₀ is true, or failing to define the hypotheses precisely before starting a calculation. In binomial tests, students often forget to include the “more extreme” probabilities correctly. Always state your conclusion in context: “There is sufficient evidence to reject H₀ at the 5% level” rather than just “Reject H₀.” In CCEA papers, marks are allocated for clear statements of H₀, H₁, test statistic, p-value/critical value, and a contextual conclusion.

典型错误包括：混淆单尾与双尾检验、将 p 值误解为 H₀ 为真的概率、以及在开始计算前没有准确定义假设。在二项检验中，学生常忘记正确纳入“更极端”的概率。请务必在上下文中给出结论：“在 5% 的显著性水平下，有足够证据拒绝 H₀”，而不只是“拒绝 H₀”。在 CCEA 试卷中，明确陈述 H₀、H₁、检验统计量、p 值/临界值以及符合语境的结论都能得到相应分值。

• Write down H₀ and H₁ before calculating

  计算前先写下 H₀ 和 H₁

• Sketch the distribution and mark the rejection region

  画出分布示意图并标出拒绝域

• Check whether the test is one- or two-tailed

  先确认是单尾还是双尾检验

• Use the correct formula sheet for critical values

  使用正确的公式表查找临界值

---

12. Summary | 总结

Hypothesis testing is a logical framework that connects sample data to conclusions about a population. Whether you evaluate a binomial proportion, a normal mean, or a chi-squared association, the structure remains consistent: state hypotheses, choose significance level, compute test statistic, find p-value or compare with critical region, and write a contextual conclusion. IB and CCEA both value clear methodology and correct interpretation. Reviewing the worked examples from your specification and practising a variety of past paper questions will build the confidence to handle any hypothesis testing scenario under exam conditions.

假设检验是一个将样本数据与总体结论联系起来的逻辑框架。无论你是检验二项比例、正态均值还是卡方关联性，其结构都是一致的：陈述假设、选择显著性水平、计算检验统计量、得出 p 值或与临界域比较、并写出融入背景的结论。IB 和 CCEA 都看重清晰的方法和正确的解读。回顾教学大纲中的例题，并练习各类历年真题，将使你在考试中从容应对任何假设检验情境。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）