Hypothesis Testing for IGCSE OCR Maths | IGCSE OCR 数学：假设检验考点精讲

📚 Hypothesis Testing for IGCSE OCR Maths | IGCSE OCR 数学：假设检验考点精讲

Hypothesis testing is a statistical method used to make decisions about a population parameter based on sample data. In the IGCSE OCR specification, you will learn to set up null and alternative hypotheses, carry out tests using binomial probabilities, and interpret results in context. This guide covers every key point you need to master for your exam.

假设检验是一种基于样本数据对总体参数进行判断的统计方法。在 IGCSE OCR 考纲中，你将学习如何设定原假设和备择假设、利用二项分布概率进行检验，并在实际问题中解释结果。本指南涵盖了你需要掌握的所有核心考点。

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

Hypothesis testing is a formal procedure that helps us decide whether there is enough evidence to support a particular claim about a population. We start with a tentative assumption, called the null hypothesis, and then examine sample data to see if it contradicts that assumption.

假设检验是一个正规的程序，帮助我们判断是否有足够证据支持关于总体的某个主张。我们从一个试探性假设（原假设）出发，然后检查样本数据是否与之矛盾。

In simple terms, we ask: ‘Could the observed result have happened by chance, or is it so unusual that we must reject our starting assumption?’ This process is widely used in science, medicine, and quality control.

简单来说，我们问：“观察到的结果可能只是偶然发生的吗？还是它太不寻常了，我们必须推翻最初的假设？” 这一过程广泛应用于科学、医学和质量控制中。

2. Stating the Hypotheses | 陈述假设

Every hypothesis test involves two competing statements: the null hypothesis (H₀) and the alternative hypothesis (H₁). The null hypothesis usually states that there is no effect or no difference, while the alternative hypothesis represents what we are trying to prove.

每个假设检验都包含两个对立的陈述：原假设（H₀）和备择假设（H₁）。原假设通常陈述没有影响或没有差异，而备择假设代表我们试图证明的内容。

• H₀: The status quo, e.g. ‘the coin is fair’ – in symbols, H₀: p = 0.5.

  H₀：现状，例如“硬币是公平的”——用符号表示为 H₀: p = 0.5。

• H₁: The claim we suspect might be true, e.g. ‘the coin is biased towards heads’ – H₁: p > 0.5.

  H₁：我们怀疑可能成立的断言，例如“硬币偏向正面”——H₁: p > 0.5。

In IGCSE problems, p is often a binomial probability of success. Hypotheses are always about the population parameter, never about the sample statistic.

在 IGCSE 题目中，p 通常是二项分布中的成功概率。假设始终针对总体参数，而非样本统计量。

3. Significance Level | 显著性水平

The significance level, denoted by α (alpha), is the probability threshold below which we consider the observed result to be statistically significant. It represents the maximum acceptable risk of wrongly rejecting H₀.

显著性水平，记作 α（alpha），是我们认为观察结果具有统计显著性的概率阈值。它表示错误地拒绝 H₀ 的可接受最大风险。

Common values for α in IGCSE exams are 5% (0.05) and 1% (0.01). If the probability of obtaining the observed result (or more extreme) under H₀ is less than α, we reject H₀.

在 IGCSE 考试中，α 的常见取值为 5% (0.05) 和 1% (0.01)。如果在 H₀ 成立的条件下得到观察结果（或更极端结果）的概率小于 α，我们拒绝 H₀。

Think of α as your ‘cut-off’ for deciding whether something is too unlikely to have occurred by chance alone.

可以将 α 理解为判断事情是否不可能仅凭偶然发生的“分界线”。

4. Test Statistic and Critical Region | 检验统计量与拒绝域

The test statistic is the value calculated from sample data. In OCR IGCSE, it is usually the number of successes observed, X, from a binomial distribution X ~ B(n, p).

检验统计量是从样本数据计算得出的值。在 OCR IGCSE 中，通常是从二项分布 X ~ B(n, p) 中观测到的成功次数 X。

The critical region (or rejection region) is the set of values of the test statistic that would lead to rejecting H₀. Its boundary values are called critical values.

拒绝域（或拒绝区域）是检验统计量中会导致拒绝 H₀ 的数值集合。其边界值称为临界值。

For example, if n = 20, H₀: p = 0.5, H₁: p > 0.5 at α = 0.05, we find the smallest x such that P(X ≥ x) ≤ 0.05. That x is the critical value, and the region x, x+1, …, 20 is the critical region.

例如，若 n = 20，H₀: p = 0.5，H₁: p > 0.5，α = 0.05，我们找到最小的 x 使得 P(X ≥ x) ≤ 0.05。该 x 就是临界值，而区域 x, x+1, …, 20 即为拒绝域。

5. One-Tailed vs. Two-Tailed Tests | 单尾与双尾检验

A one-tailed test is used when the alternative hypothesis specifies a direction: H₁: p > value or H₁: p < value. All the significance level is concentrated in one tail of the distribution.

当备择假设指定了方向（H₁: p > 某值 或 H₁: p < 某值）时，使用单尾检验。所有显著性水平集中在分布的一尾。

A two-tailed test is used when we are simply testing for any difference: H₁: p ≠ value. Here, α is split equally between the two tails, so we look for outcomes that are either much higher or much lower than expected under H₀.

当我们只是检验是否存在差异（H₁: p ≠ 某值）时，使用双尾检验。此时 α 平均分配到两尾，因此我们要寻找在 H₀ 下过高或过低的结果。

In IGCSE, the problem statement often hints at the type of test: words like ‘higher’, ‘improved’ suggest a one-tailed test, while ‘changed’ or ‘different’ suggests a two-tailed test.

在 IGCSE 中，题目表述常暗示检验的类型：“更高”、“改善”等词语提示单尾检验，“改变”或“不同”提示双尾检验。

6. Using Critical Values and Rejection Rules | 使用临界值与拒绝规则

Once the critical value(s) are found, the decision rule is simple: if the test statistic falls inside the critical region, we reject H₀; otherwise, we do not reject H₀.

一旦找到临界值，决策规则很简单：如果检验统计量落入拒绝域，则拒绝 H₀；否则，不拒绝 H₀。

For a one-tailed test (H₁: p > p₀), find the smallest critical value c where P(X ≥ c) ≤ α. Then the critical region is X ≥ c. For H₁: p < p₀, find the largest c with P(X ≤ c) ≤ α; region is X ≤ c.

对于单尾检验（H₁: p > p₀），找到最小的临界值 c 使得 P(X ≥ c) ≤ α。拒绝域为 X ≥ c。对于 H₁: p < p₀，找到最大的 c 使得 P(X ≤ c) ≤ α；拒绝域为 X ≤ c。

For two-tailed tests with symmetric binomial (p=0.5), halve α for each tail. Find c₁ and c₂ such that P(X ≤ c₁) ≤ α/2 and P(X ≥ c₂) ≤ α/2. Critical region: X ≤ c₁ or X ≥ c₂.

对于对称二项分布（p=0.5）的双尾检验，将 α 对半分配至两尾。找到 c₁ 和 c₂ 使得 P(X ≤ c₁) ≤ α/2 且 P(X ≥ c₂) ≤ α/2。拒绝域为 X ≤ c₁ 或 X ≥ c₂。

Always state your conclusion in the context of the problem: ‘There is sufficient evidence at the 5% level to reject the null hypothesis, suggesting the coin is biased towards heads.’

始终在问题情境中陈述结论：“在 5% 显著性水平下，有足够证据拒绝原假设，说明该硬币偏向正面。”

7. The p-Value Method | p 值方法

The p-value is the probability, under H₀, of obtaining a result at least as extreme as the one observed. It provides an alternative way to make the decision: if p-value ≤ α, reject H₀.

p 值是在 H₀ 下，得到至少与观测结果同样极端的结果的概率。它提供了另一种决策方式：若 p 值 ≤ α，则拒绝 H₀。

For H₁: p > p₀, the p-value = P(X ≥ observed value). For H₁: p < p₀, p-value = P(X ≤ observed value). For a two-tailed test, the p-value is often twice the smaller tail probability (for symmetrical distributions).

对于 H₁: p > p₀，p 值 = P(X ≥ 观测值)。对于 H₁: p < p₀，p 值 = P(X ≤ 观测值)。对于双尾检验，p 值通常为较小尾概率的两倍（对于对称分布）。

In exams, the p-value method is often quicker when the observed value is given, as you only need to compute the tail probability and compare with α.

在考试中，当给出观测值时，p 值方法通常更快捷，因为只需计算尾概率并与 α 比较即可。

8. Hypothesis Testing with the Binomial Distribution | 二项分布假设检验

All IGCSE OCR hypothesis tests assume a binomial model. You must be able to:

所有 IGCSE OCR 假设检验均假设二项分布模型。你必须能够：

• Identify n and p from the context.

  从情境中确定 n 和 p。

• State H₀ and H₁ using correct notation.

  使用正确的符号陈述 H₀ 和 H₁。

• Calculate probabilities using P(X = k) = ⁿCₖ pᵏ (1-p)ⁿ⁻ᵏ, or use cumulative binomial tables provided.

  使用 P(X = k) = ⁿCₖ pᵏ (1-p)ⁿ⁻ᵏ 计算概率，或使用提供的累积二项分布表。

• Interpret the result in the given context.

  在给定情境中解释结果。

A typical problem: ‘A manufacturer claims that at most 5% of its products are defective. A random sample of 20 products contains 3 defectives. Test at the 5% level whether the claim is justified.’

一道典型题目：“某制造商声称其产品的不合格率至多为 5%。随机抽取 20 件产品，发现 3 件不合格。在 5% 水平下检验该声称是否合理。”

Here, H₀: p = 0.05, H₁: p > 0.05. Observed X=3. Find P(X ≥ 3) under p=0.05. If this probability ≤ 0.05, reject H₀.

这里 H₀: p = 0.05, H₁: p > 0.05。观测值 X=3。在 p=0.05 下求 P(X ≥ 3)。如果该概率 ≤ 0.05，则拒绝 H₀。

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

Errors can occur in hypothesis testing. A Type I error is rejecting H₀ when it is actually true. The probability of a Type I error is exactly α, the significance level.

假设检验中可能出现错误。第一类错误是当 H₀ 实际为真时拒绝 H₀。发生第一类错误的概率正是显著性水平 α。

A Type II error is failing to reject H₀ when H₁ is true. Its probability is denoted by β and is not controlled directly in IGCSE, but you should understand the concept.

第二类错误是当 H₁ 为真时未能拒绝 H₀。其概率记作 β，在 IGCSE 中不直接要求控制，但你应理解其概念。

There is a trade-off: decreasing α reduces the chance of a Type I error but increases the chance of a Type II error for a fixed sample size.

两者存在权衡：降低 α 会减少第一类错误的机会，但对于固定样本量会增加第二类错误的机会。

In exams, you may be asked to identify, for a given scenario, what a Type I or Type II error would mean in plain English.

考试中可能会要求你针对给定场景，用通俗语言说明第一类或第二类错误代表什么。

10. Worked Example | 例题精讲

Problem: A student claims that a die is biased towards six. In 30 throws, a six appears 9 times. Test at the 5% significance level whether the claim is supported.

问题：某学生声称一颗骰子偏向于点数为六。在 30 次投掷中，六点出现了 9 次。在 5% 显著性水平下检验该声称是否得到支持。

• H₀: p = 1/6, H₁: p > 1/6; one-tailed test.

  H₀: p = 1/6，H₁: p > 1/6；单尾检验。

• Under H₀, X ~ B(30, 1/6). Observed value = 9.

  在 H₀ 下，X ~ B(30, 1/6)。观测值 = 9。

• Find p-value = P(X ≥ 9) = 1 − P(X ≤ 8). Using binomial tables or calculator, P(X ≤ 8) ≈ 0.876, so p-value ≈ 0.124.

  求 p 值 = P(X ≥ 9) = 1 − P(X ≤ 8)。查表或使用计算器，P(X ≤ 8) ≈ 0.876，因此 p 值 ≈ 0.124。

• Since p-value = 0.124 > 0.05, we do not reject H₀.

  由于 p 值 = 0.124 > 0.05，我们不拒绝 H₀。

• Conclusion: There is insufficient evidence at the 5% level to suggest the die is biased towards six.

  结论：在 5% 显著性水平下，没有足够证据表明骰子偏向六点。

Always check whether the p-value is less than or equal to α before making your conclusion.

在得出结论前务必检查 p 值是否小于或等于 α。

11. Common Exam Pitfalls | 常见考试误区

• Confusing p and p̂: Hypotheses are about the population proportion p, not the sample proportion.

  混淆 p 与 p̂：假设针对的是总体比例 p，而不是样本比例。

• Using two-tailed when one-tailed is required: Read the wording carefully. ‘Increase’, ‘more than’, ‘higher’ → one-tailed. ‘Different’, ‘changed’ → two-tailed.

  需要单尾时误用双尾：仔细阅读措辞。“增加”、“大于”、“更高” → 单尾。“不同”、“改变” → 双尾。

• Stating ‘accept H₀’: We never accept H₀; we only say ‘do not reject H₀’ because failing to reject does not prove H₀ is true.

  说“接受 H₀”：我们从不接受 H₀；只能说“不拒绝 H₀”，因为未能拒绝并不能证明 H₀ 为真。

• Misinterpreting p-value: The p-value is not the probability that H₀ is true. It is the probability of the observed result (or more extreme) given H₀ is true.

  错误解释 p 值：p 值不是 H₀ 为真的概率。它是在 H₀ 为真的条件下，得到当前结果（或更极端结果）的概率。

• Forgetting to halve the significance level for two-tailed tests: Always split α into two tails when H₁ has a ≠ sign.

  双尾检验忘记将显著性水平减半：当 H₁ 包含 ≠ 符号时，一定要将 α 分配到两尾。

12. Key Summary | 核心总结

Hypothesis testing on the IGCSE OCR syllabus simplifies to a clear sequence: define hypotheses, choose significance level, collect data, calculate tail probability, and compare with α. The binomial setting underpins every test, so strong binomial probability skills are essential. Remember that the conclusion must be stated in context, using phrases like ‘sufficient evidence at the X% significance level to suggest…’.

IGCSE OCR 考纲中的假设检验可简化为一个清晰的流程：定义假设、选择显著性水平、收集数据、计算尾概率，并与 α 比较。二项分布背景是每个检验的基础，因此扎实的二项分布概率计算能力至关重要。记住，结论必须结合情境，使用“在 X% 显著性水平下有充分证据表明……”的表述。

Practice interpreting both critical region and p-value approaches, as questions may require one or the other. With careful reading and systematic calculation, hypothesis testing can become one of the most reliable marks on your paper.

练习使用临界域和 p 值两种方法进行解释，因为题目可能要求使用其中一种。仔细阅读题目并系统计算，假设检验可以成为试卷上最容易得分的部分之一。

Published by TutorHao | IGCSE Maths Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）