📚 Hypothesis Testing for IGCSE Edexcel Mathematics | IGCSE Edexcel 数学：假设检验考点精讲
Hypothesis testing is a fundamental concept in inferential statistics. In the IGCSE Edexcel Mathematics syllabus, students are expected to carry out binomial hypothesis tests, understand one–tailed and two–tailed tests, and interpret p–values and critical regions. This revision guide breaks down each key idea with clear examples.
假设检验是推断统计中的基本概念。在 IGCSE Edexcel 数学大纲中，学生需要能够进行二项分布的假设检验，理解单尾与双尾检验，并解释 p 值与拒绝域。本复习指南将通过清晰的示例逐一讲解核心知识点。
1. What is Hypothesis Testing? | 什么是假设检验？
A hypothesis test is a procedure that uses sample data to assess the strength of evidence against a stated claim about a population parameter. The aim is to decide whether the observed result is so unusual that it cannot reasonably be attributed to chance alone.
假设检验是一种利用样本数据来评估反对某个总体参数声明的证据强度的程序。其目的是判断观察到的结果是否异常到无法合理地仅归因于随机因素。
2. Null and Alternative Hypotheses | 零假设与备择假设
The null hypothesis H₀ is the status quo: it assumes that any observed difference is due to sampling variability. The alternative hypothesis H₁ states that there is a genuine effect or difference. H₁ can be one‑tailed (e.g. p > 0.5 or p < 0.5) or two‑tailed (p ≠ 0.5).
零假设 H₀ 是默认立场：假定观察到的任何差异都是由抽样波动引起的。备择假设 H₁ 则声称存在真实的效果或差异。H₁ 可以是单尾的（如 p > 0.5 或 p < 0.5）或双尾的（p ≠ 0.5）。
3. Significance Level and Confidence | 显著性水平与置信水平
The significance level α (usually 0.05 or 0.01) is the maximum probability of making a Type I error — rejecting H₀ when it is actually true. The confidence level is 1 − α. Choosing α before the test is essential to avoid bias.
显著性水平 α（通常为 0.05 或 0.01）是发生第一类错误的最大概率——即错误地拒绝真实的 H₀。置信水平为 1 − α。在检验前选定 α 至关重要，以免产生偏差。
4. Test Statistic and the Binomial Distribution | 检验统计量与二项分布
For IGCSE problems, the test statistic X is the number of successes in a fixed number of independent trials n. Under H₀, X ~ B(n, p₀), where p₀ is the assumed probability of success. The evidence is assessed by how likely the observed X (or a more extreme value) occurs.
在 IGCSE 题目中，检验统计量 X 是在固定次数的独立试验 n 中成功的次数。在 H₀ 下，X ~ B(n, p₀)，其中 p₀ 是假设的成功概率。证据的强度根据观察到的 X（或更极端的值）出现的可能性来评定。
5. The p–value Approach | p 值方法
The p–value is the probability, assuming H₀ is true, of obtaining a result at least as extreme as the observed one. If p–value ≤ α, we reject H₀. If p–value > α, we do not reject H₀. The smaller the p–value, the stronger the evidence against H₀.
p 值是在 H₀ 为真的前提下，得到至少与实际观察结果一样极端的结果的概率。若 p 值 ≤ α，则拒绝 H₀；若 p 值 > α，则不拒绝 H₀。p 值越小，反对 H₀ 的证据越强。
6. The Critical Value (Rejection Region) Method | 临界值（拒绝域）方法
Find a critical value c such that P(X ≥ c | H₀) ≤ α (for an upper‑tail test) or P(X ≤ c | H₀) ≤ α (for a lower‑tail test). The region {X ≥ c} or {X ≤ c} is the rejection region. If the observed X falls in this region, H₀ is rejected.
寻找临界值 c，使得 P(X ≥ c | H₀) ≤ α（上尾检验）或 P(X ≤ c | H₀) ≤ α（下尾检验）。区域 {X ≥ c} 或 {X ≤ c} 称为拒绝域。若观察到的 X 落入该区域，则拒绝 H₀。
7. One–tailed vs Two–tailed Tests | 单尾检验与双尾检验
A one‑tailed test examines a specific direction of interest (greater than or less than). A two‑tailed test looks for any difference from the assumed value. For a two‑tailed test, the p–value is doubled when the binomial distribution is symmetric (p₀ = 0.5), or we split α equally between both tails.
单尾检验关注指定的方向（大于或小于）。双尾检验则检测与假设值的任何差异。对于双尾检验，当二项分布对称时（p₀ = 0.5），将单尾概率乘以 2 作为 p 值；或者将 α 平分到两侧尾部。
8. Step‑by‑Step Procedure | 解题步骤总结
1. State H₀ and H₁ clearly. 2. Choose significance level α. 3. Write down the distribution of X under H₀. 4. Compute the p–value or find the critical region. 5. Compare the p–value with α, or the test statistic with the critical region. 6. Write a conclusion in the context of the problem, using phrases like ‘sufficient evidence to reject H₀’ or ‘not enough evidence to reject H₀’.
1. 明确写出 H₀ 和 H₁。2. 选择显著性水平 α。3. 写出在 H₀ 下 X 的分布。4. 计算 p 值或确定拒绝域。5. 将 p 值与 α 比较，或将检验统计量与拒绝域比较。6. 在问题语境下写出结论，使用“有足够证据拒绝 H₀”或“证据不足，无法拒绝 H₀”等表述。
9. Worked Example: One–tailed Lower Tail Test | 典型例题：单尾下尾检验
Problem: A manufacturer claims that fewer than 10% of its light bulbs are defective. A random sample of 20 bulbs contains 1 defective. Test this claim at the 5% significance level.
问题：某制造商声称其灯泡次品率低于 10%。随机抽取 20 只灯泡，发现 1 只次品。在 5% 显著性水平下检验该声称。
Solution – Hypotheses: H₀: p = 0.10, H₁: p < 0.10 (one‑tailed lower). α = 0.05.
解答 – 假设：H₀: p = 0.10，H₁: p < 0.10（单尾下尾）。α = 0.05。
Under H₀, X ~ B(20, 0.10). p–value = P(X ≤ 1 | p=0.10) = P(X=0) + P(X=1). Using the binomial formula: P(X=0) = (0.90)²⁰ ≈ 0.1216, P(X=1) = 20 × 0.10 × (0.90)¹⁹ ≈ 0.2702. p–value ≈ 0.3918.
在 H₀ 下，X ~ B(20, 0.10)。p 值 = P(X ≤ 1 | p=0.10) = P(X=0) + P(X=1)。用二项公式计算：P(X=0) = (0.90)²⁰ ≈ 0.1216，P(X=1) = 20 × 0.10 × (0.90)¹⁹ ≈ 0.2702。p 值 ≈ 0.3918。
Since 0.3918 > 0.05, we do not reject H₀. There is insufficient evidence at the 5% level to support the claim that the proportion of defective bulbs is less than 10%.
因为 0.3918 > 0.05，不拒绝 H₀。在 5% 水平下，没有足够证据支持次品率低于 10% 的声称。
10. Worked Example: Two–tailed Test | 典型例题：双尾检验
Problem: Is a coin fair? You toss a coin 20 times and obtain 15 heads. Test at the 5% significance level.
问题：一枚硬币是否公平？抛掷 20 次，得到 15 次正面。在 5% 显著性水平下检验。
Solution – Hypotheses: H₀: p = 0.5, H₁: p ≠ 0.5 (two‑tailed). α = 0.05. Under H₀, X ~ B(20, 0.5). Observed X = 15.
解答 – 假设：H₀: p = 0.5，H₁: p ≠ 0.5（双尾）。α = 0.05。在 H₀ 下，X ~ B(20, 0.5)。观测值 X = 15。
For a two‑tailed test with symmetric binomial, p–value = 2 × P(X ≥ 15 | p=0.5). From tables or calculator, P(X ≥ 15) = 1 − P(X ≤ 14) ≈ 0.0207. Thus p–value = 2 × 0.0207 = 0.0414.
对于对称二项分布的双尾检验，p 值 = 2 × P(X ≥ 15 | p = 0.5)。查表或计算器可得 P(X ≥ 15) = 1 − P(X ≤ 14) ≈ 0.0207。因此 p 值 = 2 × 0.0207 = 0.0414。
Since 0.0414 < 0.05, we reject H₀. There is sufficient evidence at the 5% level to conclude that the coin is biased.
因为 0.0414 < 0.05，拒绝 H₀。在 5% 水平下有足够证据认为硬币不均匀。
11. Common Mistakes and Pitfalls | 常见错误与注意事项
1. Confusing the direction of the inequality in one‑tailed tests. Always check whether the alternative hypothesis points to ‘greater than’ or ‘less than’. 2. Forgetting to double the tail probability in a two‑tailed test. 3. Writing ‘accept H₀’ instead of ‘do not reject H₀’. 4. Using the wrong binomial parameters under H₀. 5. Failing to relate the conclusion back to the context of the question.
1. 单尾检验中混淆不等式的方向。务必检查备择假设是“大于”还是“小于”。2. 在双尾检验中忘记将单尾概率乘以 2。3. 将“不拒绝 H₀”写成“接受 H₀”。4. 在 H₀ 下使用了错误的二项分布参数。5. 结论未能结合问题的实际背景。
12. Summary and Key Exam Tips | 总结与考点提示
Always state the hypotheses precisely, remember the significance level α, and show the probability calculation or critical region clearly. For Edexcel IGCSE exam papers, the binomial cumulative probability tables are often provided, but you should also be comfortable using the formula. Practise interpreting p–values and writing conclusions in non‑technical language for full marks.
始终准确地陈述假设，牢记显著性水平 α，并清晰地展示概率计算或拒绝域的推导。在 Edexcel IGCSE 试卷中，通常会提供二项累积概率表，但你也应熟练运用公式。多练习解释 p 值并用非技术性语言撰写结论，方能拿到满分。
Published by TutorHao | Mathematics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导