📚 Hypothesis Testing for IB Mathematics | IB 数学：假设检验 考点精讲
Hypothesis testing is a cornerstone of inferential statistics, allowing us to make decisions about population parameters based on sample data. In the IB Mathematics curriculum, particularly in Analysis & Approaches (AA) and Applications & Interpretation (AI) at Higher Level, you will be expected to formulate null and alternative hypotheses, calculate test statistics and p-values, and draw correct conclusions in context. This masterclass covers the essential concepts, step-by-step procedures, common distributions used, and typical pitfalls to help you excel in your exams.
假设检验是推断统计学的基石，它让我们能够根据样本数据对总体参数做出决策。在 IB 数学课程中，尤其是高级别的分析与方法（AA）和应用与解释（AI），你需要能够提出零假设和备择假设，计算检验统计量和 p 值，并在具体情境中得出正确结论。这篇考点精讲涵盖了基本概念、分步流程、常用的分布以及典型误区，帮助你在考试中取得优异成绩。
1. Introduction to Hypothesis Testing | 假设检验简介
Hypothesis testing is a statistical method used to determine whether there is enough evidence in a sample to reject a statement about a population parameter. It starts with an initial claim, known as the null hypothesis, and an alternative claim that challenges it.
假设检验是一种统计方法，用来判断样本中是否有足够证据来拒绝关于总体参数的某个声明。它从一个初始声明（零假设）和一个对立的备择假设开始。
The core idea is based on probability: we assume the null hypothesis is true and then examine how likely the observed sample result (or something more extreme) would occur. If this probability is very small, we cast doubt on the null hypothesis.
核心思想基于概率：我们假设零假设为真，然后考察观测到的样本结果（或更极端的情况）发生的可能性有多大。如果这个概率非常小，我们就对零假设产生怀疑。
Every hypothesis test involves a pre-defined significance level, a test statistic computed from the data, and a decision rule. The final conclusion is always stated in the context of the original problem.
每个假设检验都包含预先设定的显著性水平、根据数据计算的检验统计量以及决策规则。最终结论必须在原始问题的情境中给出。
2. Null and Alternative Hypotheses | 零假设与备择假设
The null hypothesis, denoted H₀, is a statement of no effect, no difference, or status quo. It is the hypothesis that we assume to be true unless the data provides convincing evidence against it. For example, H₀: μ = 50 means the population mean is 50.
零假设，记作 H₀，是关于“无效应、无差异或维持现状”的声明。除非数据提供令人信服的相反证据，我们假定 H₀ 为真。例如，H₀: μ = 50 表示总体均值为 50。
The alternative hypothesis, denoted H₁ or Hₐ, represents what we want to prove: there is an effect, a difference, or a change. It can be one-sided (μ > 50 or μ < 50) or two-sided (μ ≠ 50). The choice of H₁ determines whether the test is one-tailed or two-tailed.
备择假设，记作 H₁ 或 Hₐ，代表我们想要证明的效应、差异或变化。它可以是单侧的（μ > 50 或 μ < 50），也可以是双侧的（μ ≠ 50）。H₁ 的选择决定了检验是单尾还是双尾。
In IB exams, you must clearly write both hypotheses using proper notation and refer to the population parameter (e.g., μ for mean, p for proportion). Always state H₀ and H₁ before carrying out any calculations.
在 IB 考试中，你必须用正确的符号清晰地写出两个假设，并引用总体参数（如均值用 μ，比例用 p）。在进行任何计算之前，务必先写出 H₀ 和 H₁。
3. Significance Level and p-value | 显著性水平与 p 值
The significance level, denoted by α (alpha), is the threshold probability at which we are willing to reject the null hypothesis. Common choices are α = 0.05, 0.01, or 0.10. It represents the maximum acceptable risk of making a Type I error.
显著性水平，记作 α，是我们愿意拒绝零假设的概率阈值。常见的选择有 α = 0.05、0.01 或 0.10。它代表可接受的 I 类错误的最大风险。
The p-value is the probability of obtaining a test result at least as extreme as the one observed, assuming that H₀ is true. A small p-value (≤ α) indicates strong evidence against H₀, leading to rejection. A large p-value (> α) means insufficient evidence to reject H₀.
p 值是指在 H₀ 为真的前提下，得到与观测结果至少一样极端的检验结果的概率。小的 p 值（≤ α）表明反对 H₀ 的证据很强，应当拒绝 H₀；大的 p 值（> α）意味着没有足够证据拒绝 H₀。
In IB questions, you will often compute the p-value directly or compare a test statistic to a critical value. Always compare the p-value to α to make your decision, not the other way round.
在 IB 问题中，你经常需要直接计算 p 值，或者将检验统计量与临界值进行比较。做决策时，总是把 p 值与 α 比较，而不是反过来。
4. One-tailed vs Two-tailed Tests | 单尾与双尾检验
A one-tailed (or one-sided) test is used when the alternative hypothesis specifies a direction of the effect, such as H₁: μ > 100. The critical region is located entirely in one tail of the sampling distribution, and the p-value is the probability in that tail only.
当备择假设明确效应的方向时，如 H₁: μ > 100，使用单尾检验。此时拒绝域完全位于抽样分布的一个尾部，p 值只从该尾部计算概率。
A two-tailed (or two-sided) test is appropriate when we are testing for any difference from the null value, H₁: μ ≠ 100. The significance level α is split equally between the two tails, and the p-value involves both extremes. For symmetric distributions, the p-value is twice the one-tailed probability.
当我们检验的是偏离零假设值的任何差异，即 H₁: μ ≠ 100，适合使用双尾检验。显著性水平 α 被平均分配到两个尾部，p 值涉及两端极端情况。对于对称分布，p 值是单尾概率的两倍。
Choosing the correct tail is crucial. Read the wording carefully: phrases like “increased”, “more than”, “greater” suggest an upper-tailed test; “decreased”, “less than” suggest a lower-tailed test; “changed”, “different” point to a two-tailed test.
选择正确的尾部至关重要。仔细阅读题干措辞：“增加”、“多于”、“高于”暗示上尾检验；“减少”、“低于”暗示下尾检验；“改变”、“有差异”指向双尾检验。
5. Test Statistics and Critical Values | 检验统计量与临界值
A test statistic is a standardized value calculated from sample data, used to decide whether to reject H₀. Its formula depends on the parameter being tested and the distribution assumed. For a population mean with known variance, the test statistic is z = (x̄ – μ₀) / (σ / √n).
检验统计量是从样本数据计算的标准化值，用于决定是否拒绝 H₀。其公式取决于被检验的参数和假定的分布。对于已知方差的总体均值，检验统计量为 z = (x̄ – μ₀) / (σ / √n)。
The critical value is the boundary (or boundaries) that separates the rejection region from the non-rejection region on the sampling distribution. It is determined by the significance level α and the type of test (one- or two-tailed). If the test statistic falls in the critical region, we reject H₀.
临界值是抽样分布上分隔拒绝域和非拒绝域的边界，由显著性水平 α 和检验类型（单尾或双尾）决定。如果检验统计量落入拒绝域，我们就拒绝 H₀。
In the IB, you may be asked to find critical values from the normal, t, or binomial tables (or using your GDC). When using the critical value method, always sketch the distribution and shade the rejection region for clarity.
在 IB 中，你可能需要从正态分布、t 分布或二项分布表（或使用 GDC）查表找出临界值。使用临界值法时，最好画出分布示意图并标出拒绝域，使过程更清晰。
6. Type I and Type II Errors | I 类错误与 II 类错误
A Type I error occurs when we reject a true null hypothesis. The probability of making a Type I error is exactly the significance level α. For example, concluding that a new drug works when in fact it does not, with α = 0.05, means we have a 5% risk of this error.
当我们拒绝了一个真实的零假设时，发生了 I 类错误。犯 I 类错误的概率恰好为显著性水平 α。例如，若在 α = 0.05 下得出新药有效的结论而实际无效，我们就有 5% 的风险犯此类错误。
A Type II error happens when we fail to reject a false null hypothesis. Its probability is denoted by β. Power of the test is 1 – β, the probability of correctly rejecting a false H₀. IB exam questions may ask you to interpret these errors in context.
当我们未能拒绝一个错误的零假设时，发生了 II 类错误，其概率记为 β。检验的功效为 1 – β，即正确拒绝错误 H₀ 的概率。IB 考题可能要求结合情境解释这两类错误。
Reducing α reduces the chance of a Type I error but increases the chance of a Type II error (all else being equal). The only way to reduce both errors simultaneously is to increase the sample size, which narrows the sampling distribution.
降低 α 可以减少 I 类错误的机会，但同时会增加 II 类错误的机会（其他条件不变）。同时降低两类错误的唯一方法是增大样本容量，这会使抽样分布变窄。
7. Hypothesis Test for a Population Mean (σ Known) – Z-test | 总体均值的假设检验（方差已知）— Z 检验
When the population standard deviation σ is known and the data come from a normal population (or n ≥ 30 by the Central Limit Theorem), we use a z-test for the mean. The test statistic is
z = (x̄ – μ₀) / (σ / √n)
当总体标准差 σ 已知，且数据来自正态总体（或根据中心极限定理 n ≥ 30）时，我们使用均值的 z 检验。检验统计量为
z = (x̄ – μ₀) / (σ / √n)
Once the z-statistic is calculated, you can either compare it to the critical z-value from standard normal tables or compute the p-value. For a two-tailed test with H₁: μ ≠ μ₀, the p-value = 2P(Z > |z|) assuming the normal distribution.
计算出 z 统计量后，你可以将其与标准正态分布表中的临界 z 值比较，也可以计算 p 值。对于 H₁: μ ≠ μ₀ 的双尾检验，p 值 = 2P(Z > |z|)，基于正态分布假设。
In the IB, you are expected to interpret the result: “Since z = 2.31 > 1.96, we reject H₀ at the 5% significance level.” Always link the conclusion back to the wording of the problem.
在 IB 中，你需要对结果做出解释：“由于 z = 2.31 > 1.96，我们在 5% 显著性水平下拒绝 H₀。” 始终要把结论与题干的措辞联系起来。
8. Hypothesis Test for a Population Mean (σ Unknown) – t-test | 总体均值的 t 检验
When the population standard deviation is unknown and is estimated by the sample standard deviation s, we use a t-test. The test statistic follows a t-distribution with n – 1 degrees of freedom:
t = (x̄ – μ₀) / (s / √n)
当总体标准差未知且用样本标准差 s 估计时，我们使用 t 检验。检验统计量服从自由度为 n – 1 的 t 分布：
t = (x̄ – μ₀) / (s / √n)
The critical values for the t-distribution depend on degrees of freedom and the significance level. As the sample size increases, the t-distribution approaches the standard normal distribution, so for very large n the t-test and z-test give similar results.
t 分布的临界值取决于自由度和显著性水平。随着样本容量增大，t 分布趋近于标准正态分布，因此对于非常大的 n，t 检验和 z 检验给出相似结果。
IB exam questions often require you to state the degrees of freedom, use a GDC to obtain the p-value, or read t-tables. Always check the assumptions: the sample is random and the underlying population is approximately normal, which is robust for moderate sample sizes.
IB 考题经常要求你给出自由度，使用 GDC 求得 p 值，或查阅 t 分布表。务必要检查假设条件：样本是随机的，且总体近似正态，这对于中等样本量是稳健的。
9. Hypothesis Test for a Population Proportion | 总体比例的假设检验
For testing a claim about a population proportion p, we use a z-test based on the normal approximation to the binomial, provided that np₀ ≥ 10 and n(1–p₀) ≥ 10. The test statistic is:
z = (p̂ – p₀) / √[p₀(1 – p₀) / n]
where p̂ is the sample proportion.
对于总体比例 p 的检验，我们使用基于二项分布正态近似的 z 检验，前提是 np₀ ≥ 10 且 n(1–p₀) ≥ 10。检验统计量为：
z = (p̂ – p₀) / √[p₀(1 – p₀) / n]
其中 p̂ 是样本比例。
The decision rule is the same as the z-test for a mean: reject H₀ if the z-statistic falls in the critical region or if the p-value ≤ α. When calculating the p-value, use the standard normal distribution.
决策规则与均值的 z 检验相同：若 z 统计量落入拒绝域或 p 值 ≤ α，则拒绝 H₀。计算 p 值时使用标准正态分布。
For small samples where the normal approximation is not valid, IB may expect you to use the exact binomial distribution instead (see the next section). Always verify the conditions before choosing the method.
对于正态近似不成立的小样本情形，IB 可能要求你改用精确的二项分布（见下一节）。选择方法前务必验证条件。
10. Hypothesis Testing Using the Binomial Distribution | 二项分布下的假设检验
When testing a population proportion with a small sample, or when the IB explicitly instructs, we use the exact binomial distribution X ~ B(n, p₀). The p-value is found directly from binomial probabilities.
当用小样本检验总体比例，或 IB 明确要求时，我们使用精确的二项分布 X ~ B(n, p₀)。p 值直接从二项分布概率求得。
For an upper-tailed test H₁: p > p₀, the p-value = P(X ≥ observed number of successes). For a lower-tailed test H₁: p < p₀, the p-value = P(X ≤ observed successes). For a two-tailed test, the p-value is usually 2 × min[P(X ≤ x), P(X ≥ x)] but capped at 1.
对于上尾检验 H₁: p > p₀，p 值 = P(X ≥ 观测成功次数)。对于下尾检验 H₁: p < p₀，p 值 = P(X ≤ 观测成功次数)。对于双尾检验，p 值常取 2 × min[P(X ≤ x), P(X ≥ x)]，但不超过 1。
IB candidates must be comfortable using a GDC to find binomial cumulative probabilities. Remember to state clearly whether you are using the p-value method or critical region method, and always reference the binomial model.
IB 考生必须熟练使用 GDC 求二项分布累积概率。记得清楚说明你使用的是 p 值法还是拒绝域法，并始终指明所依据的二项模型。
11. p-value Method and Decision Making | p 值法与决策
The p-value method is widely used in IB examinations because it provides a quantitative measure of evidence against H₀. The decision rule is simple: if p-value ≤ α, reject H₀; if p-value > α, do not reject H₀.
p 值法在 IB 考试中被广泛使用，因为它为反对 H₀ 的证据提供了量化度量。决策规则很简单：若 p 值 ≤ α，拒绝 H₀；若 p 值 > α，不拒绝 H₀。
When writing your conclusion, never say “accept H₀”. Instead, state that there is insufficient evidence to reject H₀, or the data do not provide enough evidence against the null hypothesis. Use the wording from the problem context (e.g., “there is significant evidence at the 5% level that the mean time has increased”).
书写结论时，不要说“接受 H₀”。应表述为：没有足够证据拒绝 H₀，或数据未提供足够证据反对零假设。要使用题目情境中的措辞（例如，“在 5% 水平下有显著证据表明平均时间增加了”）。
Both the p-value method and the critical value method lead to the same conclusion. IB questions may ask you to perform the test using a specific method, so be proficient in both. Always report the p-value to at least 3 decimal places if asked.
p 值法和临界值法会得出相同结论。IB 题目可能要求用特定方法进行检验，因此两种方法都要熟练。如果题目有要求，通常将 p 值报告到至少三位小数。
12. IB Exam Tips and Common Pitfalls | IB 考试技巧与常见误区
First, read the question carefully and identify the parameter, the hypotheses, and the significance level. Many marks are lost by writing the hypotheses incorrectly or using the wrong parameter (mean vs. proportion).
首先，仔细读题，明确参数、假设和显著性水平。很多失分是由于错误地书写假设或使用了错误的参数（均值与比例混淆）。
Common pitfalls include: forgetting to state the distribution of the test statistic; using a z-test when a t-test is required; neglecting to check conditions (normality, large sample, independence); reporting the p-value without comparing to α; and misinterpreting the conclusion.
常见误区有：忘记说明检验统计量的分布；在应该使用 t 检验时用了 z 检验；忽略条件检验（正态性、大样本、独立性）；只给出 p 值而未与 α 比较；结论解读错误。
In IB Papers, you may be asked to “test at the 5% significance level whether…”. Always perform the complete test, showing the test statistic, p-value or critical value, and a contextualized conclusion. A well-labelled sketch can earn method marks.
在 IB 试卷中，你可能遇到“在 5% 显著性水平下检验是否…”的要求。要完成完整的检验，展示检验统计量、p 值或临界值，以及上下文相关的结论。标注清晰的示意图可获得方法分。
Practice using your GDC efficiently for binomial, normal, and t-distribution probabilities. Knowing how to store values and use statistical functions will save time. Finally, double-check whether the question expects an exact binomial test or a normal approximation.
高效使用 GDC 处理二项分布、正态分布和 t 分布的概率；懂得如何存储数值和调用统计功能能节省时间。最后，再次确认题目是要求精确二项检验还是正态近似。
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Cancel reply