Understanding the t-Distribution in IB Maths | 理解IB数学中的t分布

📚 Understanding the t-Distribution in IB Maths | 理解IB数学中的t分布

The t-distribution, also known as Student’s t-distribution, is one of the most important concepts in IB Mathematics: Analysis and Approaches HL. It is widely used in statistical inference, particularly when sample sizes are small and population standard deviations are unknown.

t分布，又称学生t分布，是IB数学分析与方法HL中最重要的概念之一。它广泛应用于统计推断，尤其是在样本量较小且总体标准差未知的情况下。

1. Why Do We Need the t-Distribution? | 为什么我们需要t分布？

In many real-world scenarios, we do not know the population standard deviation σ. When the sample size is large (n ≥ 30), the Central Limit Theorem allows us to use the normal distribution as an approximation. However, when n is small, the normal distribution is not accurate enough for estimating confidence intervals or conducting hypothesis tests.

在许多实际情境中，我们并不知道总体标准差σ。当样本量较大（n≥30）时，中心极限定理允许我们使用正态分布进行近似。然而，当n较小时，正态分布对于估计置信区间或进行假设检验来说不够精确。

William Sealy Gosset, writing under the pseudonym “Student,” developed the t-distribution in 1908 to address this problem. The t-distribution accounts for the extra uncertainty introduced by estimating σ from the sample itself.

威廉·西利·戈塞特以笔名”Student”于1908年提出了t分布，用以解决这一问题。t分布考虑了通过样本估计σ所带来的额外不确定性。

2. Definition of the t-Distribution | t分布的定义

Let X₁, X₂, …, Xₙ be a random sample from a normal population with mean μ and standard deviation σ. The t-statistic is defined as:

设X₁, X₂, …, Xₙ是来自均值为μ、标准差为σ的正态总体中的随机样本。t统计量定义为：

t = (x̄ – μ) / (s / √n)

where x̄ is the sample mean, s is the sample standard deviation, and n is the sample size. The t-statistic follows a t-distribution with ν = n – 1 degrees of freedom.

其中x̄是样本均值，s是样本标准差，n是样本容量。t统计量服从自由度为ν = n – 1的t分布。

The probability density function of the t-distribution is given by:

t分布的概率密度函数为：

f(t) = Γ((ν+1)/2) / (√(νπ) · Γ(ν/2)) · (1 + t²/ν)^(-(ν+1)/2)

where Γ is the gamma function and ν is the degrees of freedom. In the IB syllabus, you are not required to memorize this formula, but understanding its shape and properties is essential.

其中Γ是伽马函数，ν是自由度。在IB教学大纲中，你不需要记住这个公式，但理解其形状和性质至关重要。

3. Key Properties of the t-Distribution | t分布的关键性质

The t-distribution has several important properties that you must know for your IB exams:

t分布有几个你必须为IB考试掌握的重要性质：

Symmetric: The t-distribution is symmetric about t = 0, just like the standard normal distribution.

对称性：t分布关于t = 0对称，就像标准正态分布一样。
Bell-shaped: It has a bell shape, but with heavier tails than the normal distribution.

钟形：它具有钟形曲线，但尾部比正态分布更厚。
Degrees of freedom: The shape depends on the degrees of freedom ν. As ν increases, the t-distribution approaches the standard normal distribution.

自由度：其形状取决于自由度ν。随着ν增大，t分布趋近于标准正态分布。
Variance: The variance of the t-distribution is ν/(ν-2), which is defined for ν > 2.

方差：t分布的方差为ν/(ν-2)，在ν > 2时有定义。

4. Degrees of Freedom | 自由度

Degrees of freedom (df) is a fundamental concept in statistical inference. In the context of a one-sample t-test, df = n – 1. This represents the number of independent pieces of information available after estimating the sample mean.

自由度（df）是统计推断中的基本概念。在单样本t检验的背景下，df = n – 1。这代表在估计样本均值后，可用的独立信息数量。

For two-sample t-tests, the calculation of degrees of freedom is more complex. There are two cases:

对于双样本t检验，自由度的计算更为复杂。有两种情况：

Pooled t-test (equal variances assumed): df = n₁ + n₂ – 2.

合并t检验（假设方差相等）：df = n₁ + n₂ – 2。
Welch’s t-test (unequal variances): df is calculated using the Welch-Satterthwaite equation, which in IB is typically left to your GDC.

韦尔奇t检验（方差不相等）：df使用韦尔奇-萨特思韦特方程计算，在IB中通常交给图形计算器处理。

Understanding degrees of freedom is crucial because the critical values of the t-distribution change with df. Smaller df produces larger critical values, meaning we need stronger evidence to reject the null hypothesis.

理解自由度至关重要，因为t分布的临界值随df变化。较小的df会产生较大的临界值，意味着我们需要更强的证据来拒绝原假设。

5. t-Distribution vs. Normal Distribution | t分布与正态分布的比较

It is essential to understand how the t-distribution differs from the standard normal distribution:

理解t分布与标准正态分布的差异至关重要：

Property	Standard Normal Z	t-Distribution
Property \| 性质	Standard Normal Z \| 标准正态Z	t-Distribution \| t分布
Shape \| 形状	Fixed, independent of n	Depends on df = n – 1
Tails \| 尾部	Lighter (less probability in tails)	Heavier (more probability in tails)
Use when \| 使用条件	σ known or n ≥ 30	σ unknown and n small
As n → ∞ \| 当n→∞时	—	Converges to N(0, 1)

In your IB exams, you will often use a GDC (such as the TI-84 or Casio fx-9860G) to find t critical values and p-values. However, you must also be familiar with reading t-distribution tables by hand, as some exam questions may require this skill.

在你的IB考试中，你通常会使用图形计算器（如TI-84或Casio fx-9860G）来查找t临界值和p值。然而，你也必须熟悉手动读取t分布表，因为有些考试题目可能需要这项技能。

6. One-Sample t-Test | 单样本t检验

The one-sample t-test is used to determine whether the mean of a population differs from a hypothesized value μ₀. The hypotheses are:

单样本t检验用于判断总体均值是否与假设值μ₀存在差异。假设为：

H₀: μ = μ₀ vs. H₁: μ ≠ μ₀ (two-tailed) or μ > μ₀ / μ < μ₀ (one-tailed)

The test statistic is calculated as:

检验统计量计算如下：

t = (x̄ – μ₀) / (s / √n)

with df = n – 1. You then compare this calculated t-value with the critical value from the t-table, or use your GDC to find the p-value directly.

其中df = n – 1。然后你将计算出的t值与t表中的临界值进行比较，或者直接使用图形计算器查找p值。

Example: A teacher claims that the mean score of students on a test is 70. A sample of 12 students has a mean score of 67.5 with a standard deviation of 5.2. Test this claim at the 5% significance level.

示例：一位老师声称学生考试的平均分为70。从12名学生组成的样本中得到平均分为67.5，标准差为5.2。在5%显著性水平下检验这一说法。

Solution: t = (67.5 – 70) / (5.2 / √12) = -2.5 / 1.501 = -1.665. The critical value for df = 11 at α = 0.05 is ±2.201. Since |t| = 1.665 < 2.201, we fail to reject H₀. There is insufficient evidence to say the mean differs from 70.

解答：t = (67.5 – 70) / (5.2 / √12) = -2.5 / 1.501 = -1.665。df = 11，α = 0.05时的临界值为±2.201。由于|t| = 1.665 < 2.201，我们无法拒绝H₀。没有充分证据表明均值与70存在差异。

7. Two-Sample t-Test | 双样本t检验

The two-sample t-test compares the means of two independent populations. In IB Maths, you need to know when to use the pooled versus the unpooled (Welch’s) version.

双样本t检验用于比较两个独立总体的均值。在IB数学中，你需要知道何时使用合并版本与独立版本（韦尔奇检验）。

Pooled t-test assumptions: Both populations are approximately normal, the two samples are independent, and the population variances are equal (σ₁² = σ₂²).

合并t检验的假设：两个总体近似服从正态分布，两个样本相互独立，且总体方差相等（σ₁² = σ₂²）。

The pooled t-statistic is:

合并t统计量为：

t = (x̄₁ – x̄₂) / (sₚ · √(1/n₁ + 1/n₂))

where the pooled standard deviation is:

其中合并标准差为：

sₚ = √(((n₁-1)s₁² + (n₂-1)s₂²) / (n₁ + n₂ – 2))

with df = n₁ + n₂ – 2.

其中df = n₁ + n₂ – 2。

Welch’s t-test: When population variances cannot be assumed equal, use Welch’s t-test. The test statistic is:

韦尔奇t检验：当无法假设总体方差相等时，使用韦尔奇t检验。检验统计量为：

t = (x̄₁ – x̄₂) / √(s₁²/n₁ + s₂²/n₂)

In IB exams, you will typically use your GDC to perform these calculations. Make sure you can interpret the output correctly.

在IB考试中，你通常会使用图形计算器进行这些计算。确保你能正确解读计算结果。

8. Confidence Intervals Using the t-Distribution | 使用t分布构建置信区间

When the population standard deviation is unknown, we construct confidence intervals for the population mean using the t-distribution:

当总体标准差未知时，我们使用t分布为总体均值构建置信区间：

x̄ ± t*(α/2, df) · (s / √n)

where t*(α/2, df) is the critical t-value for the desired confidence level.

其中t*(α/2, df)是所需置信水平对应的t临界值。

Example: A sample of 9 light bulbs has a mean lifetime of 1200 hours with a standard deviation of 30 hours. Construct a 95% confidence interval for the true mean lifetime.

示例：从9个灯泡组成的样本中得到平均寿命为1200小时，标准差为30小时。构建真实平均寿命的95%置信区间。

Solution: With df = 8 and α/2 = 0.025, the critical value is t* = 2.306. The margin of error is 2.306 × 30/√9 = 2.306 × 10 = 23.06. The confidence interval is (1176.94, 1223.06) hours.

解答：df = 8，α/2 = 0.025时，临界值为t* = 2.306。误差范围是2.306 × 30/√9 = 2.306 × 10 = 23.06。置信区间为(1176.94, 1223.06)小时。

Note that this interval is wider than it would be if we used the normal distribution. This reflects the additional uncertainty from estimating σ.

请注意，这个区间比使用正态分布时的区间更宽。这反映了因估计σ而产生的额外不确定性。

9. Assumptions and Conditions | 假设与条件

The t-distribution is valid only when certain conditions are met. For the IB exam, you must be able to state and verify these assumptions:

t分布仅在满足特定条件时才有效。对于IB考试，你必须能够陈述并验证这些假设：

Random sampling: The sample must be randomly selected from the population.

随机抽样：样本必须从总体中随机选取。
Normality: The population should be approximately normal, especially for small samples. For moderate samples, the t-procedure is robust to modest departures from normality. For large samples, the CLT ensures the sample mean is approximately normal.

正态性：总体应近似服从正态分布，特别是对于小样本。对于中等样本，t检验方法对轻微偏离正态性具有稳健性。对于大样本，中心极限定理保证了样本均值近似服从正态分布。
Independence: Observations must be independent of each other.

独立性：观测值之间必须相互独立。
Unknown σ: The population standard deviation must be unknown — this is what makes the t-distribution necessary.

σ未知：总体标准差必须未知——这正是需要t分布的原因。

In exam questions, you may be asked to check these conditions before proceeding with a t-test. Remember that when n ≥ 30, many textbooks and the IB suggest using the normal distribution (z-test) as an approximation, even if σ is unknown.

在考试题目中，你可能会被要求在继续进行t检验之前检查这些条件。请记住，当n≥30时，许多教科书和IB建议使用正态分布（z检验）作为近似，即使σ未知。

10. Paired t-Test | 配对t检验

The paired t-test is used when the data consists of matched pairs — for example, measuring the same subjects before and after an intervention. In this case, we work with the differences dᵢ = x₁ᵢ – x₂ᵢ.

当数据由匹配对组成时使用配对t检验——例如，在干预前后测量同一批受试者。此时，我们使用差值dᵢ = x₁ᵢ – x₂ᵢ。

The test statistic is:

检验统计量为：

t = d̄ / (s_d / √n)

where d̄ is the mean difference, s_d is the standard deviation of the differences, and n is the number of pairs. The degrees of freedom are n – 1.

其中d̄是平均差值，s_d是差值的标准差，n是配对数。自由度为n – 1。

Example: A group of 6 students takes a maths test before and after a revision course. The differences (after – before) are: 5, 3, -1, 2, 4, 6. Test at the 5% level whether the course improves scores.

示例：一组6名学生在上复习课程前后参加了数学测试。差值（后-前）为：5, 3, -1, 2, 4, 6。在5%显著性水平下检验该课程是否能提高分数。

Solution: d̄ = 3.167, s_d = 2.483, n = 6. t = 3.167 / (2.483/√6) = 3.167 / 1.014 = 3.124. Critical value with df = 5 at α = 0.05 (one-tailed) is 2.015. Since t = 3.124 > 2.015, we reject H₀ and conclude that the revision course significantly improves scores.

解答：d̄ = 3.167，s_d = 2.483，n = 6。t = 3.167 / (2.483/√6) = 3.167 / 1.014 = 3.124。df = 5、α = 0.05（单尾）时的临界值为2.015。由于t = 3.124 > 2.015，我们拒绝H₀，并得出结论：复习课程显著提高了分数。

11. Common Mistakes and Pitfalls | 常见错误与陷阱

Students often make the following mistakes in t-distribution questions. Avoid these in your exam:

学生在处理t分布相关题目时经常犯以下错误。考试中请避免这些失误：

Using z instead of t: When σ is unknown, always use the t-distribution, not the normal distribution, for small samples.

使用z而不是t：当σ未知且样本量较小时，始终使用t分布，而非正态分布。
Wrong degrees of freedom: For a one-sample test, df = n – 1. For a two-sample pooled test, df = n₁ + n₂ – 2. Getting these wrong will lead to incorrect critical values.

自由度错误：单样本检验中df = n – 1。双样本合并检验中df = n₁ + n₂ – 2。弄错这些会导致临界值不正确。
Wrong tails: Always check whether the alternative hypothesis is one-tailed or two-tailed before finding the critical value.

错误选择单双尾：在查找临界值之前，务必确认备择假设是单尾还是双尾。
Ignoring assumptions: Using a t-test when conditions are not met can lead to invalid conclusions.

忽略假设条件：在条件不满足时使用t检验可能会导致无效的结论。
Confusing paired and independent tests: If data are paired, you must use the paired t-test, not the independent two-sample test.

混淆配对检验与独立检验：如果数据是配对的，你必须使用配对t检验，而不能用独立双样本检验。

12. t-Distribution in the IB Formula Booklet | IB公式册中的t分布

The IB Mathematics: Analysis and Approaches HL formula booklet provides a t-distribution table. You should be familiar with its layout: the rows correspond to degrees of freedom, and the columns correspond to different tail probabilities.

IB数学分析与方法HL公式册中提供了t分布表。你应该熟悉其排版：行对应自由度，列对应不同的尾部概率。

In recent IB exams, students are expected to use their GDCs for t-tests. However, you should still know how to:

在近期的IB考试中，学生被期望使用图形计算器进行t检验。然而，你仍然应该知道如何：

State hypotheses correctly, including the correct symbol for the parameter (μ or μ₁ – μ₂)

正确陈述假设，包括使用正确的参数符号（μ或μ₁ – μ₂）
Write the test statistic formula in terms of the problem’s variables

用题目中的变量写出检验统计量公式
Calculate degrees of freedom correctly

正确计算自由度
Report the p-value from the GDC with appropriate precision

从图形计算器中以适当的精确度报告p值
Draw a correct conclusion in context, using words like “significant” or “not significant”

结合语境得出正确结论，使用”显著”或”不显著”等词语

For example, a full mark solution would include the conclusion: “Since the p-value (0.031) is less than the significance level (0.05), we reject H₀. There is sufficient evidence at the 5% level to conclude that the mean weight differs from 50 kg.”

例如，一个满分的解答应包含这样的结论：”由于p值（0.031）小于显著性水平（0.05），我们拒绝H₀。在5%显著性水平下有充分的证据表明平均体重与50千克存在差异。”

13. Practice Questions | 练习题目

Here are some practice questions to test your understanding:

以下是一些检验你理解的练习题：

Question 1: A company claims that its cereal boxes contain an average of 500 grams. A random sample of 10 boxes has a mean of 495 grams and a standard deviation of 8 grams. Test the claim at the 1% significance level.

题目1：一家公司声称其麦片盒平均装有500克。从10个盒子的随机样本中得到平均为495克，标准差为8克。在1%显著性水平下检验该说法。

Question 2: Two different teaching methods are compared. 14 students are taught using method A (mean = 72, s = 6) and 12 using method B (mean = 68, s = 7). At the 5% level, is there a significant difference between the two methods?

题目2：比较两种不同的教学方法。14名学生使用方法A教学（均值=72，s = 6），12名学生使用方法B教学（均值=68，s = 7）。在5%显著性水平下，两种方法之间是否存在显著差异？

Question 3: Repeat the Example in Section 8, but now with n = 16 and s = 40. What is the 90% confidence interval?

题目3：重做第8节中的示例，但现在n = 16，s = 40。90%置信区间是多少？

Try these questions yourself and check your answers with a GDC. Practice is the key to mastering t-distribution questions in the IB exam.

自己尝试这些题目，并使用图形计算器核对答案。练习是掌握IB考试中t分布题目的关键。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Understanding the t-Distribution in IB Maths | 理解IB数学中的t分布