IB Math: Finding Probabilities in the t-Distribution | IB数学：t分布中求概率的方法

📚 IB Math: Finding Probabilities in the t-Distribution | IB数学：t分布中求概率的方法

The t-distribution, also known as Student’s t-distribution, is one of the most important probability distributions in IB Mathematics Analysis and Approaches (AA) and Applications and Interpretation (AI), particularly in the topic of statistical inference. It is used when the population standard deviation is unknown and the sample size is small.

t分布，又称学生氏t分布，是IB数学分析与方法（AA）及数学应用与解释（AI）中最重要的概率分布之一，尤其在统计推断部分有着核心地位。它通常在总体标准差未知且样本量较小时使用。

---

1. What is the t-Distribution? | 什么是t分布

The t-distribution is a continuous probability distribution that arises when estimating the mean of a normally distributed population in situations where the sample size is small and the population standard deviation is unknown. It was developed by William Sealy Gosset under the pseudonym ‘Student’.

t分布是一种连续型概率分布，当样本量较小且总体标准差未知，需要估计正态分布总体的均值时，t分布便自然出现。它由威廉·西利·戈塞特以”学生”（Student）为笔名提出。

Formally, if \( X_1, X_2, …, X_n \) are independent random variables following a normal distribution \( N(\mu, \sigma^2) \), then the statistic

严格来说，若 \( X_1, X_2, …, X_n \) 是服从正态分布 \( N(\mu, \sigma^2) \) 的独立随机变量，则统计量

t = (x̄ − μ) / (s / √n)

follows a t-distribution with \( n − 1 \) degrees of freedom, where \( x̄ \) is the sample mean and \( s \) is the sample standard deviation.

服从自由度为 \( n − 1 \) 的t分布，其中 \( x̄ \) 是样本均值，\( s \) 是样本标准差。

---

2. t-Distribution vs Standard Normal Distribution | t分布与标准正态分布的区别

The t-distribution is very similar in shape to the standard normal distribution: it is symmetric, bell-shaped, and centred at zero. However, the t-distribution has heavier tails, meaning it assigns more probability to extreme values.

t分布与标准正态分布在形态上非常相似：都具有对称性、钟形曲线，且以0为中心。然而，t分布的尾部更厚，意味着它赋予极端值更大的概率。

Feature | 特征	Standard Normal Z | 标准正态Z	t-Distribution | t分布
Shape | 形状	Fixed bell curve | 固定钟形	Bell curve, heavier tails | 钟形，尾部更厚
Parameters | 参数	Mean 0, SD 1 | 均值0，标准差1	Degrees of freedom | 自由度
Tail thickness | 尾部厚度	Fixed | 固定	Decreases as df increases | 随自由度增大而变薄
Use | 用途	σ known, large n | σ已知，大样本	σ unknown, small n | σ未知，小样本

As the degrees of freedom increase, the t-distribution approaches the standard normal distribution. In practice, for df > 30, the two distributions are nearly identical.

随着自由度的增大，t分布逐渐趋近于标准正态分布。在实际应用中，当自由度大于30时，两者几乎无法区分。

---

3. Degrees of Freedom | 自由度

The degrees of freedom (df) of a t-distribution is the key parameter that determines its exact shape. In the context of a one-sample t-test, df = n − 1, where n is the sample size.

t分布的自由度（df）是决定其确切形状的关键参数。在单样本t检验中，自由度 = n − 1，其中n为样本量。

For example, if a sample of size 12 is taken, the t-distribution has df = 11. The degrees of freedom reflect the amount of independent information available in the sample after estimating the population standard deviation.

例如，若抽取样本量为12的样本，则对应的t分布自由度为11。自由度反映了在估计总体标准差之后，样本中可供使用的独立信息量。

In IB exam questions, you may be given the degrees of freedom directly, or you may need to compute it from the sample size. Always check whether the question provides df explicitly.

在IB考试题中，题目可能直接给出自由度，也可能需要你根据样本量自行计算。务必仔细审题，确认是否已明确给出自由度。

---

4. Structure of the t-Distribution Table | t分布表的结构

A standard t-distribution table displays critical values for different degrees of freedom and significance levels. The table typically shows one-tailed probabilities (e.g., 0.10, 0.05, 0.025, 0.01, 0.005) along the top row and degrees of freedom along the first column.

标准的t分布表展示了不同自由度和显著性水平下的临界值。表格顶部通常列出单尾概率（例如0.10、0.05、0.025、0.01、0.005），第一列则为自由度。

Each entry in the table is the t-value \( t_{\alpha, df} \) such that the area to its right under the t-curve equals \( \alpha \). Because the t-distribution is symmetric, the area to the left of \( −t_{\alpha, df} \) also equals \( \alpha \).

表中的每个数值是t临界值 \( t_{\alpha, df} \)，使得t曲线下该值右侧的面积等于 \( \alpha \)。由于t分布具有对称性，\( −t_{\alpha, df} \) 左侧的面积同样等于 \( \alpha \)。

df | 自由度	0.10	0.05	0.025	0.01
1	3.078	6.314	12.706	31.821
5	1.476	2.015	2.571	3.365
10	1.372	1.812	2.228	2.764
20	1.325	1.725	2.086	2.528

To find a probability using the table, you locate your degrees of freedom in the first column and read across to the appropriate critical value. However, the table only gives specific tail probabilities — it does not give arbitrary probabilities like P(T < 1.234). For those, you need a calculator.

使用t分布表求概率时，先在第一列找到你的自由度，再横向读取对应的临界值。但表只能给出特定尾部概率，无法求任意概率如P(T < 1.234)。对于这类问题，你需要使用计算器。

---

5. Using a GDC to Find t-Probabilities | 用图形计算器求t概率

In IB exams, the use of a graphing display calculator (GDC) is expected. The most efficient way to find probabilities from a t-distribution is to use the built-in distribution functions on your GDC.

在IB考试中，使用图形计算器（GDC）是基本要求。求t分布概率最高效的方法是利用GDC内置的分布函数功能。

TI-Nspire CX / TI-84 Plus Instructions | TI-Nspire CX / TI-84 Plus 操作指南

• Press 2nd → DISTR (on TI-84) to open the distribution menu.

  按 2nd → DISTR（在TI-84上）打开分布菜单。

• Select tcdf( (t cumulative distribution function).

  选择 tcdf(（t累积分布函数）。

• Enter the lower bound, upper bound, and degrees of freedom. For P(T < a), use lower bound −10⁹⁹ and upper bound a.

  输入下限、上限和自由度。求P(T < a)时，下限输入 −10⁹⁹，上限输入 a。

• For P(T > a), use lower bound a and upper bound 10⁹⁹.

  求P(T > a)时，下限输入 a，上限输入 10⁹⁹。

On the TI-Nspire CX, navigate to Menu → Statistics → Distributions → t Cdf, and enter the lower bound, upper bound, and df.

在TI-Nspire CX上，进入 Menu → Statistics → Distributions → t Cdf，输入下限、上限和自由度即可。

---

6. One-Tailed and Two-Tailed Probabilities | 单尾与双尾概率

Understanding whether a question requires a one-tailed or two-tailed probability is crucial. A one-tailed probability corresponds to the area in a single tail of the distribution; a two-tailed probability corresponds to the total area in both tails.

判断题目需要单尾还是双尾概率至关重要。单尾概率对应分布中单个尾部的面积；双尾概率则对应两个尾部的面积之和。

For a t-distribution with df = 10, suppose we want P(T > 1.812). From the table, this is 0.05 — a one-tailed probability. The two-tailed probability P(|T| > 1.812) = P(T < −1.812) + P(T > 1.812) = 0.05 + 0.05 = 0.10.

对于自由度为10的t分布，假设我们要求P(T > 1.812)。查表可知该概率为0.05——这是单尾概率。而双尾概率P(|T| > 1.812) = P(T < −1.812) + P(T > 1.812) = 0.05 + 0.05 = 0.10。

In hypothesis testing, a one-tailed test uses a single tail area, while a two-tailed test splits the significance level equally between both tails. Always read the question carefully to identify which type of probability is required.

在假设检验中，单尾检验使用单个尾部面积，而双尾检验将显著性水平平均分到两个尾部。务必仔细审题，判断题目要求的是哪种概率。

---

7. Finding P(T < a) | 求P(T < a)

When asked to find the probability that a t-random variable is less than a given positive value a, you need the cumulative area under the t-curve from negative infinity up to a.

当题目要求t随机变量小于某个正数a的概率时，你需要计算t曲线下从负无穷到a的累积面积。

Example | 例题

Let T follow a t-distribution with df = 15. Find P(T < 1.753).

设T服从自由度为15的t分布，求P(T < 1.753)。

Solution | 解答

• Using a GDC, enter tcdf(−10⁹⁹, 1.753, 15).

  使用GDC，输入tcdf(−10⁹⁹, 1.753, 15)。

• The result is approximately 0.9500.

  结果约为0.9500。

• Alternatively, from the table for df = 15, the value 1.753 corresponds to a right-tail probability of 0.05, so the left-tail probability is 1 − 0.05 = 0.95.

  或者，查表可知自由度为15时，1.753对应的右尾概率为0.05，因此左尾概率为1 − 0.05 = 0.95。

This shows the connection between table lookup and calculator methods: the table gives tail areas, and the complement rule gives the cumulative probability.

这体现了查表法与计算器法之间的联系：表格给出尾部面积，而互补法则给出累积概率。

---

8. Finding P(T > a) | 求P(T > a)

To find the probability that T exceeds a positive value a, we can either use the GDC directly or use the complement rule: P(T > a) = 1 − P(T ≤ a).

要求T超过某个正数a的概率，可以直接使用GDC，也可以利用互补法则：P(T > a) = 1 − P(T ≤ a)。

Example | 例题

Let T follow a t-distribution with df = 8. Find P(T > 2.306).

设T服从自由度为8的t分布，求P(T > 2.306)。

Solution | 解答

• Using a GDC, enter tcdf(2.306, 10⁹⁹, 8).

  使用GDC，输入tcdf(2.306, 10⁹⁹, 8)。

• The result is 0.0250.

  结果为0.0250。

• Using the table, for df = 8 and right-tail probability 0.025, the critical value is exactly 2.306. So P(T > 2.306) = 0.025.

  查表可知，自由度为8、右尾概率为0.025时，临界值正好是2.306。因此P(T > 2.306) = 0.025。

This example demonstrates that the t-table entries are, in fact, the critical values corresponding to standard tail probabilities.

这个例子说明，t分布表中的数值正是与标准尾部概率对应的临界值。

---

9. Finding P(a < T < b) | 求P(a < T < b)

For an interval probability, we subtract the cumulative probability at the lower bound from the cumulative probability at the upper bound: P(a < T < b) = P(T < b) − P(T < a).

对于区间概率，我们用上限处的累积概率减去下限处的累积概率：P(a < T < b) = P(T < b) − P(T < a)。

Example | 例题

Let T follow a t-distribution with df = 20. Find P(−1.725 < T < 2.086).

设T服从自由度为20的t分布，求P(−1.725 < T < 2.086)。

Solution | 解答

• Using a GDC, enter tcdf(−1.725, 2.086, 20).

  使用GDC，输入tcdf(−1.725, 2.086, 20)。

• The calculator gives approximately 0.9350.

  计算器输出约为0.9350。

• Using the table: P(T < 2.086) = 0.975 and P(T < −1.725) = 0.05. Thus the interval probability = 0.975 − 0.05 = 0.925. Note: the slight difference is due to rounding of t-values in the table.

  查表计算：P(T < 2.086) = 0.975，P(T < −1.725) = 0.05。因此区间概率 = 0.975 − 0.05 = 0.925。注意：细微差异源于表中t值的四舍五入。

This highlights a key advantage of GDC: it gives more precise probabilities without rounding errors from table lookup.

这体现了GDC的一大优势：它能给出更精确的概率，避免查表时因四舍五入产生的误差。

---

10. Inverse Problem: Finding t Given a Probability | 逆问题：已知概率求t值

IB exams often require you to work backwards: given a probability, find the corresponding t-value. This is called the inverse t-distribution problem.

IB考试经常要求你反向推导：已知概率，求对应的t值。这被称为t分布的逆问题。

Example | 例题

Let T follow a t-distribution with df = 12. Find the value of t such that P(T < t) = 0.95.

设T服从自由度为12的t分布，求满足P(T < t) = 0.95的t值。

Solution | 解答

• Using a GDC, use the inverse t function: invT(0.95, 12) on TI-84, or use the inverse distribution menu on TI-Nspire.

  使用GDC的逆t函数：在TI-84上输入invT(0.95, 12)，或在TI-Nspire上使用逆分布菜单。

• The result is approximately 1.782.

  结果约为1.782。

• Using the table: look for df = 12 and one-tailed probability 0.05 (since P(T > t) = 0.05 when P(T < t) = 0.95). The table gives t = 1.782.

  查表法：找到自由度为12、单尾概率0.05（因为P(T < t) = 0.95时P(T > t) = 0.05）。表中给出t = 1.782。

For a two-tailed inverse problem, e.g., P(|T| > t) = 0.05 with df = 12, the required t is the critical value for α = 0.025 in each tail, which is 2.179.

对于双尾逆问题，例如df = 12时P(|T| > t) = 0.05，所需的t值对应每尾α = 0.025的临界值，即2.179。

---

11. Common Errors and Exam Tips | 常见错误与考试技巧

Students often confuse the t-distribution with the standard normal distribution. Remember to check whether the population standard deviation is known: if σ is unknown and s is used, the t-distribution is required.

学生经常混淆t分布与标准正态分布。记住要检查总体标准差是否已知：如果σ未知而使用s，则必须使用t分布。

Another common error is using the wrong degrees of freedom. For a one-sample t-procedure, df = n − 1; for a two-sample t-procedure, the df formula is more complex and is often computed by the GDC.

另一个常见错误是使用错误的自由度。对于单样本t过程，df = n − 1；对于双样本t过程，自由度公式更复杂，通常由GDC自动计算。

• Always write down the distribution you are using: \( T \sim t(df) \).

  务必写出你使用的分布：\( T \sim t(df) \)。

• When using tcdf, remember that the lower bound must be less than the upper bound.

  使用tcdf时，记住下限必须小于上限。

• Use −10⁹⁹ and 10⁹⁹ as approximations for negative and positive infinity.

  用−10⁹⁹和10⁹⁹分别近似负无穷和正无穷。

• If the question asks for P(T > a) and a is negative, the result will be greater than 0.5.

  如果题目求P(T > a)且a为负数，结果必定大于0.5。

• In hypothesis testing, always draw a sketch of the t-distribution and shade the relevant region — this helps you avoid sign errors.

  在假设检验中，务必画出t分布草图并标出相关区域——这有助于避免符号错误。

When using the t-table, never confuse one-tailed and two-tailed headings. The top row of a typical table may say “one tail” or “two tails” — always verify which one you are reading.

使用t分布表时，切勿混淆单尾和双尾的列标题。典型表格的顶行可能标注”单尾”或”双尾”——务必确认你正在读取的是哪一类。

---

12. Summary: Step-by-Step Method | 总结：分步方法

To find probabilities in a t-distribution, follow this systematic approach:

要在t分布中求概率，请遵循以下系统化步骤：

Step | 步骤	Action | 操作
1	Identify the degrees of freedom df. | 确定自由度df。
2	Determine the required probability form: P(T < a), P(T > a), or P(a < T < b). | 确定所需概率形式：P(T < a)、P(T > a)或P(a < T < b)。
3	Use the GDC’s tcdf function with correct bounds and df. | 使用GDC的tcdf函数，输入正确的边界和df。
4	If using the table, locate df in the first column and read the critical value for the given tail probability. | 若查表，在第一列找到df，读取对应尾部概率的临界值。
5	For inverse problems, use invT on the GDC or find the critical value from the table. | 对于逆问题，使用GDC的invT函数或从表中查找临界值。

Mastering the t-distribution is essential for IB Math exams, especially for Paper 2 and the internal assessment. With regular practice using both GDC and tables, you will build confidence and accuracy.

掌握t分布对IB数学考试至关重要，尤其是Paper 2和内部评估（IA）。通过定期使用GDC和查表进行练习，你将逐步建立信心并提高准确率。

---

Published by TutorHao | IB Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）