📚 Understanding Percentage Points of Student's t-Distribution | 理解学生t分布的百分位点
The Student's t-distribution is a fundamental probability distribution in statistics, extensively used in A-Level Mathematics for constructing confidence intervals and conducting hypothesis tests on the mean of a normally distributed population when the population standard deviation is unknown. The percentage points of the t-distribution, listed in statistical tables, provide the critical values needed to determine rejection regions and p-value boundaries. Mastering how to read and apply these percentage points is essential for success in Edexcel exam questions on sampling and hypothesis testing.
学生t分布是统计学中一个基本的概率分布，在A-Level数学中被广泛用于在总体标准差未知的情况下，对服从正态分布的总体均值构建置信区间和进行假设检验。t分布的百分位点列于统计表中，提供了确定拒绝域和p值边界所需的临界值。掌握如何读取和应用这些百分位点对于在爱德思考试中成功解答抽样和假设检验题目至关重要。
1. Introduction to Student's t-Distribution | 学生t分布简介
The Student's t-distribution arises naturally when we estimate the mean of a normal distribution using a sample, but we must estimate the standard deviation from the same sample. Unlike the normal distribution, the t-distribution is heavier-tailed, particularly when the sample size is small. As the degrees of freedom increase, the t-distribution approaches the standard normal distribution N(0,1).
当我们使用样本估计正态分布的均值，但同时必须从同一样本估计标准差时，自然就产生了学生t分布。与正态分布不同，t分布具有更厚的尾部，尤其是在样本量较小的时候。随着自由度的增加，t分布趋近于标准正态分布N(0,1)。
2. What Are Percentage Points? | 什么是百分位点？
ch that the probability of the t-statistic exceeding that value is a given percentage, α, under the null hypothesis. For a one-tailed test, the upper-tail percentage point is the value that cuts off area α in the right tail. A two-tailed percentage point corresponds to splitting α equally into both tails, so the critical values have absolute value that leaves α/2 in each tail.
在t分布的语境下，百分位点（通常记为t_α,ν 或 t临界值）是指这样一个值：在原假设下，t统计量超过该值的概率等于给定的百分比α。对于单尾检验，上尾百分位点是在右尾截去面积α的值。双尾百分位点则对应于将α均分到两个尾部，因此临界值的绝对值在每个尾部留出α/2的面积。
3. The t-Distribution and Degrees of Freedom | t分布与自由度
size n. A smaller ν produces a more spread-out distribution with fatter tails, reflecting greater uncertainty due to estimating the standard deviation. As ν increases, the t-distribution becomes more concentrated around zero and its tails become lighter, matching the standard normal distribution when ν → ∞.
t分布的形状完全取决于一个称为自由度（ν）的参数，对于大小为n的单个样本，自由度通常等于n−1。较小的ν会产生更分散且尾部更厚的分布，反映了由于估计标准差而带来的更大不确定性。随着ν增加，t分布变得更集中于零附近，尾部变轻，当ν→∞时与标准正态分布一致。
4. Structure of the t-Table | t分布表的结构
o 50, and then jumps to 60, 80, 100, 1000, and infinity (the normal distribution). The top row gives the tail probabilities (p). For one-tailed tests, the probabilities are 0.10, 0.05, 0.025, 0.01, and 0.005. For two-tailed tests, the corresponding probabilities would be 0.20, 0.10, 0.05, 0.02, and 0.01 respectively, but the table usually labels the column by the one-tail probability, so a two-tailed test at the 5% significance level uses the p = 0.025 column.
爱德思公式手册提供了t分布的临界值表。最左列列出自由度（ν），从1到50，然后跳到60、80、100、1000和无穷大（正态分布）。顶行列出了尾部概率（p）。对于单尾检验，概率为0.10、0.05、0.025、0.01和0.005。对于双尾检验，对应的概率分别为0.20、0.10、0.05、0.02和0.01，但表格通常按单尾概率标记列，因此在5%显著性水平下的双尾检验使用p=0.025列。
A small excerpt of the table appears below:
下表是一个简短的摘录：
| ν | p=0.05 (one-tail) | p=0.025 (one-tail) |
|---|---|---|
| 1 | 6.314 | 12.706 |
| 5 | 2.015 | 2.571 |
| 10 | 1.812 | 2.228 |
| 20 | 1.725 | 2.086 |
| ∞ | 1.645 | 1.960 |
5. Reading One-Tail Percentage Points | 读取单尾百分位点
the required significance level α. For example, with ν = 15 and a one-tailed test at the 5% des sufficient evidence to reject the null hypothesis in the direction of the alternative.
要找到单尾临界值，先找到与自由度对应的行，然后移动到与所需显著性水平α匹配的列。例如，当ν = 15且进行5%水平的单尾检验（α = 0.05）时，表上给出t(15) = 1.753。这意味着t统计量大于1.753时，就有足够的证据在备择假设的方向上拒绝原假设。
6. Reading Two-Tail Percentage Points | 读取双尾百分位点
d test at significance level α, each tail contains α/2. The critical values are symmetric: – el table, you use the one-tail probability equal to α/2. So for a two-tailed test at the 5% p = 0.025. With ν = 10, the critical value is t(10) = 2.228. The rejection regions are t < –2.228 and t > 2.228.
对于显著性水平为α的双尾检验，每个尾部包含α/2的区域。临界值是对称的：–t和+t。在爱德思的表格中，你使用等于α/2的单尾概率来查表。因此，对于5%显著性水平的双尾检验，应查看p = 0.025列。当ν = 10时，临界值为t(10) = 2.228。拒绝域为t < –2.228和t > 2.228。
7. Critical Values and Hypothesis Testing | 临界值与假设检验
n a hypothesis test for a population mean with unknown variance, the test statistic t = (x̄ − μ₀) / (s/√n) follows a t-distribution with n−1 degrees of freedom under H₀. By comparing the absolute value of the calculated t with the critical value from the table, you decide whether the observed sample is significantly different. If |t| > critical value, reject H₀.
在方差未知时对总体均值进行假设检验，检验统计量t = (x̄ − μ₀) / (s/√n)在原假设下服从自由度为n−1的t分布。通过将计算出的t的绝对值与表中的临界值进行比较，你判断观察到的样本是否显著不同。如果|t| > 临界值，则拒绝H₀。
t = (x̄ − μ₀) / (s/√n) ~ tₙ₋₁
8. Confidence Intervals Using t-Values | 使用t值构建置信区间
A confidence interval for the population mean μ when σ is unknown is given by x̄ ± t_critical × (s/√n t-value corresponds to a two-tailed probability equal to (1 − confidence level). For a 95% is 0.025, so you look up t₀.₀₂₅,ν. For example, with a sample of size 20 (ν = 19), the 95% confidence interval uses t(19) = 2.093.
当σ未知时，总体均值μ的置信区间由x̄ ± t临界值 × (s/√
Published by TutorHao | A-Level Mathematics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图