📚 Freedom or Liberty: Degrees of Freedom in Edexcel A-Level Statistics | 自由还是自由度：Edexcel A-Level 统计学中的自由度概念
In everyday speech, “freedom” and “liberty” are almost interchangeable: both describe the absence of constraint. In Edexcel A-Level Statistics, however, the word “freedom” takes on a precise technical meaning through the concept of degrees of freedom. This article explains what degrees of freedom are, why they appear in formulae such as sample variance, and how they affect key distributions and hypothesis tests.
在日常用语中，“freedom” 和 “liberty” 几乎可以互换使用，两者都描述不受限制的状态。然而，在 Edexcel A-Level 统计学中，“freedom” 通过自由度的概念获得了精确的技术含义。本文将解释什么是自由度、为什么它们出现在诸如样本方差等公式中，以及它们如何影响关键分布和假设检验。
1. What Does “Degrees of Freedom” Mean? | 什么是“自由度”？
Degrees of freedom, often abbreviated as df, refer to the number of independent values or quantities that can vary in a calculation without breaking any given constraints. Once a sample statistic, such as the mean, is fixed, the remaining data values lose some of their freedom to vary independently.
自由度，通常缩写为 df，指的是在一个计算中可以独立变化而不违反任何给定约束的数值或量的个数。一旦某个样本统计量（例如均值）被固定，剩余的数据值就失去了部分独立变化的自由。
For example, suppose three numbers have a mean of 10. The total is therefore 30. If the first two numbers are chosen freely as 8 and 12, then the third number must be 10 to make the total 30. Only two of the three numbers are free to vary, so the degrees of freedom are 2.
例如，假设有三个数，其均值为 10。因此总和为 30。如果前两个数可以自由选择为 8 和 12，那么第三个数必须是 10 才能使总和为 30。三个数中只有两个可以自由变化，因此自由度为 2。
2. The Sample Variance and the n − 1 Formula | 样本方差与 n − 1 公式
In Edexcel A-Level Statistics, one of the first places students meet degrees of freedom is in the formula for sample variance. When the population mean μ is unknown and must be estimated by the sample mean x̄, the sample variance is given by:
在 Edexcel A-Level 统计学中，学生最早接触自由度的地方之一是样本方差公式。当总体均值 μ 未知而必须用样本均值 x̄ 来估计时，样本方差由以下公式给出：
s² = Σ(xᵢ − x̄)² ÷ (n − 1)
The divisor (n − 1) rather than n is not arbitrary. It is exactly the degrees of freedom associated with the sample. Using n − 1 ensures that the sample variance is an unbiased estimator of the population variance σ².
除数使用 (n − 1) 而不是 n 并非随意。它正是与样本相关的自由度。使用 n − 1 可以保证样本方差是总体方差 σ² 的无偏估计量。
3. Why Divide by n − 1? An Intuitive Explanation | 为什么除以 n − 1？直观解释
When we calculate the deviations (xᵢ − x̄), the sum of all deviations is always zero. This is a fixed constraint: Σ(xᵢ − x̄) = 0. Because of this single constraint, only n − 1 of the deviations can vary independently. The last deviation is determined automatically by the others.
当我们计算离差 (xᵢ − x̄) 时，所有离差的总和始终为零。这是一个固定的约束条件：Σ(xᵢ − x̄) = 0。由于这个单一约束，只有 n − 1 个离差可以独立变化。最后一个离差由其他离差自动决定。
If we divided by n instead, we would tend to underestimate the true population variance, especially for small samples. Dividing by n − 1 corrects this bias by accounting for the fact that one degree of freedom has been used up in estimating the mean.
如果我们除以 n，往往会低估真实的总体方差，尤其是在小样本情况下。除以 n − 1 可以纠正这种偏差，因为它考虑了估计均值时已经用掉了一个自由度。
4. Degrees of Freedom in the t-Distribution | t 分布中的自由度
The t-distribution is used in Edexcel A-Level Statistics when the population standard deviation σ is unknown and must be estimated by the sample standard deviation s. The shape of the t-distribution depends entirely on its degrees of freedom, which is usually n − 1 for a one-sample t-test.
当总体标准差 σ 未知而必须用样本标准差 s 来估计时，Edexcel A-Level 统计学中会使用 t 分布。t 分布的形状完全取决于其自由度，对于单样本 t 检验，自由度通常为 n − 1。
For small degrees of freedom, the t-distribution has heavier tails than the normal distribution, reflecting greater uncertainty. As the degrees of freedom increase, the t-distribution approaches the standard normal distribution N(0, 1).
对于较小的自由度，t 分布的尾部比正态分布更厚，反映出更大的不确定性。随着自由度的增加，t 分布趋近于标准正态分布 N(0, 1)。
t = (x̄ − μ) ÷ (s ÷ √n)
Here the sample size n determines the degrees of freedom ν = n − 1.
这里样本容量 n 决定了自由度 ν = n − 1。
5. Degrees of Freedom in the Chi-Squared Distribution | 卡方分布中的自由度
The chi-squared distribution, denoted χ², arises from the sum of squares of independent standard normal variables. If Z₁, Z₂, …, Zₖ are independent N(0, 1) variables, then their sum of squares follows a chi-squared distribution with k degrees of freedom:
卡方分布，记作 χ²，来自独立标准正态变量的平方和。如果 Z₁, Z₂, …, Zₖ 是独立的 N(0, 1) 变量，那么它们的平方和服从自由度为 k 的卡方分布：
χ² = Z₁² + Z₂² + … + Zₖ²
The number of independent standard normal variables, k, is exactly the degrees of freedom. The mean of a chi-squared distribution is k, and its variance is 2k.
独立标准正态变量的个数 k 正是自由度。卡方分布的均值为 k，方差为 2k。
6. Chi-Squared Tests and Degrees of Freedom | 卡方检验与自由度
In chi-squared hypothesis tests, such as tests for independence in contingency tables or goodness-of-fit tests, the degrees of freedom determine the appropriate critical value from the χ² table. For a contingency table with r rows and c columns, the degrees of freedom are calculated as:
在卡方假设检验中，例如列联表的独立性检验或拟合优度检验，自由度决定了从 χ² 表中查得的临界值。对于具有 r 行和 c 列的列联表，自由度按下式计算：
df = (r − 1)(c − 1)
This formula counts the number of independent cell frequencies after the row and column totals have been fixed. For a goodness-of-fit test with k categories, the degrees of freedom are usually k − 1, because the total frequency is fixed.
该公式计算在行合计和列合计固定后独立的单元格频数个数。对于具有 k 个类别的拟合优度检验，自由度通常为 k − 1，因为总频数是固定的。
7. Degrees of Freedom in Regression and ANOVA | 回归与方差分析中的自由度
In linear regression, when estimating both the intercept and slope from n paired observations, the residual degrees of freedom are n − 2. Two degrees of freedom are lost because two parameters have been estimated from the data.
在线性回归中，当从 n 对观测值中同时估计截距和斜率时，残差自由度为 n − 2。损失了两个自由度，因为从数据中估计了两个参数。
In one-way analysis of variance, the total variability is partitioned into between-group and within-group components. If there are k groups and n total observations, the between-group degrees of freedom are k − 1, and the within-group degrees of freedom are n − k.
在单因素方差分析中，总变异被分解为组间变异和组内变异。如果有 k 个组和 n 个总观测值，则组间自由度为 k − 1，组内自由度为 n − k。
8. Common Mistakes and Exam Tips | 常见错误与应试技巧
Students often use n instead of n − 1 when calculating sample variance. Always ask whether the data is a population or a sample. If the question says “a sample of size n”, use the n − 1 divisor unless told otherwise.
学生在计算样本方差时经常使用 n 而不是 n − 1。始终要问清楚数据是总体还是样本。如果题目说“一个容量为 n 的样本”，除非另有说明，否则使用 n − 1 作为除数。
For chi-squared tests, do not forget that degrees of freedom depend on the table dimensions, not on the total sample size directly. A 2 × 2 table has df = (2 − 1)(2 − 1) = 1, no matter how large the total frequency is.
对于卡方检验，不要忘记自由度取决于表格的维度，而不是直接取决于总样本量。2 × 2 列联表的自由度为 df = (2 − 1)(2 − 1) = 1，无论总频数有多大。
When using statistical tables for the t-distribution or chi-squared distribution, always match the correct degrees of freedom. A wrong df can lead to an incorrect critical value and therefore an incorrect conclusion.
在查 t 分布表或卡方分布表时，始终要匹配正确的自由度。错误的自由度会导致错误的临界值，进而得出错误的结论。
9. Worked Example: Sample Variance with n − 1 | 例题：使用 n − 1 的样本方差
Five measurements are taken from a sample: 4, 6, 8, 10, 12. Calculate the sample variance.
从一个样本中取得五个测量值：4、6、8、10、12。计算样本方差。
First find the sample mean: x̄ = (4 + 6 + 8 + 10 + 12) ÷ 5 = 8. Then compute the squared deviations: (4 − 8)² = 16, (6 − 8)² = 4, (8 − 8)² = 0, (10 − 8)² = 4, (12 − 8)² = 16. Their sum is 40.
首先计算样本均值：x̄ = (4 + 6 + 8 + 10 + 12) ÷ 5 = 8。然后计算离差平方：(4 − 8)² = 16，(6 − 8)² = 4，(8 − 8)² = 0，(10 − 8)² = 4，(12 − 8)² = 16。它们的总和为 40。
Since this is a sample, divide by n − 1 = 4, not 5. Therefore s² = 40 ÷ 4 = 10. If we had divided by n = 5 instead, we would get 8, which underestimates the population variance.
由于这是一个样本，除以 n − 1 = 4，而不是 5。因此 s² = 40 ÷ 4 = 10。如果我们用 n = 5 去除，会得到 8，这会低估总体方差。
10. Worked Example: Degrees of Freedom in a Chi-Squared Test | 例题：卡方检验中的自由度
A market researcher collects data on gender and brand preference, producing a 2 × 3 contingency table. Determine the degrees of freedom for a chi-squared test of independence.
一位市场研究人员收集了性别和品牌偏好的数据，生成了一个 2 × 3 列联表。确定卡方独立性检验的自由度。
Here r = 2 and c = 3, so df = (r − 1)(c − 1) = (2 − 1)(3 − 1) = 1 × 2 = 2. The correct chi-squared distribution to use has 2 degrees of freedom.
这里 r = 2，c = 3，因此 df = (r − 1)(c − 1) = (2 − 1)(3 − 1) = 1 × 2 = 2。应使用的卡方分布具有 2 个自由度。
11. Summary: Freedom Means Independent Information | 总结：自由度意味着独立信息
In statistics, freedom is not a philosophical idea but a practical count of independent information. Every time a parameter is estimated from the data, one degree of freedom is used up. The remaining degrees of freedom appear in the divisor of sample variance, in the shape of the t and chi-squared distributions, and in the critical values for hypothesis tests.
在统计学中，自由度不是一个哲学概念，而是独立信息的实际计数。每当从数据中估计一个参数时，就会用掉一个自由度。剩余的自由度出现在样本方差的除数中、t 分布和卡方分布的形状中，以及假设检验的临界值中。
Mastering degrees of freedom helps Edexcel A-Level students choose the right formula, read statistical tables correctly, and explain their reasoning in exam answers. It is one of the core statistical ideas that link estimation, distributions, and hypothesis testing.
掌握自由度有助于 Edexcel A-Level 学生选择正确的公式、正确查阅统计表，并在考试答案中解释推理过程。它是连接估计、分布和假设检验的核心统计思想之一。
12. Key Formulae to Remember | 需要记住的关键公式
| Concept | Formula | Degrees of Freedom |
|---|---|---|
| Sample variance | s² = Σ(xᵢ − x̄)² ÷ (n − 1) | n − 1 |
| One-sample t-test | t = (x̄ − μ) ÷ (s ÷ √n) | n − 1 |
| Chi-squared test for independence | df = (r − 1)(c − 1) | (r − 1)(c − 1) |
| Chi-squared goodness-of-fit | df = k − 1 | k − 1 |
| Simple linear regression residuals | df = n − 2 | n − 2 |
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