📚 The Variance of a Normal Distribution | 正态分布的方差
In Edexcel A-Level Mathematics, the variance of a normal distribution is a fundamental parameter that measures the spread of data around the mean. Understanding variance allows you to standardise variables, perform hypothesis tests, and construct confidence intervals. This article covers everything you need to know about variance in the context of the normal distribution, from basic definitions to exam-style applications.
在爱德思A-Level数学中，正态分布的方差是衡量数据围绕均值的离散程度的基本参数。理解方差可以帮助你对变量进行标准化、执行假设检验并构建置信区间。本文涵盖正态分布中方差的全部必备知识，从基础定义到考试题型应用。
1. The Variance Parameter σ² | 方差参数 σ²
The normal distribution is fully defined by two parameters: the mean μ and the variance σ². While the mean locates the centre of the distribution, the variance controls how widely the values are spread. A larger σ² produces a flatter, wider bell curve; a smaller σ² yields a taller, narrower curve.
正态分布完全由两个参数定义：均值 μ 和方差 σ²。均值决定了分布的中心位置，而方差则控制值的离散程度。较大的 σ² 会产生更扁平、更宽的钟形曲线；较小的 σ² 则导致更高、更窄的曲线。
The variance is always positive, and its units are the square of the units of the variable. For instance, if X represents height in metres, then σ² is in m². The square root of the variance, σ, is the standard deviation and has the same units as the data.
方差始终为正数，其单位是变量单位的平方。例如，若 X 表示身高（单位为米），则 σ² 的单位是 m²。方差的平方根 σ 就是标准差，与数据的单位相同。
2. Definition and Notation | 定义与符号
For a random variable X ~ f size n, the sample variance s² is used as an estimate of σ², calculated by s² = Σ(xᵢ − x̄)² / (n − 1) for an unbiased estimator.
对于随机变量 X ~ N(μ, σ²)，方差是偏离均值的平方的期望值：σ² = E[(X − μ)²]。在实践中，当我们处理容量为 n 的样本时，样本方差 s² 被用作 σ² 的估计值，其无偏估计量的计算公式为 s² = Σ(xᵢ − x̄)² / (n − 1)。
The notation used in Edexcel exams distinguishes clearly between population variance σ² and sample variance s². You will often be given σ² or asked to assume it is known for the purpose of testing the mean. If σ² is unknown, the sample variance s² replaces it, and the t-distribution becomes relevant.
爱德思考试中使用的符号明确区分了总体方差 σ² 和样本方差 s²。你经常会拿到 σ² 的数值，或者被要求在检验均值时假设其已知。如果 σ² 未知，则用样本方差 s² 代替，此时需要用到 t 分布。
3. Properties of Variance in Normal Distributions | 正态分布中方差的特性
: Var(X ± Y) = Var(X) + Var(Y). This additive property underpins many statistical procedures, such as combining uncertainties.
独立性是一个关键性质：如果两个正态变量 X 和 Y 独立，它们的和或差的方差等于各自方差之和：Var(X ± Y) = Var(X) + Var(Y)。这种可加性是许多统计方法的基础，例如不确定度的合并。
Note that even though X ± Y is normally distributed, the variance is always additive — independent variables, variances add, whether you are adding or subtracting the variables.
请注意，尽管 X ± Y 服从正态分布，方差始终是相加的——你绝不应该将方差相减。这是考试题目中常见的错误来源，因此请记住：对于独立变量，无论变量是相加还是相减，方差总是相加。
4. Standardising: From N(μ, σ²) to N(0,1) | 标准化：从 N(μ, σ²) 到 N(0,1)
Standardising converts any normal random variable X ~ The resulting Z has a mean of 0 and a variance of 1. This step is essential for using probability tables, as exam boards provide cumulative probabilities for Z ~ N(0,1).
标准化使用公式 Z = (X − μ) / σ 将任意正态随机变量 X ~ N(μ, σ²) 转化为标准正态变量 Z。所得的 Z 均值为 0，方差为 1。由于考试机构提供 Z ~ N(0,1) 的累积概率表，这一步对于查表求概率至关重要。
Variance plays a crucial role here: dividing by σ scales the squared deviations, transforming the variance from σ² to 1. If the variance were zero, standardisation would be impossible, but that would imply a degenerate distribution where all values are identical — not a true normal distribution.
在这里方差起着关键作用：除以 σ 可以缩放离差的平方，将方差从 σ² 转换为 1。如果方差为零，标准化就无法进行，但那意味着所有值都相同的退化分布——并非真正的正态分布。
5. Effect of Linear Transformations on Variance | 线性变换对方差的影响
shifts the mean but leaves the variance unchanged. This is widely tested in Edexcel problems, especially when combining normal variables.
如果随机变量 X 的方差 Var(X) = σ²，那么线性变换 Y = aX + b 会将方差变为 Var(Y) = a² σ²。加法常数 b 会移动均值，但方差保持不变。这在爱德思的考题中广泛涉及，尤其是在组合正态变量时。
Be careful: the standard deviation scales by |a|, not a. Students often mistakenly multiply the variance by a instead of a². A quick check is to consider a simple case: doubling each data value multiplies the variance by 4, which makes sense because the squared distances from the mean increase fourfold.
请注意：标准差按 |a| 缩放，而非 a。学生常常错误地给方差乘以 a 而不是 a²。一个快速检验方法是考虑一个简单情况：将每个数据值翻倍会使方差乘以 4，这很合理，因为与均值的平方距离变成了原来的四倍。
6. Sample Variance as an Unbiased Estimator | 作为无偏估计量的样本方差
In Edexcel Statistics, the sample variance s² = Σ(xᵢ − x̄ nce σ². The division by n − 1, rather than n, corrects for the fact that the sample mean x̄ is used instead of the true μ, which costs one degree of freedom.
在爱德思统计学中，样本方差 s² = Σ(xᵢ − x̄)² / (n − 1) 是总体方差 σ² 的无偏估计量。除以 n − 1 而非 n，可以修正因使用样本均值 x̄ 而非真实 μ 而带来的偏差，这消耗了一个自由度。
ulation variance is unknown and you have a sample, you must use s² in further calculations, such as estimating the variance of the sample mean or carrying out a t-test. Failing to use n − 1 is a classic mark-losing mistake.
S2 的考试大纲要求你认识到 E(s²) = σ²。当总体方差未知而你有一个样本时，你必须在进一步的计算中使用 s²，例如估计样本均值的方差或执行 t 检验。没有使用 n − 1 是一个典型的丢分错误。
7. Distribution of the Sample Mean: Variance σ²/n | 样本均值的分布：方差 σ²/n
If a random sample of size n is drawn from X ~ N(μ, σ²), then the sample mean X̄ follows a normal distribution with the same mean μ but with a reduced variance: Var(X̄ . This shrinking variance reflects the increased precision that comes with larger samples.
若从 X ~ N(μ, σ²) 中抽取一个容量为 n 的随机样本，则样本均值 X̄ 服从正态分布，均值仍为 μ，但方差减小了：Var(X̄) = σ² / n。方差的缩小反映了样本容量增大带来的精度提升。
When σ² is unknown, the estimated variance of X̄ becomes s²/n. In this case, the distribution of the standardised sample mean follows a t-distribution with n − 1 degrees of freedom, which is a required concept for confidence intervals and hypothesis tests in S2.
当 σ² 未知时，X̄ 的估计方差变为 s²/n。在这种情况下，标准化的样本均值服从自由度为 n − 1 的 t 分布，这是 S2 中置信区间和假设检验所要求的概念。
8. Using Sample Variance in Hypothesis Tests (t-tests) | 在假设检验中使用样本方差（t 检验）
one-sample t-test for the population mean when σ² is unknown. The test statistic is t = (x̄ − μ₀) / (s/√n), which uses the sample standard deviation s as a plug-in estimator for σ. The critical values come from the t-distribution.
爱德思A-Level数学经常考查在 σ² 未知时对总体均值执行单样本 t 检验的能力。检验统计量为 t = (x̄ − μ₀) / (s/√n)，它使用样本标准差 s 作为 σ 的替代估计量。临界值来自 t 分布。
It is essential to adjust the degrees of freedom to n − 1 and to use the correct tail probability. The variance estimate s² directly influences the size of the test statistic: a smaller s makes the denominator smaller and t larger, potentially pushing the result into the rejection region.
必须将自由度调整为 n − 1，并使用正确的尾部概率。方差估计值 s² 直接影响检验统计量的大小：较小的 s 使分母变小、t 变大，从而可能将结果推入拒绝域。
9. Pooled Variance for Two-Sample t-Tests | 双样本 t 检验的合并方差
ance estimator. The pooled variance sp² = [(n₁−1)s₁² + (n₂−1)s₂²] / (n₁ + n₂ − 2) combines information from both samples to estimate the common σ².
当比较两个独立正态总体的均值、它们的方差未知但假设相等时，爱德思要求使用合并方差估计量。合并方差 sp² = [(n₁−1)s₁² + (n₂−1)s₂²] / (n₁ + n₂ − 2) 综合了两个样本的信息来估计共同的 σ²。
This pooled variance is then used in the two-sample t-test statistic. The assumption of equal variances must be stated; it is part of the model and should be checked against any given information. Remember that the degrees of freedom become n₁ + n₂ − 2 in this scenario.
然后将此合并方差用于双样本 t 检验统计量。必须声明方差相等的假设；它是模型的一部分，并应根据给定的信息进行检验。请记住，在这种情况下自由度变为 n₁ + n₂ − 2。
10. Confidence Intervals for the Mean (Variance Unknown) | 均值的置信区间（方差未知）
When σ² is unknown, a 100(1 − α)% confidence interval for μ is given by x̄ ± tα/2, n−1 × (s/√n). Here, tα/2, n−1 is the critical value from the t-distribution with n − 1 degrees of freedom. The interval width depends on s, illustrating how the sample variance directly affects the precision of estimation.
当 σ² 未知时，μ 的 100(1 − α)% 置信区间由 x̄ ± tα/2, n−1 × (s/√n) 给出。其中 tα/2, n−1 是自由度为 n − 1 的 t 分布的临界值。区间宽度取决于 s，这说明了样本方差如何直接影响估计的精度。
If σ² were known, the interval would use the z-value instead of t. The switch from z to t accounts for the extra uncertainty introduced by estimating the variance. As n grows large, the t-distribution approaches N(0,1), so the distinction becomes negligible for large samples.
如果 σ² 已知，区间将使用 z 值而非 t 值。从 z 到 t 的转换，是为了考虑因估计方差而引入的额外不确定性。随着 n 增大，t 分布趋近于 N(0,1)，因此对于大样本，这种差异可以忽略。
11. Common Mistakes and Pitfalls | 常见错误与陷阱
r is confusing variance and standard deviation when reading questions. If a problem states 'standard deviation = 5' distribution formulas. Similarly, when calculating Z = (X − μ) / σ, always use σ, not σ².
一个常见错误是在读题时混淆方差和标准差。如果问题给出‘标准差 = 5’，那么 σ² = 25，你必须在应用正态分布公式之前将其平方。同样地，计算 Z = (X − μ) / σ 时，始终应使用 σ 而非 σ²。
Another pitfall is forgetting to divide by n when working with the sample mean. Students often apply the population variance directly to X̄ instead of using σ²/n. Finally, when the variance is unknown, many candidates incorrectly use the normal distribution critical values instead of the t-table.
另一个陷阱是在处理样本均值时忘记除以 n。学生经常将总体方差直接应用到 X̄ 上，而不是使用 σ²/n。最后，当方差未知时，很多考生错误地使用了正态分布临界值，而不是查 t 表。
12. Summary and Exam Strategy | 总结与考试策略
The variance of a normal distribution is far more than a simple measure of spread — it is a gateway to standardisation, estimation, and inference. For Edexcel exams, ensure you can quickly identify whether σ² is known or must be estimated, and decide correctly between z-procedures and t-procedures.
正态分布的方差远不止是一个离散程度指标——它是通往标准化、估计和推断的大门。在爱德思考试中，请确保你能快速识别 σ² 是已知还是需要估计，并正确地在 z 方法和 t 方法之间做出选择。
Practice writing down the parameters clearly: state the distribution of X, X̄, and any transformed variables with their variances. Show your scale factor a² clearly when combining variables, and always use unbiased estimators with n − 1 when working with sample data. Attention to these details will secure full marks on variance-related questions.
练习时请清晰地写下参数：陈述 X、X̄ 以及任何变换变量的分布及其方差。在组合变量时要清楚地写出缩放因子 a²，并在使用样本数据时始终使用包含 n − 1 的无偏估计量。注意这些细节，这将助你在与方差相关的题目中拿到满分。
Published by TutorHao | 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