📚 Variance and Standard Deviation | 方差与标准差
Variance and standard deviation are fundamental measures of dispersion in statistics. They quantify how much a set of data values spreads out around the mean. Understanding these concepts is essential for any IB Mathematics student, whether in Analysis & Approaches or Applications & Interpretation, as they appear in descriptive statistics, probability distributions, and inferential methods.
方差与标准差是统计学中最基本的离散度量。它们量化了一组数据值围绕均值的分散程度。对于任何 IB 数学学生，无论是在分析与方法还是应用与解释课程中，理解这些概念都至关重要，因为它们出现在描述性统计、概率分布和推断方法中。
1. What Is Variance? | 什么是方差？
Variance measures the average squared deviation of each data point from the mean. By squaring the deviations, we eliminate negative signs and give more weight to larger deviations. For a population, the variance is denoted by σ² (sigma squared).
方差衡量每个数据点与均值之差的平方的平均值。通过平方偏差，我们消除了负号，并赋予较大偏差更大的权重。对于总体，方差用 σ²（西格玛平方）表示。
Population variance formula:
σ² = ∑(xᵢ − μ)² / N
where μ is the population mean, xᵢ represents each value, and N is the population size.
总体方差公式：
σ² = ∑(xᵢ − μ)² / N
其中 μ 是总体均值，xᵢ 代表每个值，N 是总体大小。
If we ignore the squaring step, the sum of deviations from the mean is always zero, so variance uses squares to provide a meaningful non-negative summary of spread.
如果忽略平方步骤，均值偏差的总和始终为零，因此方差使用平方来提供一个有意义的非负离散程度总结。
2. Sample Variance: Why n−1? | 样本方差：为什么是 n−1？
When working with a sample rather than the whole population, the variance formula uses n−1 in the denominator instead of N. This correction, known as Bessel's correction, ensures that the sample variance is an unbiased estimator of the population variance.
当使用样本而不是整个总体时，方差公式分母使用 n−1 而不是 N。这一修正称为贝塞尔校正，可确保样本方差是总体方差的无偏估计量。
Sample variance formula:
s² = ∑(xᵢ − x̄)² / (n − 1)
where x̄ is the sample mean and n is the sample size.
样本方差公式：
s² = ∑(xᵢ − x̄)² / (n − 1)
其中 x̄ 是样本均值，n 是样本大小。
By dividing by n−1, we slightly increase the variance estimate to compensate for the fact that a sample tends to underestimate the true variability. The degrees of freedom lost in estimating the mean explain this adjustment.
通过除以 n−1，我们略微增大了方差估计值，以弥补样本往往低估真实变异性的这一事实。估计均值所损失的自由度解释了这一调整。
3. Standard Deviation: Definition and Interpretation | 标准差：定义与解释
Standard deviation is simply the positive square root of variance. It returns the spread measure to the original units of the data, making it easier to interpret. Population standard deviation is σ, and sample standard deviation is s.
标准差就是方差的非负平方根。它将离散度量还原为数据的原始单位，使其更易于解释。总体标准差为 σ，样本标准差为 s。
σ = √σ² and s = √s².
Because standard deviation shares units with the data, we often say things like “the scores have a mean of 70 with a standard deviation of 8 points”.
σ = √σ² 且 s = √s²。
由于标准差与数据共享单位，我们经常说诸如“分数的均值为 70，标准差为 8 分”之类的话。
A small standard deviation indicates that data points cluster closely around the mean, whereas a large one signals greater spread. In a normal distribution, roughly 68% of values lie within one standard deviation of the mean.
较小的标准差表示数据点紧密聚集在均值周围，而较大的标准差则表示分布更广。在正态分布中，大约 68% 的值落在均值的一个标准差范围内。
4. Calculating Variance and Standard Deviation by Hand | 手工计算方差和标准差
Manual computation follows a clear five-step process. First, find the mean of the data set. Second, subtract the mean from each data value to obtain the deviations. Third, square every deviation. Fourth, sum the squared deviations. Finally, divide by N (for population) or n−1 (for sample) and take the square root if standard deviation is needed.
手工计算遵循清晰的五步过程。首先，计算数据集的均值。其次，从每个数据值中减去均值得到偏差。第三，将每个偏差平方。第四，求偏差平方和。最后，除以 N（总体）或 n−1（样本），如果需要标准差则取平方根。
Example: Find the sample standard deviation of 5, 7, 9, 11.
Mean x̄ = (5+7+9+11)/4 = 8.
Deviations: −3, −1, 1, 3. Squares: 9, 1, 1, 9. Sum = 20.
Sample variance s² = 20/(4−1) = 6.67 (to 3 s.f.). Sample standard deviation s = √6.67 ≈ 2.58.
示例：求 5, 7, 9, 11 的样本标准差。
均值 x̄ = (5+7+9+11)/4 = 8。
偏差：−3, −1, 1, 3。平方：9, 1, 1, 9。和 = 20。
样本方差 s² = 20/(4−1) = 6.67（保留三位有效数字）。样本标准差 s = √6.67 ≈ 2.58。
Organising work in a table helps avoid arithmetic mistakes, especially with larger data sets. IB exam questions often expect you to show such a table.
用表格组织计算有助于避免算术错误，尤其是数据量较大时。IB 考试题目通常期望你展示这样一个表格。
5. Using Technology: GDC Tips | 使用技术：图形计算器技巧
IB Mathematics allows and encourages the use of a graphical display calculator (GDC) to compute variance and standard deviation efficiently. You simply enter the list of numbers and read the needed statistic.
IB 数学允许并鼓励使用图形显示计算器（GDC）来高效计算方差和标准差。你只需输入数字列表，然后读取所需的统计量。
On a TI-Nspire or TI-84, after entering data into a list, use the one-variable statistics command. The calculator will display both sx (sample standard deviation) and σx (population standard deviation). Be careful to select the one that matches the context of the problem.
在 TI-Nspire 或 TI-84 上，将数据输入列表后，使用单变量统计命令。计算器将同时显示 sx（样本标准差）和 σx（总体标准差）。注意选择与问题背景匹配的那个。
Many students lose marks by mixing up sx and σx. If the question states “the data represents a sample of …” use sx; if it states “the entire population” use σx. The same distinction applies to variance: square the appropriate standard deviation afterwards if required.
许多学生因混淆 sx 和 σx 而丢分。如果题目说“数据代表……的一个样本”，则使用 sx；如果说“整个总体”，则使用 σx。同样的区别也适用于方差：如果需要，之后将相应的标准差平方即可。
6. Variance and Standard Deviation for Grouped Data | 分组数据的方差与标准差
ecomes σ² = ∑f(m − μ)² / N, where f is the frequency, m is the class midpoint, and N = ∑f.
当数据以频率表的形式呈现时，使用组中值来估计方差。总体方差公式变为 σ² = ∑f(m − μ)² / N，其中 f 是频数，m 是组中值，N = ∑f。
For sample grouped data, s² = ∑f(m − x̄ s assumes that within each interval the data are evenly spread, so it is an approximation.
对于样本分组数据，s² = ∑f(m − x̄)² / (n − 1)，其中 n = ∑f。这假设每个区间内的数据是均匀分布的，因此它是一个近似值。
Your GDC can handle grouped data by entering midpoints in one list and frequencies in another, then using the two-variable statistics or frequency list option. Always check whether the intervals are continuous and that midpoints are correctly calculated.
你的 GDC 可以通过在一个列表中输入组中值，在另一个列表中输入频数，然后使用双变量统计或频率列表选项来处理分组数据。始终检查区间是否连续，以及组中值是否正确计算。
7. Properties of Variance and Standard Deviation | 方差与标准差的性质
b) = Var(X). Adding a constant shifts all values equally, so the spread remains the same.
方差具有几个在 IB 考试中频繁考查的代数性质。如果我们将随机变量 X 加上常数 b，方差不改变：Var(X + b) = Var(X)。加常数使所有值等量平移，因此离散程度保持不变。
nstant a, the variance is multiplied by a²: Var(aX) = a² Var(X), and consequently sd(aX) = |a| sd(X). Combining both, Var(aX + b) = a² Var(X).
当我们将 X 乘以常数 a 时，方差乘以 a²：Var(aX) = a² Var(X)，因此标准差 sd(aX) = |a| sd(X)。结合两者，Var(aX + b) = a² Var(X)。
= Var(X) + Var(Y) for independent X and Y. Standard deviations do not add in the same way.
这些性质在处理独立随机变量的线性组合时至关重要，此时方差相加：对于独立的 X 和 Y，Var(X ± Y) = Var(X) + Var(Y)。标准差不能以同样的方式相加。
8. Variance of a Discrete Random Variable | 离散随机变量的方差
the expected squared deviation from the mean: Var(X) = E[(X − μ)²] = ∑ (x − μ)² P(X = x).
对于具有概率分布 P(X = x) 的离散随机变量 X，方差定义为与均值之差的平方的期望：Var(X) = E[(X − μ)²] = ∑ (x − μ)² P(X = x)。
This avoids having to subtract the mean repeatedly. Both approaches yield the same result.
一个更便捷的计算公式是 Var(X) = E(X²) − [E(X)]²，其中 E(X²) = ∑ x² P(X = x)。这避免了反复减去均值。两种方法得出相同的结果。
table to include columns for x·P(X=x) and x²·P(X=x), sum them, and then apply the formula.
IB 考试题目经常提供一个概率表，并要求你计算期望值和方差。要有条理：扩展表格以包含 x·P(X=x) 和 x²·P(X=x) 的列，求和，然后应用公式。
9. Common Mistakes and How to Avoid Them | 常见错误及如何避免
One frequent error is using the wrong denominator: N for a sample or n−1 for a population. Always read the question carefully to decide whether the data represent a sample or the entire population.
一个常见错误是使用了错误的分母：对样本使用 N 或对总体使用 n−1。务必仔细读题，判断数据是代表样本还是整个总体。
Another mistake is confusing variance with standard deviation. If the question asks for variance, do not give the square root. Likewise, if it asks for standard deviation, remember to take the square root of the variance.
另一个错误是混淆方差与标准差。如果题目要求方差，不要给出平方根。同样，如果要求标准差，记得对方差取平方根。
Students sometimes apply the linear transformation rule incorrectly. Remember: adding a constant does not affect variance or standard deviation, but multiplying by a constant multiplies variance by the square of that constant. Check signs: standard deviation is always non-negative.
学生们有时会错误地应用线性变换规则。记住：加常数不影响方差或标准差，但乘以常数会使方差乘以该常数的平方。检查符号：标准差始终为非负数。
10. Checking Your Work and Exam Strategies | 检查作业与考试策略
Variance must be non-negative. If you obtain a negative variance, you have almost certainly made a calculation error. Standard deviation should be roughly between zero and the range of the data; if it exceeds the range, double-check your work.
方差必须是非负的。如果你得到一个负的方差，几乎可以肯定是计算错误。标准差应大致介于零和数据范围之间；如果它超过范围，请仔细检查你的作业。
Use your GDC to verify hand calculations in the exam. However, show enough working to demonstrate understanding, as some marks are awarded for method. A clear table of deviations and squared deviations is often sufficient.
在考试中使用你的 GDC 验证手工计算。然而，要展示足够的过程以证明理解，因为有些分数是给方法的。一个清晰的偏差和偏差平方表通常就足够了。
When interpreting results, always relate the standard deviation back to the context. Saying “the standard deviation is 12.3” is less meaningful than “the typical spread of the data around the mean is about 12 units”.
在解释结果时，始终将标准差与上下文联系起来。说“标准差为 12.3”不如“数据围绕均值的典型离散程度约为 12 个单位”有意义。
11. Variance in Bivariate and Further Topics | 二元及进阶主题中的方差
In IB, variance concepts extend to regression and correlation. The coefficient of determination r² represents the proportion of variance in the dependent variable explained by the independent variable. Understanding variance deepens the meaning of r².
在 IB 中，方差概念延伸到回归和相关。判定系数 r² 表示因变量中由自变量解释的方差比例。理解方差加深了对 r² 含义的认识。
tion B(n, p), variance = np(1−p). The normal distribution uses σ² as a parameter directly.
在二项分布和正态分布等概率分布中，方差公式是标准结果：对于二项分布 B(n, p)，方差 = np(1−p)。正态分布直接使用 σ² 作为参数。
Mastering variance now makes later topics much easier. You will see variance in hypothesis testing, confidence intervals, and the study of residuals. Treat it as a building block of statistical thinking.
现在掌握方差会让后面的主题容易得多。你会在假设检验、置信区间和残差研究中看到方差。把它当作统计思维的基石。
12. Summary and Key Takeaways | 总结与要点回顾
- Variance = average squared distance from the mean; standard deviation = its square root. 方差 = 到均值的平均平方距离；标准差 = 其平方根。
- Use σ² and σ for populations, s² and s for samples. 总体使用 σ² 和 σ，样本使用 s² 和 s。
- Bessel's correction (n−1) makes sample variance unbiased. 贝塞尔校正（n−1）使样本方差无偏。
- GDC offers quick computation; always distinguish sx from σx. GDC 提供快速计算；始终区分 sx 和 σx。
- Var(aX + b) = a² Var(X); adding constants does not change spread. Var(aX + b) = a² Var(X)；加常数不改变离散度。
- For discrete random variables, Var(X) = E(X²) − [E(X)]². 对于离散随机变量，Var(X) = E(X²) − [E(X)]²。
- Contextual interpretation makes your answers stronger and exam-ready. 情境化解释让你的答案更扎实，更符合考试要求。
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