📚 Diversity (Measures of Dispersion) | 多样性（离散程度）
In statistics, the term 'diversity' is often used to describe the spread, variability, or dispersion within a data set. While measures of central tendency like the mean and median tell us about the typical value, measures of dispersion tell us how much the data values differ from one another – essentially, how diverse the set is. Understanding this diversity is crucial for drawing meaningful conclusions from data.
在统计学中，“多样性”一词常用来描述数据集的分布、变异或离散程度。虽然集中趋势的度量（如平均数和中位数）告诉我们典型数值是什么，但离散程度度量则告诉我们数据值之间相差有多大——本质上是数据集的多样性程度。理解这种多样性对于从数据中得出有意义的结论至关重要。
1. Introduction to Diversity in Data | 数据多样性简介
Diversity in a data set refers to how observations are scattered around a central point. A data set can have the same mean but very different levels of spread. For example, the test scores 50, 50, 50, 50 have a mean of 50 with zero diversity, while 0, 50, 100, 50 also have a mean of 50 but display much greater diversity. Measuring this diversity helps us compare the consistency, reliability, and risk associated with different sets.
数据集中的多样性是指观测值围绕中心点的分散程度。一个数据集可以具有相同的平均值，但分散程度却截然不同。例如，考试分数 50、50、50、50 的平均值为 50，多样性为零；而 0、50、100、50 的平均值也是 50，但多样性却大得多。测量这种多样性有助于我们比较不同数据集的一致性、可靠性和风险。
2. Why Measure Diversity? | 为什么测量多样性？
Measures of diversity give context to the average. In manufacturing, two machines may produce components with the same mean length, but one might have higher variability, resulting in more defective products. In finance, two investment portfolios might have the same average return, but the one with higher variability carries greater risk. Without a measure of spread, the mean alone can be misleading.
多样性度量赋予平均值以背景。在制造业中，两台机器可能生产出平均长度相同的零件，但一台可能变化更大，导致更多次品。在金融领域，两个投资组合可能有相同的平均回报，但变异更大的那个风险更高。如果没有分散程度的度量，仅凭平均值可能会产生误导。
3. Range | 极差
e between the largest and smallest values in a data set. For a set of numbers, Range = Max – highly sensitive to outliers and does not reflect the distribution of intermediate values.
极差是最简单的多样性度量。它是数据集中最大值与最小值之间的差值。对于一组数值，极差 = 最大值 – 最小值。虽然极差计算简便，但它对异常值高度敏感，并且不能反映中间数值的分布情况。
For example, given the data: 12, 15, 18, 22, 35, the range is 35 – 12 = 23. If an outlier 100 is added, the range jumps to 88, exaggerating the diversity.
例如，给定数据：12、15、18、22、35，极差为 35 – 12 = 23。如果加入一个异常值 100，极差飙升到 88，夸大了多样性。
4. Interquartile Range (IQR) | 四分位距
The interquartile range measures the spread of the middle 50% rs. It is the difference between the upper quartile (Q₃) and lower quartile (Q₁): IQR = Q₃ – st be ordered. Q₁ is the median of the lower half, and Q₃ is the median of the upper half.
四分位距衡量中间 50% 数据的分散程度，因而对异常值具有抵抗力。它是上四分位数 (Q₃) 与下四分位数 (Q₁) 之差：IQR = Q₃ – Q₁。要找出 Q₁ 和 Q₃，必须先将数据排序。Q₁ 是下半部分的中位数，Q₃ 是上半部分的中位数。
12. Lower half: 5, 7, 8 gives Q₁ = 7. Upper half: 14, 15, 20 gives Q₃ = 15. Hence IQR = 15 – 7 = 8. This tells us that the central half of the data spans 8 units.
如果数据集为 5、7、8、12、14、15、20，中位数为 12。下半部分：5、7、8 得出 Q₁ = 7。上半部分：14、15、20 得出 Q₃ = 15。因此 IQR = 15 – 7 = 8。这告诉我们中间一半的数据跨度为 8 个单位。
5. Variance and Standard Deviation | 方差与标准差
Variance and standard deviation are the most widely used measures of diversity because they take every data point into account. Variance is the average of the squared differences from the mean. Standard deviation is the square root of variance, bringing the measure back to the original units. For a population, the variance is denoted σ², and for a sample it is s², often using n–1 as the divisor to obtain an unbiased estimate.
方差和标准差是最常用的多样性度量，因为它们考虑了每一个数据点。方差是各数据与均值之差的平方的平均值。标准差是方差的平方根，使度量回归原始单位。对于总体，方差记为 σ²；对于样本，记为 s²，常用 n–1 作为除数以获得无偏估计。
Formula for population variance: σ² = Σ(xᵢ – μ)² / N. Sample variance: s² = Σ(xᵢ – x̄)² / (n – 1). Sample standard deviation: s = √[Σ(xᵢ – x̄)² / (n – 1)].
总体方差公式：σ² = Σ(xᵢ – μ)² / N。样本方差：s² = Σ(xᵢ – x̄)² / (n – 1)。样本标准差：s = √[Σ(xᵢ – x̄)² / (n – 1)]。
6. Calculating Variance for Ungrouped Data | 计算未分组数据的方差
To calculate sample variance manually: first find the sample mean x̄ ach data value, square the result, and sum all these squared deviations. Finally, divide by (n – 1) for a sample or by n for a population.
要手动计算样本方差：首先求出样本均值 x̄ = Σx / n。然后，用每个数据值减去均值，将结果平方，并求所有平方偏差之和。最后，样本方差除以 (n – 1)，总体方差除以 n。
Example: data set 4, 8, 6, 10. n = 4. x̄ = (4+8+6+10)/4 = 28/4 = 7. Deviations: (4–7) = –3 → 9; (8–7)=1 → 1; (6–7)= –1 → 1; (10–7)=3 → 9. Sum of squares = 20. Sample variance s² = 20 / (4–1) = 20/3 ≈ 6.67. Sample standard deviation s = √6.67 ≈ 2.58.
示例：数据集 4, 8, 6, 10。n = 4。x̄ = (4+8+6+10)/4 = 28/4 = 7。偏差：(4–7) = –3 → 9；(8–7)=1 → 1；(6–7)= –1 → 1；(10–7)=3 → 9。平方和 = 20。样本方差 s² = 20 / (4–1) = 20/3 ≈ 6.67。样本标准差 s = √6.67 ≈ 2.58。
7. Calculating Variance for Grouped Data | 计算分组数据的方差
ed in a frequency table, use the midpoints of each class as x-values. The sample mean is x̄ = Σ f x / Σ f. Then sample variance s² = Σ f (x – x̄)² / (Σ f – 1). For large data sets, the computational formula s² = (Σ f x² – (Σ f x)² / Σ f) / (Σ f – 1) may be more convenient.
当数据以频数表形式呈现时，使用每个组的中点作为 x 值。样本均值 x̄ = Σ f x / Σ f。然后样本方差 s² = Σ f (x – x̄)² / (Σ f – 1)。对于大型数据集，计算式 s² = (Σ f x² – (Σ f x)² / Σ f) / (Σ f – 1) 可能更方便。
Consider ages of students: 10–12 (f=5, midpoint 11), 13–15 (f=8, midpoint 14), 16–18 (f=7, midpoint 17). Σ f = 20. Σ f x = 5×11 + 8×14 + 7×17 = 55 + 112 + 119 = 286, x̄ e Σ f x² = 5×121 + 8×196 + 7×289 = 605 + 1568 + 2023 = 4196. Using the formula: s² = (4196 – 286²/20) / (20–1) = (4196 – 4089.8) / 19 = 106.2 / 19 ≈ 5.59.
考虑学生年龄：10–12 (f=5, 中点 11), 13–15 (f=8, 中点 14), 16–18 (f=7, 中点 17)。Σ f = 20, Σ f x = 5×11 + 8×14 + 7×17 = 55+112+119=286, x̄ = 14.3。然后计算 Σ f x² = 5×121+8×196+7×289=4196。使用公式：s² = (4196 – 286²/20) / 19 = (4196 – 4089.8) / 19 ≈ 5.59。
8. Properties of the Standard Deviation | 标准差的性质
Adding or subtracting a constant to all data values does not change the standard deviation, as the spread remains the same. However, multiplying or dividing all values by a constant scales the standard deviation by that same factor. For example, if each data point is doubled, the standard deviation also doubles.
对所有数据值加上或减去一个常数不会改变标准差，因为分散程度保持不变。但是，将所有值乘以或除以一个常数会使标准差按相同比例缩放。例如，如果每个数据点都扩大一倍，标准差也会扩大一倍。
Standard deviation is always non‑negative, and a value of zero indicates no diversity at all. It is measured in the same units as the original data, which makes it interpretable and useful for comparing different data sets.
标准差总是非负的，值为零表明完全没有多样性。它的单位与原始数据相同，因此具有可解释性，并且对于比较不同数据集很有用。
9. Comparing Measures of Dispersion | 比较离散度量
Each measure of diversity has its strengths and weaknesses. The range is quick but unstable. The IQR is robust against outliers but ignores tail information. Standard deviation uses all data and is the most informative, but it can be heavily influenced by extreme values. The choice depends on the data distribution and the purpose of analysis. When data contains outliers, the IQR is preferred; for symmetric distributions without outliers, standard deviation is ideal.
每种多样性度量各有优缺点。极差计算快速但不稳定。四分位距对异常值稳健，但忽略尾部信息。标准差使用全部数据，信息最丰富，但容易受极端值影响。选择何种度量取决于数据分布和分析目的。当数据包含异常值时，优先使用四分位距；对于没有异常值的对称分布，标准差最为理想。
10. Practical Applications of Dispersion | 离散程度的实际应用
In quality control, a small standard deviation in product dimensions indicates a stable manufacturing process. In meteorology, the variability of temperature data helps climate scientists assess extreme weather risks. In education, the spread of exam scores reveals differences in student achievement and test reliability. Finance analysts use standard deviation of returns (volatility) to quantify investment risk.
在质量控制中，产品尺寸的标准差小意味着生产过程稳定。在气象学中，温度数据的变异性帮助气候科学家评估极端天气风险。在教育领域，考试分数的分布揭示学生成绩差异和测试可靠性。金融分析师使用回报的标准差（波动性）来量化投资风险。
11. Common Pitfalls and Exam Tips | 常见错误与应考技巧
When computing variance, many students forget to square the deviations or mistakenly use n instead of n–1 for a sample. Always check whether data is a population or a sample. With grouped data, ensure midpoints are correctly identified and used in the formulas. Exam questions may ask for interpretation, not just calculation – be ready to explain what a standard deviation value means in context.
在计算方差时，许多学生忘记将偏差平方，或者对样本错误地使用 n 而非 n–1。务必检查数据是总体还是样本。对于分组数据，确保正确识别并使用组中点。考试题目可能要求解释，而不仅是计算——准备好说明在特定情境下标准差数值的含义。
Another common error is confusing units: variance is in squared units, while standard deviation is in the original units. Always present your final measure of diversity with appropriate units and rounded sensibly.
另一个常见错误是混淆单位：方差的单位是平方单位，而标准差与原始数据单位相同。务必以适当单位呈现最终的多样性度量，并合理舍入。
12. Summary of Key Formulas | 关键公式总结
The core formulas for diversity are essential for A‑Level Mathematics. Below is a concise reference:
以下是 A‑Level 数学中多样性的核心公式，供简明参考：
Range = Max – Min
IQR = Q₃ – Q₁
Sample Variance s² = Σ(x – x̄)²/(n – 1)
Population Variance σ² = Σ(x – μ)²/N
Standard Deviation s = √[Σ(x – x̄)²/(n – 1)]
Memorising these and understanding when each is applicable will help you tackle exam problems on diversity with confidence.
熟记这些公式并理解各自的适用条件，将帮助你自信地应对有关多样性的考试题目。
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