Descriptive Statistics: Summarizing and Visualizing Data | 描述性统计：数据的汇总与可视化

📚 Descriptive Statistics: Summarizing and Visualizing Data | 描述性统计：数据的汇总与可视化

Descriptive statistics is the branch of statistics that focuses on summarizing, organizing, and presenting data in a meaningful way. It provides simple summaries about the sample and the measures, using tables, graphs, and numerical calculations. Instead of making inferences or predictions, descriptive statistics simply describe what the data shows, helping to detect patterns, identify outliers, and understand the distribution’s shape. This foundation is crucial for further statistical analysis, including inferential statistics.

描述性统计是统计学的分支，其重点在于以有意义的方式对数据进行汇总、整理和呈现。它利用表格、图形和数值计算提供关于样本和指标的简要总结。与推断或预测不同，描述性统计仅仅描述数据所显示的信息，帮助发现规律、识别异常值并理解分布形态。这一基础对于进一步的统计分析（包括推断统计）至关重要。

1. Introduction to Descriptive Statistics | 描述性统计简介

Descriptive statistics involves methods for collecting, summarizing, and displaying data. It aims to condense large amounts of information into understandable formats, such as charts and summary numbers. The two main types are measures of central tendency (where the data cluster) and measures of dispersion (how spread out the data are). These, together with graphical representations, give a complete picture of the dataset.

描述性统计涉及数据的收集、汇总和展示方法，旨在将大量信息凝练成易于理解的格式，如图表和汇总数字。它主要包括集中趋势指标（数据聚集的位置）和离散程度指标（数据的分散程度）两大类。这些与图形表示相结合，便能完整地描绘数据集的状况。

2. Data Types and Measurement Scales | 数据类型与测量尺度

Data can be classified as categorical (qualitative) or numerical (quantitative). Categorical data represent groups, such as eye colour or brand preference. Numerical data are further split into discrete (countable, like number of students) and continuous (measurable, like height). Recognizing the data type is essential because it determines which descriptive methods and graphs are appropriate.

数据可分为分类（定性）数据或数值（定量）数据。分类数据代表组别，如眼睛颜色或品牌偏好。数值数据进一步分为离散型（可数的，如学生人数）和连续型（可测量的，如身高）。识别数据类型至关重要，因为它决定了哪些描述方法和图形是合适的。

3. Organizing Data: Frequency Distributions | 数据整理：频数分布

A frequency distribution table groups data into classes and records how many observations fall into each class. For discrete data with few values, we can list each value. For continuous data, we create intervals (e.g., 10–20, 20–30). The table may also include relative frequency (proportion) and cumulative frequency. This organized view facilitates the calculation of descriptive measures and the drawing of graphs.

频数分布表将数据分组，并记录每个组内观测值的个数。对于取值较少的离散数据，可以列举每个值；对于连续数据，我们创建区间（如10–20、20–30）。表格还可以包含相对频数（比例）和累积频数。这种有序的视图便于计算描述指标和绘制图形。

Example of a simple frequency table:

简单频数表示例：

Score Interval	Frequency
0–10	5
10–20	12
20–30	8
30–40	3

4. Measures of Central Tendency | 集中趋势指标

The mean (x̄) is the arithmetic average, computed by summing all values and dividing by the number of observations: x̄ = ( Σx ) / n. It is sensitive to extreme values. The median is the middle value when data are ordered; it is resistant to outliers. The mode is the most frequently occurring value in a dataset. For symmetric distributions, the mean and median are close; for skewed data, they differ.

均值（x̄）是算术平均值，通过将所有数值相加再除以观测值个数得出：x̄ = ( Σx ) / n。它对极端值敏感。中位数是数据排序后位于中间的值，能抵抗异常值的影响。众数是数据集中出现频率最高的值。在对称分布中，均值与中位数接近；在偏斜数据中，两者存在差异。

Mean (grouped data): x̄ = Σ(f × m) / Σf, where m is the class midpoint.

分组数据均值：x̄ = Σ(f × m) / Σf，其中 m 为组中值。

5. Measures of Dispersion | 离散程度指标

Range = maximum – minimum is the simplest measure of spread, but it ignores the distribution’s interior. The interquartile range (IQR) = Q₃ – Q₁ covers the middle 50% and is robust against outliers. Variance measures the average squared deviation from the mean; for a sample, s² = Σ(x – x̄)² / (n – 1). Standard deviation (s) is the square root of variance, giving spread in the original units.

全距 = 最大值 – 最小值，是最简单的离散度指标，但忽略了数据内部分布。四分位距（IQR）= Q₃ – Q₁，涵盖中间50%的数据，且对异常值具有稳健性。方差衡量观测值偏离均值的平均平方距离；对于样本，s² = Σ(x – x̄)² / (n – 1)。标准差（s）是方差的平方根，以原始单位表示离散程度。

s² = Σ(x – x̄)² / (n – 1) and s = √[ Σ(x – x̄)² / (n – 1) ]

s² = Σ(x – x̄)² / (n – 1) 及 s = √[ Σ(x – x̄)² / (n – 1) ]

6. Percentiles and Quartiles | 百分位数与四分位数

The k-th percentile is a value below which k% of the observations fall. The 25th percentile is Q₁, the 50th is Q₂ (median), and the 75th is Q₃. To find a percentile, first order the data, then calculate the position L = (k/100) × n. If L is not an integer, round up to the next whole number. For grouped data, linear interpolation is used to estimate percentiles between class boundaries.

第 k 百分位数是一个值，低于该值的观测值占 k%。第25百分位数即 Q₁，第50百分位数即 Q₂（中位数），第75百分位数即 Q₃。求百分位数时，先将数据排序，然后计算位置 L = (k/100) × n。若 L 不是整数，则向上取整。对于分组数据，使用线性插值法在组边界间估计百分位数。

L = (k/100) × n → if L is not integer, take the ceiling value.

L = (k/100) × n → 若 L 非整数，则向上取整。

7. The Five-Number Summary and Boxplots | 五数概括与箱线图

The five-number summary consists of the minimum, Q₁, median (Q₂), Q₃, and maximum. A boxplot (box-and-whisker plot) graphically displays this summary, with a box from Q₁ to Q₃ and a line at the median. Whiskers extend to the smallest and largest values within 1.5 × IQR from the quartiles; points beyond are considered outliers and shown as individual dots. Boxplots quickly reveal symmetry, skewness, and outliers.

五数概括由最小值、Q₁、中位数（Q₂）、Q₃和最大值组成。箱线图（盒须图）以图形方式展示这一概括：箱子从 Q₁ 到 Q₃，中间一条线表示中位数；须线延伸至四分位距1.5倍范围内的最值点；超出此范围的点被视为异常值，以圆点表示。箱线图能迅速揭示对称性、偏度和异常值。

8. Graphical Displays for One Variable: Histograms and Cumulative Frequency Curves | 单变量图形展示：直方图与累积频率曲线

A histogram is a bar graph for continuous grouped data, where the area of each bar represents frequency (or density). Adjacent bars touch to reflect continuous scale. The shape of a histogram indicates modality and skewness. A cumulative frequency curve (ogive) plots cumulative frequency against upper class boundaries, useful for estimating medians and percentiles visually.

直方图是用于连续分组数据的条形图，其中每个条形的面积代表频数（或密度）。相邻条形相互接触，以体现数据的连续性。直方图的形态可显示众数和偏态。累积频率曲线（折线图）将累积频率与组上界进行描点，便于直观估计中位数和百分位数。

9. Stem-and-Leaf Plots | 茎叶图

A stem-and-leaf plot splits each data value into a ‘stem’ (all but the final digit) and a ‘leaf’ (the final digit). This plot preserves the original data while showing the shape of the distribution. Back-to-back stem-and-leaf plots allow comparison of two datasets by sharing a common stem. They are particularly useful for small to moderate datasets and for identifying modes and gaps.

茎叶图将每个数据值分为“茎”（除最后一位数字外的所有数字）和“叶”（最后一位数字）。这种图既能保留原始数据，又能展示分布形态。背靠背茎叶图通过共用茎部可以比较两个数据集。它尤其适用于小到中等规模的数据集，并能帮助识别众数和间隙。

10. Bivariate Data: Scatterplots and Correlation | 双变量数据：散点图与相关性

Descriptive statistics also examines the relationship between two numerical variables. A scatterplot is a graph of ordered pairs (x, y) that reveals patterns, direction, and strength of association. Pearson’s correlation coefficient r measures the linear strength and direction (from –1 to +1). A positive r indicates that as x increases, y tends to increase; negative r indicates the opposite. Note that correlation does not imply causation.

描述性统计也考察两个数值变量间的关系。散点图是由有序对（x, y）构成的图形，能揭示变量间关联的模式、方向和强度。皮尔逊相关系数 r 衡量线性相关程度与方向（取值范围为 –1 至 +1）。r 为正表示 x 增大时 y 也倾向于增大；r 为负则反之。请注意，相关关系并不意味因果关系。

r = Σ[(x – x̄)(y – ȳ)] / √[ Σ(x – x̄)² Σ(y – ȳ)² ]

11. Interpreting Shapes: Skewness and Symmetry | 形态解释：偏度与对称

Symmetrical distributions have the mean and median approximately equal. In a positively skewed (right-skewed) distribution, the mean > median, and the tail extends to the right. In a negatively skewed (left-skewed) distribution, the mean < median, and the tail extends to the left. Skewness affects the interpretation of central tendency and the choice of test statistics in later analysis.

对称分布的均值与中位数大致相等。在正偏（右偏）分布中，均值 > 中位数，且尾部向右延伸。在负偏（左偏）分布中，均值 < 中位数，尾部向左延伸。偏度会影响集中趋势的解释以及后续分析中检验统计量的选择。

Additional measures such as the sample skewness coefficient or kurtosis describe the shape more precisely, but visual inspection through histograms and boxplots often suffices for descriptive purposes.

样本偏度系数或峰度等附加指标能更精确地描述形态，但通过直方图和箱线图进行目视检查通常已能满足描述需求。

12. Summary and Tips for IB Exams | 总结与IB考试技巧

In the IB curriculum, descriptive statistics tasks require clear presentation of data, accurate calculation of summary measures, and correct interpretation of graphs. Always label axes, show units, and explain what a statistic reveals in context. When calculating from grouped data, use midpoints consistently and state any assumptions. Practice switching between different representations, as questions often link a frequency table with a boxplot or histogram. Remember to distinguish between sample and population formulas, and to check for outliers using the 1.5 IQR rule.

在IB课程中，描述性统计任务要求清晰地呈现数据、准确计算汇总指标并正确解读图形。务必为坐标轴添加标签、标明单位，并解释统计量在背景中揭示了什么。在根据分组数据计算时，须统一使用组中值，并说明所作假设。练习在不同表示形式之间切换，因为题目常常将频数表与箱线图或直方图联系起来。注意区分样本与总体公式，并利用1.5倍IQR规则检查异常值。

Finally, when describing distributions, comment on shape (symmetric/skewed), centre (median or mean), spread (IQR or standard deviation), and any unusual features. This structured approach earns full marks on descriptive questions.

最后，在描述数据分布时，需从以下方面进行评述：形态（对称/偏斜）、中心（中位数或均值）、离散度（IQR或标准差）以及任何异常特征。这种结构化的答题方式能帮助在描述性题目中获得满分。

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Descriptive Statistics: Summarizing and Visualizing Data | 描述性统计：数据的汇总与可视化