Lecture 4 — Descriptive Statistics (描述性统计)
1. Introduction (简介)
Business Statistics (商业统计学)
- Application of statistical tools in business and economics.
- 在商业与经济中应用统计工具。
Course Info (课程信息)
- Prof. Rongjuan Chen, Spring 2025, Wenzhou-Kean University.
- 陈荣娟教授，2025春季，温州肯恩大学。
2. Descriptive Statistics Overview (描述性统计概览)
Definition (定义)
- Summarizing and presenting data in tables, graphs, or numbers.
- 以表格、图形或数值总结与展示数据。
Data Types (数据类型)
- Categorical data: labels or names (分类数据)。
- Quantitative data: numerical values (定量数据)。
3. Frequency Distribution (频数分布)
Definition (定义)
- Tabular summary showing frequency of each class.
- 展示各类出现次数的表格汇总。
Purpose (目的)
- Reveal patterns not easily observed.
- 揭示数据中不易直接发现的模式。
4. Example: Maine’s Inn — Raw Data (缅因旅馆示例——原始数据)
Guest Ratings (顾客评分)
- Excellent, Above Average, Average, Below Average, Poor.
- 优秀、较好、一般、较差、差。
Purpose (目的)
- Demonstrate how to build a frequency distribution.
- 演示如何构建频数分布。
5. Pivot Table (数据透视表)
Definition (定义)
- Tool for quick summarization of categorical data.
- 用于快速汇总分类数据的工具。
Example (例子)
- Poor=2, Below Average=3, Average=5, Above Average=9, Excellent=1.
- 差=2，较差=3，一般=5，较好=9，优秀=1。
6. Relative & Percent Frequency (相对与百分比频率)
Relative Frequency (相对频率)
- Frequency ÷ Total.
- 频数 ÷ 总数。
Percent Frequency (百分比频率)
- Relative frequency × 100%.
- 相对频率 × 100%。
7. Bar Chart (条形图)
Definition (定义)
- Graphical representation of categorical data.
- 用柱形表示分类数据的图表。
Example (例子)
- “Above Average” has the tallest bar (9 counts).
- “较好”栏最高 (9 次)。
8. Cumulative Frequency (累积频率)
Definition (定义)
- Adding category frequencies sequentially.
- 依次累加各类别频数。
Example (例子)
- Average = 70% cumulative; Below Average = 85%.
- 一般=累计70%，较差=累计85%。
9. Pareto Chart (帕累托图)
Definition (定义)
- Combination of bar chart + line chart (cumulative %).
- 条形图 + 折线图 (累计百分比) 的结合。
Principle (原理)
- Pareto rule (80/20 rule).
- 帕累托法则 (80/20 法则)。
10. Example: Red Lobster Complaint Data (红龙虾餐厅投诉数据)
Complaint Categories (投诉类别)
- Overpriced, Small portions, Wait time, etc.
- 价格高、分量少、等待时间长等。
Cumulative % (累计百分比)
- Overpriced=45.8%, Overpriced+Small portions=81.9%.
- 价格过高=45.8%，价格高+分量少=81.9%。
11. Example: Red Lobster Pareto Chart (红龙虾帕累托图)
Visualization (可视化)
- Bars show frequencies, line shows cumulative %.
- 条形显示频数，折线显示累计百分比。
Insights (洞察)
- Few problems account for most complaints.
- 少数问题占大多数投诉。
12. Pie Chart (饼图)
Definition (定义)
- Circle divided into slices representing proportions.
- 圆形分割表示比例。
Data Basis (数据基础)
- Based on frequency, relative frequency, or percent frequency.
- 基于频数、相对频率或百分比。
Example (例子)
- Above Average = 45% slice.
- “较好”占饼图 45%。