Introducing Statistics: What Can We Learn from Data?
| English | Chinese | Pinyin |
|---|---|---|
| data | 数据 | shù jù |
| Statistics | 统计学 | tǒng jì xué |
| uncertainty | 不确定性 | bù què dìng xìng |
| individuals | 个体 | gè tǐ |
| variables | 变量 | biàn liàng |
| statistical question | 统计问题 | tǒng jì wèn tí |
Learning from data that varies
- Every day the world throws numbers at us — test scores, prices, heights. Statistics 统计学 is the science of learning from them.
- The catch: real data 数据 vary. No two people are identical, so answers come with wiggle room.
- Statistics is about finding patterns in that variation — and being honest about the uncertainty 不确定性 that remains.
- This whole course is a toolkit for turning messy data into trustworthy conclusions.
Data that varies
Real data spread out across values — statistics finds the pattern in that variation.
Individuals and variables
- A data set records individuals 个体 (the cases — people, animals, objects) and their variables 变量 (the characteristics measured).
- Rows are usually individuals; columns are variables.
- Example: a class roster — each student is an individual; their height, grade, favorite subject are variables.
- Identifying who and what is the first step of any analysis.
In a data set of $200$ students' commute times, the individuals are...
Individuals are the cases (students); commute time is the variable.
In a typical data table, the columns usually represent...
Rows = individuals, columns = variables.
Asking a statistical question
- A statistical question 统计问题 anticipates variability — it expects the answers to differ.
- "How tall is Ana?" is not statistical (one answer). "How tall are students in this school?" is (many varying answers).
- A good statistical question can be answered with data and accounts for spread.
- Framing the right question shapes everything that follows.
Which is a statistical question?
It anticipates variability — many differing answers.
A question with a single definite answer is a statistical question.
Statistical questions anticipate variability.
Patterns vs. uncertainty
- Patterns in data let us generalize — to make claims beyond the exact cases we measured.
- But variability means those claims carry uncertainty: we're rarely $100\%$ sure.
- The art is drawing useful conclusions while honestly reporting how uncertain they are.
- Statistics never says "definitely"; it says "very likely, within this margin."
Because data vary, conclusions drawn from them carry ____.
Variability means we are rarely $100\%$ certain.
Select all true statements about statistics.
Statistics generalizes with uncertainty; it never claims certainty.
A statistical question anticipates variability — a question with a single definite answer ("What is the capital of France?") is not statistical. And beware overreach: a pattern in data supports a generalization with uncertainty, never a claim of certainty. Data can mislead if you forget the variability behind it.
A survey records $200$ students' commute times.
- Individuals: the $200$ students. Variable: commute time (a number that varies).
- Statistical question: "What is the typical commute time, and how much does it vary?"
- A conclusion like "most students commute $15$–$30$ min" generalizes, but with uncertainty from the sample.
Statistics is the science of learning from data that vary. A data set has individuals (cases) and variables (measurements). A statistical question anticipates variability and is answered with data. Patterns let us generalize, but variability always leaves uncertainty — report it honestly.