Statistics for Everyday Life
Probability vs Statistics
Probability and statistics are deeply linked but point in opposite directions. One starts from a model and predicts data; the other starts from data and infers the model.
| Aspect | Probability | Statistics |
|---|---|---|
| Starting point | A known model or process | Observed data |
| Direction of reasoning | Model → likely outcomes | Data → the underlying model |
| Typical question | "Given a fair die, what’s the chance of a 6?" | "Given these rolls, is the die fair?" |
| Role | Predicts what data to expect | Infers what produced the data |
| Certainty | Deductive from assumptions | Inductive, with uncertainty |
When to use probability
You use probability when the rules are known and you want to predict outcomes — games of chance, risk models, forecasting from assumptions.
When to use statistics
You use statistics when you have data and want to learn what generated it — estimating, testing, and drawing conclusions under uncertainty.
Frequently asked questions
- What’s the simplest way to tell them apart?
- Probability goes from model to data ("if the coin is fair, how often will it land heads?"); statistics goes from data to model ("it landed heads 70 times in 100 — is it fair?"). They are inverse problems.
- Do you need probability to do statistics?
- Yes — probability provides the mathematical foundation that lets statistics quantify uncertainty, such as p-values and confidence intervals. Statistics applies probability to real data.
- Which is harder?
- Neither is inherently harder, but they demand different thinking. Probability is more deductive (working out consequences of assumptions); statistics is more inductive (drawing uncertain conclusions from messy data).