Statistics & Probability

Statistical Questions

Recognizing questions that anticipate variability in the data.

Intermediate15 min lesson3 min readUpdated August 12, 2026Author not yet attributed

What Makes a Question Statistical?

Consider two questions: 'How old is the school's principal?' and 'How old are the students at this school?' Before reading on, think about what's genuinely different about these two questions, beyond just who they're asking about.

Definition — Statistical Question

A question that anticipates variability in the data collected to answer it — one where different individuals in the group being studied are expected to have different values, not a single fixed answer.

'How old is the principal?' has exactly one correct answer — there's only one principal, and asking the question twice gets the same number both times. 'How old are the students?' is completely different: every student might have a different age, and answering the question means describing a whole collection of different values, not a single one.

Worked Example — Classifying Questions as Statistical or Not

Which of these is a statistical question: 'What is the height of the Eiffel Tower?' or 'What are the heights of the trees in this park?' The Eiffel Tower's height is one fixed value — not statistical. The heights of many different trees will vary from tree to tree — that's a statistical question.

Worked Example — Turning a Non-Statistical Question Statistical

'How many pets does Maya have?' isn't statistical — it has one specific answer. Rewritten as 'how many pets do the students in this class have?' it becomes statistical, since different students are expected to give different answers.

Tip

Ask yourself: if this question were asked to many different people or things instead of just one, would you expect the exact same answer every time, or different answers? Different answers means it's a statistical question.

Common Mistakes

Assuming any question that involves a number is automatically a statistical question.
A question is statistical only if it's expected to produce different values across the group being asked — a single, fixed numerical fact isn't statistical just because it involves a number.
Confusing a question about one specific individual with a question about a whole group.
Check whether the question is really asking about one fixed thing or about a collection of things that could differ from each other — the group, and the expected variation within it, is what makes a question statistical.

Key Takeaways

A statistical question anticipates variability — different individuals in a group are expected to give different answers.
A question with one fixed, single-valued answer is not a statistical question.
Rephrasing a question to ask about a varying group, rather than one individual, can turn it into a statistical question.

Summary

Recognizing a statistical question is the first step toward analyzing real data. The final lesson looks at how to summarize the variability such a question produces, using measures of center and spread.

Track your progress

Back to Statistics & Probability Measures of Center & Spread→