Not sure which lesson to do first? Find your level and get a plan →

Statistics · Grade 3-5

Statistical Question

A statistical question is a question that anticipates variability in answers — it cannot be answered with a single fixed number because different data points will give different responses.


Quick Answer

A statistical question is a question that anticipates variability in answers — it cannot be answered with a single fixed number because different data points will give different responses. In plain words, you expect the answers to differ. 'How tall is our teacher?' has one answer, so it is not statistical. 'How tall are the students in our class?' has many different answers, so it is. Spot this before a survey: a statistical question means collecting many answers and looking at how they spread out.

Why This Matters

Knowing whether a question is statistical tells you what work to do. 'How many pages is this book?' needs one look at the last page. 'How many pages do fourth graders read in a week?' needs a survey, a table and a graph, and the answer is a typical amount plus a range, like 'most read 40 to 80 pages'. Mix them up and you either collect data you don't need or give one number where the answers really vary.

Intuitive Explanation

Ask one friend 'How old are you?' and you get one answer: 9. Ask 'How old are the kids at our swimming club?' and you get many answers: 7, 8, 8, 9, 10, 11, 11, 12. The second question is a statistical question, because you expect the answers to differ from person to person. That difference in the answers is called variability.

A good test is to imagine asking the question to several people, or measuring several times. If everyone would give the same answer, it is not statistical: 'How many legs does a spider have?' is always 8. If the answers would change, it is: 'How many legs do the creatures in our garden have?' gets 6 for an ant and 8 for a spider.

A question with a number in it is not automatically statistical. 'How many students are in our class?' has a number answer, 25, but only one of it. Watch for questions about one person or one thing, like 'How long is my pencil?'. Turn it into a question about a group, 'How long are the pencils in our class?', and it becomes statistical.

Think of it like this. A statistical question is like asking 'What will lunch be?' every day for a month: the answer keeps changing, so you need many answers to understand it. A non-statistical question is like asking 'How many days are in a week?': you get 7 every single time, so one look is enough.

Core idea. A question is statistical when you expect the answers to vary across a group, so it can only be answered by collecting many answers and describing how they spread.


When to Use

Check whether a question is statistical before you plan a survey. It is when the question is about a group (students in my grade, dogs at the park) and you expect different answers, for example how many, how long or how often for each one. If the question is about one person or thing, or has one fixed answer you can look up, it is not statistical; just find that fact.

Tip. Ask: Would asking this question to many different people or things give different answers, rather than one fixed fact?

How to Recognize It

A statistical question is judged by the answers it would get, so imagine asking it before you decide.

  1. Is the question about a group, such as a class, a team, or a box of apples, rather than one person or thing?

    A group is needed. 'How tall is Sam?' is about one person and has one answer, so it is not statistical.

  2. Would you need to gather many answers, by asking, watching, or measuring, to answer it?

    Yes fits a statistical question. If the problem is about how to do the gathering, it has moved on to Data Collection.

  3. Is the problem asking you to write or pick a good question, rather than describe how spread out some answers are?

    Writing or judging the question is this concept. Describing how far apart answers you already have are is Data Variability.

  4. Could someone answer it by looking up one number, like a date on a calendar or a price tag?

    If yes, it is not statistical. 'How many days are in May?' has one answer, 31.

Statistical Question vs Data Collection vs Data Variability vs Population vs Sample

These ideas all appear at the start of a statistics project. Ask whether you are judging a question, gathering answers, describing how answers differ, or naming who the answers come from.

Statistical Question

Meaning
A statistical question is a question that you expect to get different answers to across a group.
Key test
You are judging the question itself, before any data exist.
Example
'How many hours do kids in my grade sleep?' expects answers like 8, 9, and 10.

Data Collection

Meaning
Data collection is gathering the answers, by survey, observation, or experiment, once the question is set.
Key test
You are planning or doing the asking, watching, or measuring.
Example
You ask 30 kids in your grade how many hours they slept last night.

Data Variability

Meaning
Data variability describes how much the answers in a data set differ from each other.
Key test
You already have answers and describe how spread out they are.
Example
Sleep answers of 9, 9, 9 show no variability; answers of 6, 9, 11 show a lot.

Population vs Sample

Meaning
The population is the whole group a question is about, and a sample is the part of it you actually ask.
Key test
You are naming who the answers come from: everyone, or only some.
Example
Population: all 120 kids in grade 5. Sample: the 30 of them you asked.

Formula & Notation

Worked Examples

Example 1 — Sort four questions

Easy

Which of these are statistical questions? (a) How many hours of sleep do students in my class get on a school night? (b) What time does our school day start? (c) How many letters are in my first name? (d) How many letters are in the first names of the students in my class?
Show the steps and the answer
  1. Question (a): imagine asking several classmates. You might hear 8, 9 and 10 hours.

    The answers differ from student to student, so (a) is statistical.

  2. Question (b): every student gives the same answer, for example 8:15.

    One fixed fact has no variability, so (b) is not statistical.

  3. Question (c): one name gives one count, such as 4 for 'Maya'.

    A question about one person has a single answer.

  4. Question (d): 'Maya' has 4 letters, 'Jonathan' has 8 and 'Li' has 2.

    Across the class the counts vary, so (d) is statistical.

Answer: (a) and (d) are statistical questions; (b) and (c) are not.

Takeaway: Imagine several people answering: if the answers would differ, the question is statistical.

Example 2 — About a group, but no variety

Standard

Which of these is a statistical question? (a) How many legs does each chair in our classroom have? (b) How many minutes does each student in our class take to walk to school?
Show the steps and the answer
  1. Notice that both questions are about a group and both have number answers.

    Being about a group is not enough; the answers also have to vary.

  2. Question (a): every classroom chair has 4 legs.

    If every answer is 4, there is nothing to collect or compare.

  3. Question (b): one student walks 5 minutes, another 12, another 20.

    The times differ because students live at different distances, so the answers vary.

Answer: (b) is a statistical question; (a) is not, because every chair gives the same answer, 4.

Takeaway: A statistical question needs answers that vary, not just a group to ask.

Example 3 — Find the mistake: a question about one person

Application

Sam's class is planning a reading challenge. Sam writes the survey question 'How many books did I read last month?' and says, 'It's a statistical question because the answer is a number.' Find the mistake and correct it.
Show the steps and the answer
  1. Ask how many answers the question can have: Sam read one number of books, say 3.

    A question about one person has exactly one answer, so nothing varies.

  2. Notice that having a number answer doesn't make a question statistical.

    What matters is whether different answers are expected, not whether the answer is a number.

  3. Rewrite it about the group: 'How many books did each student in our class read last month?'

    Now answers such as 0, 2, 2, 3 and 5 are expected, so the class can collect and compare them.

Answer: Sam's question is not statistical. A statistical version is 'How many books did each student in our class read last month?'

Takeaway: Turn a question about one person into a question about a group, and check that the answers would differ.

Where You'll See It

  • Teacher's age vs. teachers' ages

    'How old is my teacher?' has one fixed answer (not statistical), but 'How old are the teachers at my school?' gives different answers for different teachers — that variability makes it a statistical question.

  • My height vs. class heights

    'How tall am I?' is a single fact, while 'How tall are the students in my class?' expects a range of answers, so the second is a statistical question worth collecting data on.

  • Steps in a day

    'How many steps did I take today?' is one number, but 'How many steps do kids in my grade take in a day?' anticipates variation across kids — a statistical question.

Common Mistakes

  1. The slip: Any question with a number answer is a statistical question.

    Why it's tempting: Statistics involves numbers, so number-shaped questions feel statistical. But a question with one fixed numeric answer has no variability.

    Instead: A question is statistical only if it anticipates VARIABILITY — different data points give different answers. 'How tall is the Eiffel Tower?' is not statistical; 'How tall are students in my grade?' is.

  2. The slip: A question about a single person can be statistical if it's about a person.

    Why it's tempting: It mentions people and data, so it sounds like statistics. But asking about one individual produces one answer, not a spread.

    Instead: Statistical questions are about a GROUP and expect varied responses. Asking about one specific individual ('How old is Mr. Lee?') is not statistical.

  3. The slip: If you can collect data to answer it, it's automatically statistical.

    Why it's tempting: Data collection is part of statistics, so any data-able question seems to qualify. But you can collect a single fact too.

    Instead: Test for expected variability in the answers, not just whether data exists. The defining feature is that responses are expected to differ across the group.


Mini Practice

Try these on your own. Tap Reveal when you want to check.

  1. Which of these is a statistical question? (a) How many days are in March? (b) How many hours did the students in our class sleep last night? (c) How tall is Ms. Lee?

    Hint: Look for the question that would get different answers from different people.

  2. Change "How many pets does Jordan have?" into a statistical question about Jordan's class of 25 students.

    Hint: Ask about the whole group instead of one person.

  3. Ten students answered "How many minutes did you read last night?" Their answers were 20, 0, 15, 30, 20, 45, 10, 20, 60, 5. What were the smallest and largest answers, and do the answers show the question is statistical?

    Hint: Find the lowest and highest numbers, then ask whether everyone gave the same answer.

  4. Is "How many students are in Room 12 right now?" a statistical question? What about "How many students are in each classroom at our school?" if the school has 18 classrooms?

    Hint: Count how many answers each question would collect.

  5. Ella has already written "What is the favourite season of students in Grade 4?" Now she is handing a slip of paper to each of the 60 Grade 4 students. Is she still writing a statistical question, or has she moved on to data collection?

    Hint: Is she deciding what to ask, or gathering the answers?

  6. Ravi says, "How many days are in a week? is a statistical question, because the answer is a number and I can collect it from 30 people." Correct him.

    Hint: Would the 30 people give different answers?

Want the full set?

50 practice questions for this concept — free to try, every one with a complete worked solution showing the why, not just the answer.

Frequently Asked Questions

Is 'How many students are in our class?' statistical if I count them myself?

No. Counting takes some work, but the count ends with one number, such as 25. Nothing varies from student to student, because each student is simply in the class. Change it to something each student would answer differently, like 'How many books did each student read this month?', and it becomes statistical, because you expect answers like 0, 2, and 5.

Can a question with word answers, not number answers, be statistical?

Yes. 'What is the favorite sport of students in my class?' is statistical, because different students will name different sports: soccer, swimming, basketball. The answers are words, so you count how many chose each sport instead of finding an average. What matters is that the answers vary across the group, not whether they are numbers.

Can a question about just me ever be statistical?

It can, if it is about many times instead of one time. 'How long did I sleep last night?' has one answer. 'How long do I usually sleep on school nights?' is about many nights, and the answers will differ: 8 hours one night, 9 the next. So the group can be many days in your own life instead of many people.

Why does it matter whether a question is statistical?

It tells you what kind of work comes next. A question with one fixed answer just needs you to find that fact. A statistical question needs a plan: who to ask, how to record the answers, and a way to sum them up, such as the most common answer or a typical value. Spotting the difference early stops you from running a whole survey for something you could look up.

Learning Path

← Before

Data Collection
Statistical Question

You are here

This builds on Data Collection: you already know how to ask and record. A statistical question tells you when collecting is needed, namely when the answers will vary. Next you use it to plan better Data Collection, and Data Variability shows you how to describe how much the answers spread out, starting with the range.

See Also