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

Math · Statistics & Probability · Grade 6-8

Representativeness

A sample is representative if its makeup mirrors the population — every relevant group appears in roughly its true proportion.


Quick Answer

A sample is representative if its makeup mirrors the population — every relevant group appears in roughly its true proportion. Use the idea to decide whether a survey's findings can stand for the whole group: list the groups that might answer differently, such as ages, grades or regions, and compare each group's share of the sample with its share of the population. Sample size is a separate question. A huge sample with the wrong mix still misleads, while a modest one with the right mix can give a good estimate.

Why This Matters

A survey describes only the people in it unless the sample is representative. If a school is 60% younger students but a lunch survey is 90% older students, its results describe older students, not the school. Checking proportions group by group is how pollsters, scientists and student councils decide whether a result can be applied to everyone.

Intuitive Explanation

A box of 100 crayons has 50 blue, 30 red and 20 yellow. A handful of 10 that represents the box should have about 5 blue, 3 red and 2 yellow. If your handful is 9 blue and 1 red, it still came from the box, but it describes the box badly: someone looking only at your handful would think yellow crayons do not exist.

For people, the groups that matter are the ones that might answer the question differently. For a survey about moving the school start time, grade level matters, since older students may prefer to start later, and so does whether students ride the bus. A representative sample has each of those groups in roughly the population's proportions. If a school is 25% sixth graders, a representative sample of 80 students includes about 0.25×80=20 sixth graders.

Representative is not the same as large. A survey with 5,000 responses, 90% from one grade, represents that grade, not the school. It is not the same as random, either. Random selection is the method that usually produces a representative sample, but any single random sample can still come out lopsided, so you check the result's proportions against what you know about the population.

Think of it like this. A representative sample is like a museum's scale model of a city. Parks, roads and buildings take up the same share of space as in the real city, so you can learn the layout from it. A model missing a whole neighbourhood misleads however carefully the rest was built. Unlike a model, a sample only has to match on the features that matter for the question.

Core idea. A sample is representative when its key characteristics match the population's in the right proportions.


When to Use

Use representativeness when you must judge whether a sample can stand for a population, or when you plan a sample and need how many from each group. Look for the population's proportions, such as 60% of one group and 40% of another, and compare the sample's shares with them on every variable that could affect the answer. If the question is about the method that chose people, look at sampling bias; if it is only about how many were asked, that is sample size.

Tip. Ask: Do the sample's group proportions match the population's?

How to Recognize It

Before you let a sample speak for a whole group, list the traits that could change the answer to the survey question, then look up the population's share of each.

  1. Could different groups in the population answer differently, such as younger and older students asked about bedtimes?

    If yes, those groups must appear in about their true shares, or the result leans toward the overcounted group. If a trait has nothing to do with the question, like eye color for bedtimes, the sample does not need to match on it.

  2. Is the issue just how many people were asked, with nothing said about who they were?

    Then it is a question of sample size, which controls how much results wobble by chance. A large sample with the wrong mix stays wrong; size and makeup are separate checks.

  3. Is the question about how the sample was chosen, such as from volunteers or by a random draw?

    Then it is about sampling bias and sampling methods. Representativeness judges the result, the sample's makeup, so even a randomly chosen sample deserves a look at its shares.

  4. When you plan a survey, do you know each group's share of the population?

    If yes, you can build a matching sample: a school that is 30% grade 6, 30% grade 7 and 40% grade 8 needs 30, 30 and 40 students in a sample of 100. This is called stratified sampling.

Representativeness compared with Sample size, Sampling bias, and Random sampling

Representativeness, sample size, sampling bias and random sampling are all checks on the same survey, so a big or randomly drawn sample is easily assumed to be a good copy. Ask whether you are judging the sample's makeup, its size, or the method that produced it.

Representativeness

Meaning
Use this when you must judge whether a sample's makeup mirrors the population well enough to generalize.
Key test
Do the sample's group proportions match the population's?
Example
A school is 60% younger and 40% older students. A representative sample of 50 should have about how many of each?

Sample size

Meaning
Is how many were chosen, not whether they reflect the population's mix.
Key test
Use when discussing precision, not representativeness.
Formula
n
Example
Surveying 2,000 people

Sampling bias

Meaning
Is the failure of representativeness from a skewed selection method.
Key test
Use when explaining why a sample doesn't mirror the population.
Example
Only surveying the basketball team

Random sampling

Meaning
Is a method that promotes representativeness, not the property itself.
Key test
Use when describing how the sample was drawn.
Example
Drawing names from a hat

Worked Examples

Example 1 — A school survey

Easy

A school is 60% younger and 40% older students. A representative sample of 50 should have about how many of each?
Show the steps and the answer
  1. Write the population's proportions: 60% younger, 40% older.

    A representative sample copies these shares, so they are the targets.

  2. Apply them to the sample size: 0.60×50=30 younger and 0.40×50=20 older.

    Each group's share of the sample should equal its share of the school.

  3. Check the total: 30+20=50.

    The two groups make up the whole sample, so the counts must add to 50.

Answer: About 30 younger and 20 older

Takeaway: Multiply each group's population share by the sample size to get the counts a representative sample needs.

Example 2 — Big but not representative

Standard

A survey collects 5,000 responses, but 90% are from one grade in a school evenly split across grades. Representative?
Show the steps and the answer
  1. Count the dominant grade: 0.90×5000=4500 responses from one grade.

    Turning the percentage into a count shows how heavily one grade outweighs the rest.

  2. Find the target share: if the school has 4 grades, each is 14=25% of students, about 1,250 of 5,000 responses.

    The problem does not say how many grades there are; with more than 4, each fair share is even smaller.

  3. Compare: 90% in the sample against about 25% in the school.

    The sample's mix is far from the population's, so its results mostly describe one grade.

  4. Note why size does not rescue it: all the other grades together supply only 5000−4500=500 responses.

    A large total cannot make up for a mix that does not match the school.

Answer: No — it's large but not representative

Takeaway: Judge a sample by whether its proportions match the population, not by how many responses it has.

Example 3 — Find the mistake: checking only age

Application

A town's adults are 50% under 40 and 50% aged 40 or over; 30% live in the rural north and 70% in the city. A survey of 200 adults has 100 under 40 and 100 aged 40 or over, but 190 of them live in the city. Ana writes: "The ages match the town exactly, so the sample is representative." Find the mistake and correct it.
Show the steps and the answer
  1. Check Ana's variable: 100200=50% under 40, matching the town's 50%.

    Ana is right about age; the question is whether age is the only variable that matters.

  2. Check location: 190200=95% city and 10200=5% rural, against the town's 70% and 30%.

    A sample can match on one variable and miss badly on another.

  3. Find the target counts: a representative 200 has 0.30×200=60 rural and 0.70×200=140 city adults.

    The sample is 50 rural adults short, so rural views, on bus routes or road repairs for example, are barely heard.

  4. Correct it: choose a new sample with about 60 rural and 140 city adults while keeping the half-and-half age split.

    Every variable that could change the answers has to match, not just the first one checked.

Answer: Not representative: rural adults are 5% of the sample but 30% of the town. A representative 200 needs about 60 rural and 140 city adults, with the 50/50 age split kept.

Takeaway: Check every variable that could affect the answer. Matching on one does not make a sample representative.

Where You'll See It

  • Weighting an election poll

    Pollsters compare their samples with census figures on age, sex, region and education. If young adults are 20% of voters but only 12% of a poll's respondents, each young respondent's answer is counted about 20÷12≈1.7 times, so the weighted sample matches the population's mix.

  • Who is included in medicine trials

    A drug tested mostly on young men might work differently in older women. Regulators such as the US Food and Drug Administration ask drug makers to enrol participants who reflect the people who will use the drug, by age, sex and race, so the trial's results can be trusted for them.

  • TV ratings from a panel

    Nielsen estimates what more than 100 million US households watch from a panel of tens of thousands of homes. The panel is chosen so that region, household size, age and other traits match the country's mix; a panel of only city apartments would misjudge what rural families watch.

Common Mistakes

  1. The slip: Assuming a big sample is representative

    Instead: proportions must match, not just counts.

  2. The slip: Checking only one variable

    Instead: a sample can match on age yet miss on income.

  3. The slip: Confusing the method with the result

    Instead: random sampling helps, but you still verify the proportions match.


Mini Practice

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

  1. A town is 30% children, 50% adults and 20% seniors. How many of each should a representative sample of 200 people have?

    Hint: Apply each percent to 200.

  2. A school has 300 sixth graders, 250 seventh graders and 450 eighth graders. How many from each grade should a representative sample of 80 include?

    Hint: Find each grade's share of the 1,000 students first.

  3. A city is 40% renters. A survey of 500 people includes 120 renters. What percent of the sample are renters, and is it representative on this point?

    Hint: Divide the renters by the sample size.

  4. A school is half girls. A sample of 40 students has 30 girls. How many girls would a representative sample of 40 have?

    Hint: Apply the school's share of girls to 40.

  5. Survey A asks 2,000 people, all from one neighborhood. Survey B asks 300 people chosen at random from the whole city. Which is more likely to be representative of the city?

    Hint: Does a bigger sample help if everyone comes from one place?

  6. A sample matches the town on age (20% kids, 60% adults, 20% seniors), but everyone was surveyed at a golf club. Jay says: "It matches on age, so it is representative." Correct him.

    Hint: Could the sample differ from the town in some other way?

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

If a sample's mix is off, can the results be fixed afterwards?

Partly, by weighting. If teenagers are 45% of the town but only 10% of your sample, each teenager's answer is counted 4.5 times and each adult's counted a bit less, so the groups end up in their true shares. Pollsters do this routinely. It works only if the few teenagers you reached answer like the ones you missed, and it cannot add a group that is missing entirely.

Can a poorly chosen sample turn out representative by luck?

Sometimes. A teacher who surveys whichever students arrive first might happen to get the right mix of grades. But you cannot know that without checking, and on the traits you did not check, early arrivers may still differ, perhaps living closer to school or waking earlier. A random method makes a good mix likely on every trait at once, including ones nobody thought of; luck covers only the traits you happened to look at.

Is my own class a representative sample of my school?

Usually not, because one class shares a grade, a timetable, often a teacher and sometimes a neighborhood. On questions where those matter, like homework hours or bedtimes, your class can be far from the school's mix. It might work for a question none of them affects, such as which hand people write with, but you would need to argue that case, not assume it.

Learning Path

← Before

Sampling Bias
Representativeness

You are here

Next →

Nothing after this yet
You need Sampling Bias first: how a selection method can tilt a sample. Representativeness checks the outcome by comparing the sample's group proportions with the population's. With it you can plan a fair survey, such as deciding how many students to ask from each grade, and judge whether a poll's result applies to the whole group; it underlies later work on surveys, experiments and margins of error.

See Also