Collecting data and sampling

We can’t usually ask everyone, so we collect data from a sample. A good sample represents the whole population fairly; a biased one leans one way and gives misleading results.

Key facts

  • Population: everyone you want to know about. Sample: the ones you actually ask.
  • Random sample: everyone has an equal chance of being chosen.
  • Stratified sample: each group is represented in proportion to its size.
  • Bias comes from who you ask, how you ask, and leading questions.

Worked examples

Example 1

A school has 300 pupils in Year 7 and 200 in Year 8. Take a stratified sample of 50.

  1. Total: 500.
  2. Year 7: 300/500 × 50 = 30.
  3. Year 8: 200/500 × 50 = 20.

Answer: 30 from Year 7, 20 from Year 8

Example 2

Why is asking people leaving a gym a biased way to find how much people exercise?

  1. Everyone asked has just exercised.
  2. People who rarely exercise are left out, so the results will be too high.

Answer: They exercise more than most people