Populations, censuses and samples
- The population is everything or everyone you want to find out about. A census collects data from the whole population; a sample collects data from part of it.
- A census is accurate and complete, but slow and expensive, and impossible if items are destroyed by testing. A sample is quicker and cheaper, but may not represent the population.
- A sampling frame is a list of everyone in the population, such as a school register. Each member is a sampling unit.
- Bias means the results lean one way and don't fairly represent the population. Larger, well-chosen samples reduce the effect of chance.
Random methods
- Simple random sampling: every member has an equal chance. Number the sampling frame and use a random number generator, ignoring repeats.
- Systematic sampling: choose every kth member from a list, after a random start. k = population size ÷ sample size. For 500 people and a sample of 25, k = 20.
- Stratified sampling: split the population into groups (strata), such as year groups, and sample each in proportion to its size. Number from a stratum = stratum size ÷ population size × sample size. Then choose randomly within each stratum.
- Cluster sampling: split the population into clusters (such as schools), randomly choose some clusters and sample within them. It is cheaper but may be less representative.
Non-random methods
- Quota sampling: the interviewer chooses a set number of people from each group, such as 20 men and 20 women. It is quick but not random, so it can be biased.
- Convenience (opportunity) sampling: using whoever is easiest to reach. It is quick but very likely to be biased.
- Judgement sampling: the researcher picks people they think are representative.
Capture-recapture
- The Petersen method estimates the size of an animal population. Catch and mark a sample (M), release them, then catch a second sample (C) and count how many are marked (R).
- Estimate N = M × C ÷ R. If 40 fish are marked, and a later catch of 50 has 10 marked, N = 40 × 50 ÷ 10 = 200.
- It assumes the population doesn't change, marks aren't lost, marked animals mix back in and every animal is equally likely to be caught.
Key terms
- Population
- Everyone or everything you want to find out about.
- Census
- Collecting data from the whole population.
- Sampling frame
- A list of every member of the population.
- Simple random sample
- A sample where every member has an equal chance of being chosen.
- Systematic sample
- Choosing every kth member from a list after a random start.
- Stratified sample
- Sampling each group in proportion to its size.
- Quota sample
- Choosing a set number from each group, not randomly.
- Bias
- When results don't fairly represent the population.
Quick questions on Populations and sampling
- What is a census?
- Data collected from the whole population: The UK holds one every ten years.
- What is a sampling frame?
- A list of every member of the population: Such as a school register.
- What is a disadvantage of a census?
- It is slow and expensive: Everyone must be included.
- A school has 1,200 students and 300 are in Year 7. How many Year 7 students should be in a stratified sample of 80?
- 20: 300 ÷ 1,200 × 80 = 20.
- A list has 600 names and a sample of 30 is needed. What is k for a systematic sample?
- 20: 600 ÷ 30 = 20.
- Which sampling method gives every member an equal chance?
- Simple random sampling: Use a random number generator.
- Which sampling method is most likely to be biased?
- Convenience sampling: It uses whoever is easiest to reach.
- 30 birds are ringed. Later, 40 are caught and 6 have rings. Estimate the population.
- 200: 30 × 40 ÷ 6 = 200.
Populations and sampling: questions and answers
- What is a quota sample?
- Choosing a set number from each group, not randomly. Market researchers often use it.
- Which is an assumption of capture-recapture?
- The population size doesn't change between samples. No births, deaths or migration.