Populations and samples
- The population is the whole group you want to find out about. A sample is a smaller part of the population that you collect data from.
- A sample is quicker and cheaper than asking everyone, but it only gives an estimate.
Random samples and bias
- In a random sample, every member of the population has an equal chance of being chosen, for example by using random numbers or picking names from a hat.
- A biased sample doesn't fairly represent the population. Asking people at a gym how often they exercise would give biased results.
- A larger sample usually gives more reliable results.
Collecting data
- Primary data is collected by you. Secondary data was collected by someone else, such as census results or data from the internet.
- Good survey questions are clear and not leading. Response boxes should not overlap, and should cover every possible answer.
Capture-recapture (Higher)
- To estimate a population of animals, catch and mark some, release them, then catch a second sample. Assume the fraction marked in the second sample matches the fraction marked in the whole population.
- 50 fish are marked. Later, 40 fish are caught and 8 are marked. 8/40 = 1/5, so the population is about 50 × 5 = 250 fish.
Key terms
- Population
- The whole group being studied.
- Sample
- A smaller group chosen from the population.
- Random sample
- A sample where every member of the population has an equal chance of being chosen.
- Bias
- When results are unfair because some groups are more or less likely to be included.
- Primary data
- Data you collect yourself.
- Secondary data
- Data collected by someone else.
- Census
- Data collected from every member of the population.
- Capture-recapture
- A method of estimating a population by marking a sample, then checking a second sample.
Quick questions on Sampling
- What is a population in statistics?
- The whole group being studied: It could be all the students in a school.
- What is a sample?
- A smaller group chosen from the population: It is used to estimate facts about the population.
- What makes a sample random?
- Every member has an equal chance of being chosen: Random numbers or names from a hat can be used.
- Asking people at a gym how often they exercise gives what kind of sample?
- A biased sample: People at a gym probably exercise more than most.
- How can you make results from a sample more reliable?
- Use a larger random sample: Bigger random samples represent the population better.
- What is primary data?
- Data you collect yourself: For example, your own survey.
- What is secondary data?
- Data collected by someone else: For example, census results.
- What is wrong with the question 'Don't you agree that school uniform is great?'?
- It is a leading question: It pushes people towards one answer.
Sampling: questions and answers
- Response boxes are 0–5, 5–10 and 10–15 hours. What is wrong with them?
- They overlap. Someone who answers 5 hours could tick two boxes.
- What is a census?
- Data collected from every member of the population. The UK census asks every household.
- Why is a sample often used instead of asking everyone?
- It is quicker and cheaper. Asking everyone can take too long or cost too much.