✅ Strengths
It is fair, because everyone has the same chance. It avoids researcher bias in choosing people. It is likely to give a representative sample, so findings can be generalised.
Investigating Society » Sampling Frames and Sampling Techniques
What you'll learn this session
Study time: 30 minutes
Cambridge spec: 1.3.2
- What a sample and a sampling frame are
- How random, systematic, stratified, snowball and quota sampling are carried out
- The strengths and limitations of each technique
- How to choose a technique to fit a research aim
🔒 Unlock Full Course Content
Sign up to access the complete lesson and track your progress!
Unlock This CourseWhy we sample
Sociologists rarely have the time or money to ask everyone in the target population. Instead they choose a smaller group to study. This smaller group is called the sample. The aim is a sample that is representative, so the findings can be generalised to the whole population.
Key terms:
- Sample: the smaller group of people chosen from the target population to take part in the research.
- Sampling frame: a list of everyone in the target population, from which the sample is chosen.
- Sampling technique: the method used to choose the people for the sample.
The sampling frame
Most sampling techniques start with a sampling frame. Examples are a school register, an electoral register (a list of people who can vote in some societies), a membership list or a list of patients at a clinic.
A good frame is complete, up to date and free from mistakes. A poor frame leaves people out, for example a school register that has not been updated since the start of term. People who are not on the list have no chance of being picked, so the sample may not be representative.
Random sampling
Drawing names from a hat is random sampling: everyone has an equal chance of being picked
Random sampling: every person in the sampling frame has an equal chance of being chosen.
How it is carried out: give everyone on the frame a number, then pick numbers using a computer, or by drawing names from a hat. Nobody, including the researcher, decides who is chosen.
Example: a researcher gives each of the 1,000 students in a school a number and a computer picks 100 numbers.
❌ Limitations
It needs a complete sampling frame, which is often not available. It can be slow and costly if the people picked live far apart. By chance, a small group such as a minority may be missed.
Systematic sampling
Systematic sampling: every nth person on the sampling frame is chosen, for example every 10th name.
How it is carried out: divide the size of the frame by the sample size you want. This gives the gap (the interval). Choose a starting point, then take every person at that gap.
Example: a school has 1,000 students and the researcher wants 100. The gap is 1,000 divided by 100, which is 10. Starting at student 4, the sample is students 4, 14, 24, 34 and so on.
✅ Strengths
It is quick, simple and easy to carry out. It spreads the sample across the whole frame. There is little room for the researcher to choose people they like.
❌ Limitations
It needs a full sampling frame. If the list has a pattern, the sample can be biased, for example if the list is made up of groups of 10 students ordered from oldest to youngest, every 10th name would always be the youngest. It is not truly random, because people next to each other on the list cannot both be chosen.
Stratified sampling
Stratified sampling: the population is split into groups (strata) that share a feature, such as gender or age. The sample then contains each group in the same proportion as in the population.
How it is carried out:
- Decide which feature matters for the research.
- Divide the sampling frame into groups using that feature.
- Work out each group's share of the population.
- Pick people at random from within each group, to match that share.
Example: in a school of 1,000, 60% are girls and 40% are boys. For a sample of 100, the researcher randomly picks 60 girls and 40 boys.
✅ Strengths
It guarantees that each important group is included in the right proportion. It is more representative than a random sample for the feature chosen. Groups can be compared fairly.
❌ Limitations
It needs a detailed frame that shows each person's group. It takes more time and planning. The researcher must decide which feature matters, and may choose the wrong one.
Snowball sampling
Snowball sampling: each person introduces the researcher to others, so the sample grows like a rolling snowball
Snowball sampling: the researcher finds one or two people in the group, who then introduce them to others, who introduce them to more. The sample grows like a rolling snowball.
How it is carried out: make contact with a first member, ask them to suggest other people, then ask those people for further names, until the sample is big enough.
Example: there is no list of young carers (students who look after a family member). A researcher meets one through a youth club. That student introduces two friends who are also carers, and they each introduce more.
✅ Strengths
It works when there is no sampling frame. It helps reach groups that are hidden or do not trust outsiders. Introductions from a friend can make people more willing to take part.
❌ Limitations
The sample is not random, so it is unlikely to be representative. People tend to name friends who are similar to them, so the sample can be biased. It can take a long time to build.
Quota sampling
Quota sampling: the researcher is given a set number (a quota) of people from each group to interview, and keeps going until every quota is full. The researcher chooses who to ask.
How it is carried out: decide the groups and how many of each are needed, then approach people, often in a public place, and check which group they fit. Stop asking a group once its quota is filled.
Example: an interviewer in a shopping street must find 20 people aged 18 to 30, 20 aged 31 to 50 and 20 aged over 50.
✅ Strengths
It does not need a sampling frame. It is quick and cheap. Every group needed is included.
❌ Limitations
People are not chosen at random, so the sample may not be representative. The interviewer may choose people who look friendly, which causes bias. People who are in a street at that time of day may not be typical.
Stratified or quota?
Both use groups. In stratified sampling people are picked at random from a sampling frame. In quota sampling the interviewer chooses people until the quotas are full, with no frame.
Choosing a technique
- Is there a complete list? If yes, random, systematic or stratified are possible.
- Do particular groups need to be included in the right proportion? Use stratified or quota.
- Is the group hidden, with no list? Use snowball.
- Is time or money short? Systematic or quota are quicker.
Common mistakes
Saying a quota sample is random (it is not). Writing "it is representative" for snowball sampling. Mixing up the sampling frame (the list) with the sample (the people chosen). Giving a strength without saying why it is a strength, for example "it is quick" with no link to the research.
Exam-style question
A researcher wants to study how students in a school of 1,000 use their free time. The school has a full register showing each student's year group.
Explain one strength and one limitation of stratified sampling for this research. [4 marks]
Model answer
Strength: it makes sure each year group is included in the same proportion as in the school (1), so the sample is more likely to be representative and findings can be generalised (1). Limitation: it takes more time and planning, because the register must be split by year group and people picked from each group (1), so it is slower than choosing every 10th name (1).
Exam tip
For a "one strength and one limitation" question, name the point, then add "so..." to explain why it matters for this research. Each part is worth 2 marks.