In one line: Research is done on a sample and its conclusions are about a population. Sampling is the link between the two.
Before an election, pollsters question about a thousand people and predict how millions will vote.
It works when the thousand are chosen so that they resemble the millions: the same spread of ages, regions and incomes.
When they are not, the prediction fails, however many people were asked.
Three words to start with
Target population
The whole group the conclusions are meant for.
For a study of gaming and sleep in one town: every teenager who lives there.
Sample
The participants: those who take part.
Here: the 120 teenagers who keep a sleep diary for a week.
Representative
The sample mirrors the population in the ways that matter: age, school, how much they play.
Population · Sample · Match
A representative sample allows findings to be generalised to the population.
The method used to obtain the sample is the sampling technique.
Remember: Conclusions travel only as far as the sample resembles the population.
Free preview
This is the free notes preview
You're reading the free notes. Aimnova Pro unlocks the full study experience — and you can try it with your first topic free to keep:
- FlashcardsLock in vocabulary and key terms with spaced repetition.
- Practice questionsAnswer exam-style questions and get instant AI marking.
- Mock exams & past-paper vaultSit full mocks and see exactly how examiners award marks.
- Personalised study planA daily plan built around your exam date and weak areas.
Key idea: There is a trade-off in every choice of technique: representativeness against practicality.
The five techniques
Random sampling
Each member of the target population has an equal and independent chance of selection.
Strength: no group is favoured, so the sample is likely to be representative.
Limit: a complete sampling frame is required, a small sample can still be unrepresentative by chance, and refusals cause non-response bias.
Stratified sampling
The population is divided into strata, such as schools or age groups, and a random sample is drawn from each in proportion to its size.
Strength: key subgroups are guaranteed their share, and can be compared.
Limit: detailed information about the population is needed, and it is slow to organise.
Opportunity sampling
Participants are those who are available and willing: teenagers leaving a gaming café on a Saturday. Also called convenience sampling.
Strength: fast, cheap, useful for a pilot.
Limit: open to sampling bias and to the researcher's choice of whom to approach.
Self-selected sampling
Participants respond to an advertisement or poster of their own accord. Also called volunteer sampling.
Strength: little effort for the researcher, and participants are motivated and have clearly agreed.
Limit: those who volunteer differ from those who do not.
Snowball sampling
A small starting group recruits further participants from its own network, who recruit others in turn.
Strength: gives access to hidden or hard-to-reach groups, with trust already in place.
Limit: participants resemble one another, and they know who else took part.
Chance · Chance within groups · Convenience · Volunteers · Referrals
Random and stratified sampling are probability techniques: selection is left to chance. The other three are non-probability techniques: participants are not chosen at random from the target population.
One more you will meet: purposive sampling: Participants are selected deliberately because they have the characteristics the study needs: teenagers who play online for more than four hours a day.
It is the usual choice for interviews and focus groups, and it too is a non-probability technique.
Practice with exam-style questions
Answer exam-style questions and get AI feedback that shows you exactly what examiners want to see in a full-marks response.
Key idea: A sample is biased when the way it was obtained makes it systematically different from the target population.
Three routes
People select themselves
Those who respond to an advert have a reason to: interest in the topic, a strong opinion, or the payment.
A poster about "gaming and sleep" draws the keenest gamers.
The researcher picks
When the researcher chooses whom to approach, her own preferences enter, usually unnoticed: people like herself, people who look willing.
Groups are left out
The route of recruitment excludes some people altogether: a notice at school misses teenagers who have left school.
Refusals matter too. If those who decline differ from those who accept, the result is non-response bias.
Self-selection · Researcher's choice · Exclusion
Psychology has a standing example. Between 60 and 80% of Western studies are estimated to use university students, mostly studying psychology.
Henrich, Heine and Norenzayan (2010) named the wider pattern WEIRD: samples drawn from Western, Educated, Industrialised, Rich and Democratic societies, which are unrepresentative of most of the world.
Exam tip: Does the bias matter for this finding?
A student sample may be fine for a study of how memory works, and misleading for a study of social media and mood, where age and culture change the result.
Key idea: External validity is the extent to which findings hold beyond the participants, the setting and the time of the study.
External validity: three questions
Population validity
Can the results be applied to the target population?
Sample size and sampling technique bear directly on this one.
Ecological validity
Would the same results appear in a real-world setting?
Temporal validity
Would the same results appear in a different period?
People · Setting · Time
Two words are used for applying findings more widely.
Generalisability belongs to quantitative studies with larger samples: results are applied to a population.
Transferability belongs to qualitative studies: findings are applied, with care, to a context that resembles the one studied.
Bigger is not the same as better: Size reduces the effect of chance. It does nothing about bias.
Ten thousand readers of one gaming forum tell you about readers of that forum.
Feeling unprepared for exams?
Get a clear study plan, practice with real questions, and know exactly where you stand before exam day. No more guessing.
How this is asked: A study you are given always has a sample. Whatever the question, you need to be able to say how the sample was obtained and how that could bias the findings. Practise it as a short answer.
Four moves, four short paragraphs
- Name the technique.
- Show how it was carried out, with details from the study.
- Identify who the technique favoured or left out.
- Say what that does to the findings.
The trap: describing the sample, not the technique: "The sample was 120 teenagers aged 13 to 17" describes who.
The technique is how they were obtained: by chance, by convenience, by volunteering or by referral.
A researcher wanted to find out what secondary school teachers think about banning phones in class. The head teacher of one school chose ten teachers "who will have interesting things to say", and each was interviewed for 30 minutes. Nine of the ten supported a ban. Describe the sampling technique used in this study and outline one way it could introduce bias.
Model answer plan
See the answer broken into clear points, with guidance on how each point earns marks.