Would Station Road buy lunch?: Lena and Marco have found the site for a fourth shop on Station Road, and they want it selling 150 lunches a day.
Before they sign the lease, Marco wants to know
- Whether people nearby would buy lunch from Lena's
- What they would want
Too many to ask: The 2021 census (4.4.2) counts about 8,400 people living within a ten-minute walk of the site. He cannot ask them all.
The 8,400 are the population: everyone the question is about. The 200 people Marco actually asks are the sample. Choosing which 200 to ask is sampling.
The questionnaire itself is primary research (4.4.1). Sampling decides who gets to answer it.
Why not ask all 8,400?
Money
At ten minutes an interview, 8,400 people is 1,400 hours of someone's time. 200 people is about 33 hours: two staff for a few days.
Time
The landlord wants an answer within a month. By the time the last of 8,400 had been asked, the site would have gone to someone else.
It is not possible
Nobody could track down all 8,400 people near Station Road, at home and at work, and many would refuse to stop.
It is not needed
A well-chosen 200 tells Marco almost what 8,400 would, which is the whole point of a sample.
A sample is only useful if it looks like the population: If Marco's 200 were all retired people who live near the station, their answers would describe retired people, not the office workers who buy most lunches.
A sample that has the same mix of people as the population is representative. Only then can its answers stand in for everyone's.
So the first question in sampling is not how many people, but which people. There are three ways to choose them.
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.
Marco has three ways to pick his 200. Each one answers a different question: who is easiest to ask, who is picked by pure chance, or how many from each group.
Convenience sampling
- Ask whoever is easiest to reach
- Kofi's counter staff ask the first 200 customers into the Market Square shop
- Cheap, quick and simple: the staff and the customers are already there
- The people asked may be nothing like the population, so the results can lean one way
Random sampling
- Every person in the population has the same chance of being picked
- A computer picks 200 names from a full list of all 8,400 people near Station Road
- Nobody chooses who is asked, so the results are free of the researcher's bias
- Needs a full list, which costs time and money to build, and some people picked never reply
Quota sampling
- Set how many to ask from each group, then fill each group
- 100 office workers, 60 people at home in the day and 40 students, asked on Station Road
- Every important group is included, quickly and with no list needed
- Inside each group the interviewer chooses who to stop, so bias can creep back
Random means equal chance, not just anyone: In everyday speech 'random' means anyone at all. In sampling it means something stricter: every person in the population has exactly the same chance of being picked. For Marco, 200 in 8,400, or about 1 in 42.
Stopping whoever walks past is not random. People who never walk past have no chance at all.
Quota sampling starts from what Marco already knows about the area. The town council's figures (secondary research, 4.4.2) say that of the 8,400 people near Station Road, about half work in offices, 30% are at home during the day (retired people, parents, shift workers) and 20% study at the college.
So he sets quotas to match: 100 office workers, 60 people at home in the day and 40 students, which is 200. His interviewers stop people on Station Road until each quota is full.
Quota sampling is not random: The quotas make the mix of groups right. They do not decide who, inside a group, is asked.
An interviewer who needs ten more office workers will stop the ones who look friendly and unhurried, and skip the ones on the phone. Those are the people most likely to grab a quick lunch.
Get feedback like a real examiner
Submit your answers and get instant feedback — what you did well, what's missing, and exactly what to write to score full marks.
One question, two very different answers: Two samples of 200 people are asked the same question: would they buy lunch from Lena's on Station Road?
| Sample | Who was asked | Would buy lunch |
|---|---|---|
| Kofi's staff | 200 customers in the Market Square shop, between 7 and 9 in the morning | 160, or 80% |
| Marco's quota sample | 200 people on Station Road at lunchtime | 92, or 46% |
Who was asked: Same question, same number of people. The only difference is who was asked.
The Market Square sample leans one way. Everyone in it already buys from Lena's and likes it enough to come in. Nobody who buys Goldcrust's cheaper bread down the road was asked, and hardly anyone who lives near Station Road.
When the way people are chosen pushes the answers one way, the sample is biased (sampling bias). The 80% tells Marco about his own morning customers, not about Station Road.
Where bias creeps in
Who is easy to reach
Convenience sampling asks whoever is nearby. Here that is Lena's own loyal customers, at breakfast time, in the wrong part of town.
Who the interviewer picks
In a quota or convenience sample, a person decides whom to stop, and people choose people like themselves or people who look easy to talk to.
Who agrees to answer
Busy office workers refuse more often than retired people at home. Even a random sample leans if the people who say no think differently from those who say yes.
Who is asking
Lena's own staff, asking in Lena's shop, hear polite answers. Customers do not like saying no to the person who serves them every day (interviewer bias).
Size matters too, but in a different way. In a small sample each person's answer weighs a lot, so the result jumps about. In a big one each answer weighs little, so the result is steadier.
| People asked | One person changing their answer moves the result by | Time at ten minutes each |
|---|---|---|
| 20 | 5 percentage points (16 of 20 is 80%; 15 of 20 is 75%) | about 3 hours |
| 200 | 0.5 percentage points (160 of 200 is 80%; 159 of 200 is 79.5%) | about 33 hours |
| 2,000 | 0.05 percentage points | about 333 hours |
A bigger sample does not cure bias: If Kofi's staff asked 2,000 Market Square customers instead of 200, the result would be steadier. It would still be a result about Market Square customers.
Size makes an answer steadier; only a better way of choosing people makes it fairer. And each extra person costs time and money, so a business stops at a sample big enough to trust.
No sampling method is best everywhere. The right one depends on four things: how much money and time the business has, whether a full list of the population exists, and whether particular groups must be included.
| Cost and speed | Needs a full list? | Risk of bias | Suits a business that | |
|---|---|---|---|---|
| Convenience | Cheapest and quickest | No | High: only the people nearby | needs a quick first look on a small budget |
| Random | Slowest and dearest | Yes | Low: chance decides who is asked | has a full list and a big decision riding on the answer |
| Quota | Quick and fairly cheap | No, but it needs the population's mix | Medium: the interviewer picks within each group | must hear from several groups, in the right proportions |
What Marco chooses, and why: Marco rules out random and convenience sampling, and chooses quota sampling.
Weighing the three methods
Random sampling is out
No list of all 8,400 people near Station Road exists, and building one would take longer than the landlord will wait.
Convenience is out
Convenience at Market Square asks the wrong people.
Quota sampling is in
Quota sampling costs about as little as convenience and makes sure office workers, people at home in the day and students are all heard, in the right mix.
The catch
Marco accepts the catch and tells his interviewers to ask at different times and on both sides of the street.
The choice would change with the job. If Lena wanted the views of the cafés and hotels she supplies, she has every one of them on her invoice list, so a random sample would be easy and fair.
Whenever a question asks for an advantage or a disadvantage of a method for a business, give the feature of the method and then show what it means for that business: its budget, its customers, where and when the asking happens.
Match the method to the job: Little money and a quick first look: convenience. A full list and a big decision: random. Several groups that must all be heard: quota.
Then check who might be missing from the sample, because that is where its bias will come from.
Learn what examiners really want
See exactly what to write to score full marks. Our AI shows you model answers and the key phrases examiners look for.
How this comes up: A case describes how a business plans to sample: who will be asked, where, when and by whom. You explain an advantage, a disadvantage, or both.
The usual questions
- Four marks. Explain one advantage and one disadvantage for the business of using convenience sampling.
- Two marks, one side only. Explain one advantage for the business of using quota sampling.
- Two marks, one side only. Explain one benefit of collecting its research using random sampling.
Where the second mark comes from: Every time, the second mark of each point comes from the case: the when, the where and the who.
The four-mark pattern
- Explain the advantage. Convenience sampling is cheap and quick to organise: the sample is whoever is already there, and no list is needed.
- Show it in the case. Kofi's counter staff are in the Market Square shop anyway, so 200 interviews over two mornings cost Lena almost nothing.
- Explain the disadvantage. The people who are easy to reach may not be representative, so the results are biased.
- Show it in the case. Morning bread buyers at Market Square are Lena's own customers; the office workers who would buy lunch on Station Road are hardly asked, so the 80% overstates demand.
Trap: size is not a feature of the method: 'Only 200 people were asked' describes how the business used convenience sampling, not what convenience sampling is. The weakness is who is asked, not how many.
Trap: no case, no second mark: 'It may be biased' is one mark. 'Only Lena's own customers at 7 a.m. were asked, so the 80% overstates how many Station Road workers would buy lunch' is two.
Explain one advantage and one disadvantage for Lena's Bakery of using convenience sampling.
Model answer plan
See the answer broken into clear points, with guidance on how each point earns marks.