IB Business Management HLTopic 4 — MarketingPaper 1 & 2Core skill~9 min read
Choosing a Sample
No business can ask every customer what they think — it would cost a fortune and take years. So you ask a few hundred and assume the rest would have said something similar. That assumption is only safe if you pick those few hundred properly.
📘 What you need to know
Sampling means getting opinions from a selected group in order to draw conclusions about the whole market.
The population is everyone the business is interested in — not everyone in the country.
Results from a sample are extrapolated: what the sample says is assumed to be true of the population.
Quota sampling fixes proportions of each group in advance.
Random sampling gives every member of the population an equal chance of being chosen.
Convenience sampling uses whoever is easiest to reach or willing to take part.
Generally, the larger the sample, the more likely the results reflect the market — but the more it costs.
The idea behind sampling
Asking everyone is called a census, and for a business it is almost always impossible. So researchers take a small group, ask them, and apply the answers to the whole market. A sample of 500 chosen well beats a sample of 5,000 chosen badly.
Every decision made afterwards rests on those six dots being typical. That is why how you choose them matters more than how many you choose.
Careful with the word “population”. For a business it means the group it is interested in — local dog owners, or students at one university — not the population of a country. Assuming otherwise is a common mistake.
The three methods you need
The colours in the quota panel are the groups the researcher decided on in advance. In the convenience panel nobody decided anything — the sample is simply whoever was standing there.
Quota sampling
The researcher decides in advance how many people from each group must be interviewed, so the sample mirrors the shape of the market. A family car maker might set 25% aged 18 to 24, 50% aged 25 to 45, and 25% aged 46 and over.
It only works if you already know what the market looks like — and interviewers still choose who to stop, so bias creeps back in.
Random sampling
Every member of the population has an equal chance of being picked, usually by drawing names from a full list such as a membership database. Because nobody chooses who takes part, bias is largely avoided.
The catch is that you need a complete and accurate list of the population. And randomness can still, by chance, hand you a sample of mostly one type of person.
Convenience sampling
You ask whoever is nearest and willing: regular customers during a quiet hour, friends, people outside the shop. It is quick, cheap and produces a lot of data fast.
It is also the weakest. People known to the researcher, or standing in one place at one time of day, are not the market.
Method
Advantages
Disadvantages
Quota
Quick and cheap, and the proportions match the market, so results can be read with insight
Not random, so bias remains. You must already understand the population to set the quotas
Random
Simple to design and interpret, and bias is largely avoided because anyone can be chosen
Needs a complete, accurate list of the population, and the sample may still not be representative
Convenience
Respondents are readily available, so large amounts of data can be gathered very quickly
Likely to be biased towards people close to the researcher, and rarely represents the whole market
How a business actually chooses
Time available — with little time, random sampling from an existing list is quick to organise.
Knowledge of the population — if the business knows its market well, quota sampling gives data that lacks bias and can be read with insight.
Skills of the researchers — where staff have no research training, convenience sampling produces something usable and easy to interpret.
Budget — bigger samples cost more. The question is always whether the extra accuracy is worth the extra money.
Sample size and sample method are two separate marks. A big sample chosen badly is still biased; a small sample chosen well can be surprisingly reliable. Say both things and you have covered the question properly.
Worked examples
WORKED EXAMPLE 1
A car maker will interview 400 people using quota sampling: 25% aged 18 to 24, 50% aged 25 to 45 and 25% aged 46 and over. Calculate how many people will be interviewed in each group. [3 marks]
Step 1: Youngest group0.25 × 400 = 100Step 2: Middle group0.50 × 400 = 200Step 3: Oldest group0.25 × 400 = 100Step 4: Check it adds up100 + 200 + 100 = 400100, 200 and 100 respondentsAlways finish by checking the groups total the sample size. It catches a slip in seconds.
WORKED EXAMPLE 2
A gym wants to know why members are cancelling. It has a full membership database. Recommend a sampling method. [4 marks]
Step 1: Spot the useful detail
A full database means a complete list of the population already exists.
Step 2: Match it to a method
That is exactly what random sampling needs, and it removes the bias of only asking members who still turn up.
Step 3: Say why the others fail here
Convenience sampling would ask people at the gym — the ones who have not cancelled.
Random sampling from the membership listThat third step is the mark most students miss. Ruling out the wrong method proves you understand all three.
💡 Exam tip
Look in the case for a list of customers. A list points to random sampling; no list usually means quota or convenience.
Justify your choice against time, money and the researchers’ skill, not just accuracy.
Say why the other methods were rejected. It shows understanding and picks up the higher marks.
Quota percentages are a simple calculation. Do them carefully and check the total.
Link the sample straight back to the decision: unreliable data leads to a bad marketing mix.
Use the word representative. It is the term examiners are looking for.
⚠ Common mix-up
Random does not mean casual. Stopping people in the street is convenience sampling, not random.
Population is not a country’s population. It is the group the business cares about.
Quota is not random. The proportions are fixed, but the interviewer still picks the individuals.
Bigger is not automatically better. A large biased sample is confidently wrong.
A sample is not proof. Results are extrapolated, which always carries risk.
Sampling is not a research method. It is how you choose who to use a method on.
Up next: The Marketing Mix (4.5) — product, price, place and promotion, plus the extra three Ps that services depend on. Everything in Topic 4 so far has been getting you ready to build one.
Want this explained one-to-one?
Book a free session with an experienced IB Business Management tutor and get your trickiest topics made simple.
