Sampling methods and the reliability of research

Why sample?

The population is everyone in the group the firm wants to know about (for example all 16 to 24 year olds in a city). A sample is a smaller group chosen to represent it, because researching the whole population would cost too much and take too long.

A bigger, well-chosen sample is more representative and gives more reliable results, but costs more.

Three sampling methods

• Random sampling: every member of the population has an equal chance of being chosen (for example names drawn by computer from a full customer list). It avoids researcher bias, but needs a complete list and may by chance miss small groups.
• Quota sampling: the researcher must interview a set number of people with given characteristics (for example 30 men and 30 women aged 18 to 30), choosing who to ask. Quick and cheap, but not random, so interviewer bias can creep in.
• Stratified sampling: the population is divided into groups (strata) by a characteristic such as age, and a random sample is drawn from each group in proportion to its size in the population. It is representative of key groups but needs detailed population data.

Stratified calculation: sample from a stratum = (stratum size / population) x total sample.

Limitations of market research

• Sample size and bias: small or unrepresentative samples give misleading results.
• Question design: leading or unclear questions distort answers.
• Respondents: people may say what they think the researcher wants, or behave differently from what they say.
• Cost and time, and results may be out of date in a dynamic market.

Exam evaluation: the value of research depends on the method, sample and how quickly the market is changing.

Worked example

A gym has 2,000 members: 1,200 aged under 30 and 800 aged 30 or over. It wants a stratified sample of 150. How many from each group?

1. Under 30: 1,200 / 2,000 x 150 = 90.
2. 30 or over: 800 / 2,000 x 150 = 60.
3. Then choose randomly within each group.

Answer: 90 under 30 and 60 aged 30 or over.

Check yourself

1. In which sampling method does every member of the population have an equal chance of selection?
   • Quota
   • Random
   • Convenience
   • Snowball
   Show answer

   Random. That is the definition of random sampling.

2. An interviewer is told to stop 40 women and 40 men in a shopping centre and chooses who to approach. This is
   • random sampling
   • stratified sampling
   • a census
   • quota sampling
   Show answer

   quota sampling. Set numbers with given characteristics, chosen by the interviewer.

3. A college has 900 students, of whom 270 study business. In a stratified sample of 60, how many should be business students?
   Show answer

   18 students. 270 / 900 x 60 = 18.

4. A firm's 5,000 customers are 3,500 online-only and 1,500 in-store. It wants a stratified sample of 400. How many in-store customers should be included?
   Show answer

   120 customers. 1,500 / 5,000 x 400 = 120.

5. Explain one disadvantage of quota sampling.
   Show answer

   Because the interviewer chooses who to ask, the sample is not random, so interviewer bias may make the results unrepresentative.. Selection by the researcher can bias the sample.

6. Why might a larger sample give more reliable results?
   • It is always cheaper
   • It removes the need for good questions
   • It is more likely to represent the views of the whole population
   • It guarantees honest answers
   Show answer

   It is more likely to represent the views of the whole population. Larger samples reduce the effect of chance, though they cost more.

7. What is the main advantage of stratified sampling?
   • The sample reflects the proportions of key groups in the population
   • It needs no information about the population
   • It is the quickest possible method
   • The interviewer can choose anyone convenient
   Show answer

   The sample reflects the proportions of key groups in the population. Each stratum is represented in proportion.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Business specification (2018)

Last updated 8 November 2026. Spotted a mistake? Tell us.