Sampling Methods
Sampling methods are the different ways of choosing a smaller group (a sample) from a larger population so that the sample can be used to make conclusions about the whole population without surveying everyone. Common methods include random sampling (every member has an equal chance of selection), systematic sampling (selecting every nth member) and quota sampling (selecting a fixed number from each category). Choosing an appropriate, unbiased method is essential so the sample fairly represents the population.
Before you start
Make sure you're comfortable with these topics first:
This is our main guide to stratified sampling. Related guides:
Method
- Identify the population you want to study and, if possible, obtain a sampling frame (a list of every member of the population).
- Decide on a sample size that is large enough to be representative but practical to collect.
- For simple random sampling, number every member of the sampling frame and select using random numbers (e.g. from a random number table or generator).
- For systematic sampling, work out the sampling interval (population size divided by sample size), then select every nth member starting from a random point.
- For quota sampling, decide how many people to select from each category (e.g. age group) and then select non-randomly (e.g. the first people who fit each category) until each quota is filled.
- State any bias or limitations in the chosen method, such as a sampling frame that is out of date or a non-random element in quota sampling.
Common mistakes
MistakeFinding the systematic sampling interval by dividing the sample size by the population size, which produces a fraction instead of the number of places to move.
InsteadCalculate population size divided by sample size, then use that interval after a random start. Check that the interval times the sample size covers the population.
MistakeStarting a systematic sample at the first person on the list, rather than choosing the starting position at random within the first interval.
InsteadNumber the sampling frame, choose a random start in the first interval, and then add the interval repeatedly so the beginning of the list is not favoured.
MistakeCalling quota sampling random because the quotas match population categories, even though the people within each quota are chosen non-randomly.
InsteadSeparate the two ideas: quotas can make category totals resemble the population, but selecting the first convenient people within a quota can still introduce bias.
Worked examples
Example 1
A head teacher wants to survey 40 students out of a school of 800 students using systematic sampling. Explain how she should select the sample.
- Find the sampling interval: population divided by sample size = 800/40 = 20.
- Obtain a numbered list of all 800 students (the sampling frame).
- Choose a random starting point between 1 and 20, for example using a random number generator; say the number 7 is chosen.
- Select the 7th student, then every 20th student after that: the 27th, 47th, 67th, and so on, until 40 students have been selected.
- Final answer: sampling interval = 20; select a random start between 1 and 20, then every 20th student on the list after that.
Example 2
A museum has 1260 annual members and wants a systematic sample of 36 members. Its membership list is numbered from 1 to 1260, but it records only daytime contact details. A random start of 17 is chosen. Find the sampling interval, list the first four selected members, and suggest one limitation of this sampling frame.
- Sampling interval = population size divided by sample size = 1260 / 36 = 35.
- The random starting point 17 is within the first interval from 1 to 35.
- Add 35 each time: the first four selected members are 17, 52, 87 and 122.
- The daytime-only contact details may make members who cannot respond during the day less likely to take part, causing non-response bias even though the selection from the list is systematic.
- Final answer: interval = 35; select members 17, 52, 87 and 122 first, and note that the daytime-only contact frame may exclude some members.
Example 3
In an invented scenario, a council has 120 residents on a numbered list arranged in a repeating order: cyclist, walker, bus user, car user, then the same order again. It wants a systematic sample of 30 residents and uses random start 2. Show the interval and first four selections, identify the group represented, and explain why the sample is biased even though the interval is correct. Check whether a different random start avoids the problem.
- Sampling interval = population size / sample size = 120 / 30 = 4. The starting position 2 is within the first interval, positions 1 to 4.
- Add 4 each time. The first four selected positions are 2, 6, 10 and 14, and the thirtieth is 2 + 29 x 4 = 118, so all 30 selections are within the list.
- Positions 2, 6, 10 and 14 are all walkers. Because the list repeats every 4 positions and the interval is also 4, every selected position has the same place in the repeating pattern.
- There are 30 walkers in the list, but this sample selects all 30 walkers and no cyclists, bus users or car users. It therefore does not represent the travel modes fairly, despite using the calculated interval.
- Changing the random start to 1, 3 or 4 would select cyclists, bus users or car users respectively, still selecting only one group each time. The random start changes which group is chosen, but does not remove the repeating-pattern bias.
- Final answer: the interval is 4 and start 2 selects walkers only, so the sample is biased by the list's repeating order; a different random start still selects just one travel group.
Practice questions
Q1[2 marks]A population has 200 members and a sample of 20 is needed using systematic sampling. Find the sampling interval.Show answer
Answer: 10 (200/20 = 10)
Q2[1 mark]State one advantage of random sampling.Show answer
Answer: Every member of the population has an equal chance of being chosen, which reduces bias (accept any valid reason)
Q3[2 marks]A researcher wants a sample of 50 from a population of 500 using systematic sampling, with a random start of 4. List the first three people selected.Show answer
Answer: 4, 14, 24 (interval = 500/50 = 10, so start at 4 then add 10 each time)
Q4[1 mark]State one disadvantage of quota sampling.Show answer
Answer: It is not random, so it can introduce bias (e.g. the researcher may choose people who are easiest to approach)
Q5[1 mark]A school of 600 students wants a sample of 30. A teacher decides to only survey students in her own class of 30. State one problem with this method.Show answer
Answer: It is not representative of the whole school (it is a convenience/biased sample, not random)
Q6[2 marks]A gym has 900 members. A sample of 45 is needed using systematic sampling, with a random start of 12. Find the 4th person selected.Show answer
Answer: 72 (interval = 900/45 = 20; selections are 12, 32, 52, 72)
Exam-style questions
Written in the style of a GCSE Statistics exam paper, with a full mark scheme.
Exam tips
- Name the population and sampling frame before describing the selection, then check that the frame actually includes every group you want to study.
- For systematic sampling, show population divided by sample size, state the random starting range, and add the interval each time.
- When evaluating a method, identify who could be missed and explain how that missing group could make the results unrepresentative.
Q1[2 marks]A population has 350 members. A sample of 35 is required using systematic sampling. Find the sampling interval and state the rule used to select members after the random starting point.Show mark scheme
Tick each line you got.
Nothing ticked yet - 2 available
Q2[3 marks]A council wants to find out residents' opinions on a new park. They decide to only survey people walking through the park on a sunny Saturday afternoon. (a) State the sampling method used. (b) State one way this sample could be biased. (c) Suggest one improvement to the sampling method.Show mark scheme
Tick each line you got.
Nothing ticked yet - 3 available
Q3[4 marks]A factory produces 1200 items per day and wants a systematic sample of 24 items to check for faults. (a) Find the sampling interval. (b) If the random starting point is item number 15, list the first four items selected. (c) State one advantage of systematic sampling over simple random sampling in this context.Show mark scheme
Tick each line you got.
Nothing ticked yet - 4 available
See real GCSE Statistics past-paper questions, with official mark schemes
This topic in real past papers
We have tagged some boards' past papers question by question (not every board or year yet). This topic comes up in:
- AQA GCSE Statistics past papers: 17 questions across 14 papers, latest tagged June 2023
- Edexcel GCSE Statistics past papers: 17 questions across 10 papers, latest tagged June 2023
Free printable worksheet
Want more practice on paper? Download the Sampling Methods worksheet pack - 28 pages of exam-style questions with a full mark scheme. Free to download, no card or account. Print it for personal and classroom use.
This topic is chapter 7 of GCSE Statistics Foundation Essentials 1 and chapter 7 of GCSE Statistics Higher Essentials 1, the whole course as one free printable PDF. Its worksheet is also chapter 7 of GCSE Statistics Foundation Topic Workbook 1 and chapter 7 of GCSE Statistics Higher Topic Workbook 1, free printable practice books.
Other versions of this worksheet: