Skip to main content
ScotlandModern StudiesSyllabus dot point

How do social researchers choose a sample, and what makes a sample representative in Advanced Higher Modern Studies?

Sampling: the population and sampling frame, probability sampling (random, systematic, stratified, cluster) and non-probability sampling (quota, snowball, convenience), sample size, and representativeness.

How sampling works in SQA Advanced Higher Modern Studies. Covers populations and sampling frames, probability methods (random, systematic, stratified, cluster), non-probability methods (quota, snowball, convenience), sample size, representativeness and the trade-offs that decide which method fits a study.

Generated by Claude Opus 4.813 min answer

Reviewed by: AI editorial process; not yet individually human-reviewed

Have a quick question? Jump to the Q&A page

Jump to a section
  1. What this key area is asking
  2. Population, sampling frame and sample
  3. Probability sampling
  4. Non-probability sampling
  5. Sample size and representativeness
  6. Worked example
  7. Try this

What this key area is asking

Almost no study can examine an entire population, so researchers study a sample and use it to draw conclusions about the whole. This dot point covers how samples are chosen, the difference between probability and non-probability methods, and what makes a sample representative enough to trust. Sampling is examined directly in the question paper and is a decision you must justify in your project-dissertation.

Population, sampling frame and sample

The frame matters as much as the method. If the frame omits part of the population, no sampling technique can repair the resulting bias, so a researcher must judge how well the frame matches the population they want to describe.

Probability sampling

  • Simple random. Every member has an equal chance; selection is by lottery or random numbers. Strong for generalisation but needs a full frame.
  • Systematic. Every nth member of the frame is chosen after a random start. Quick, but risky if the list has a hidden pattern matching the interval.
  • Stratified. The population is split into strata (such as age or region) and each is sampled in proportion, guaranteeing every subgroup is represented. Ideal for comparing groups, but needs a frame that records the strata.
  • Cluster. The population is divided into clusters (such as schools or constituencies), some clusters are randomly chosen, and members within them are studied. Cheaper over a wide area, but less precise if clusters differ.

Non-probability sampling

Non-probability methods are common in qualitative and exploratory work, and in studying groups with no usable frame (for example, undocumented workers). The trade-off is clear: you gain access and speed but lose the ability to generalise, so conclusions must be stated cautiously.

Sample size and representativeness

A larger sample reduces sampling error, the random difference between the sample and the population, but raises cost and time, so size is always a compromise. Crucially, size is not the same as representativeness: a huge but skewed sample (for example, only online respondents) is less trustworthy than a smaller, well-structured one. Representativeness means the sample mirrors the population in the characteristics that matter for the study, which is what licenses generalisation.

Worked example

Try this

Q1. What is the difference between a population and a sampling frame? [2 marks]

  • Cue. The population is everyone the study is about; the sampling frame is the actual list from which the sample is drawn.

Q2. Why can findings from a convenience sample not be generalised to the whole population? [2 marks]

  • Cue. Selection is not random and the chance of selection is unknown, so the sample may not represent the population and sampling error cannot be estimated.

Exam-style practice questions

Practice questions written in the style of SQA exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.

SQA AH (research methods)12 marksEvaluate the usefulness of stratified sampling for a study of political attitudes across different age groups.
Show worked answer →

A strong answer weighs the strengths and limitations of stratified sampling against the specific aim, rather than describing it in the abstract.

Strengths: stratified sampling divides the population into strata (here, age bands) and samples each in proportion, so every age group is represented in the right share and comparisons between groups are valid. It reduces the risk that a key subgroup is missed by chance, which a simple random sample can do. Limitations: it requires an accurate sampling frame that already records age, which may not exist or may be out of date; it is more complex and costly to organise; and within each stratum selection must still be random or the benefit is lost. The judgement should conclude that for comparing attitudes across age groups stratified sampling is well suited, provided a reliable frame exists, but is harder to run than a simple random sample.

SQA AH (research methods)8 marksExplain the difference between probability and non-probability sampling and give one example of each.
Show worked answer →

The marks reward a clear distinction plus correctly classified examples.

In probability sampling every member of the population has a known, non-zero chance of selection, which supports generalising to the whole population and estimating sampling error; an example is simple random sampling. In non-probability sampling the chance of selection is unknown and selection is not random, so findings cannot be generalised with statistical confidence but the method is cheaper and useful for hard-to-reach groups; an example is snowball sampling, where participants recruit others. A full answer names the defining feature of each (known versus unknown probability of selection) and gives one valid example for each.

Related dot points

Sources & how we know this