Primary and secondary data collection; random, systematic and stratified sampling; sample size and bias; and the planning of safe, ethical fieldwork.

What this dot point is asking

Eduqas wants you to explain primary and secondary data collection, the random, systematic and stratified sampling strategies, the effect of sample size and bias, and the planning of safe, ethical fieldwork.

The answer

Primary and secondary data

Eduqas investigations rest on primary fieldwork data, gathered by the student, supported by secondary data for context and comparison. Methods divide into quantitative (numerical: measurements of pebble size or gradient, pedestrian and traffic counts, environmental quality scores) and qualitative (descriptive: questionnaire responses, perception surveys, interviews, field sketches and photographs). The Changing Places content specifically requires qualitative as well as quantitative evidence, so a human-geography enquiry should gather both. Choosing methods that genuinely answer the question, and justifying them, is what earns the skills marks.

Sampling strategies

You cannot study everything, so you sample. Random sampling minimises bias but may cluster and miss parts of the area. Systematic sampling (a fixed interval along a transect or grid) gives even, predictable coverage and is ideal where you are testing change along a gradient (distance along a beach, a transect from city centre to suburb). Stratified sampling ensures representation of recognised subgroups, for example sampling different land-use zones or age groups in proportion to their share of the population. The strategy must be explained, not arbitrary: examiners reward a method matched to the question with its strengths and weaknesses acknowledged.

Sample size, bias and safe fieldwork

A larger, well-designed sample size reduces the influence of anomalies and chance, makes the data more representative, and gives more confidence in any statistical test, though it costs more time, a trade-off to justify. Bias arises from a poorly chosen sampling method, leading questions in a questionnaire, or sampling only convenient locations, and it undermines reliability and the validity of conclusions. All fieldwork must be planned for safety and ethics: a risk assessment identifies hazards (tides and currents on a coast, river flow, traffic, weather, lone working), and ethical practice means informed consent for questionnaires, respect for privacy, and minimal environmental disturbance. Planning, justifying and safeguarding the data collection is a core part of the marked enquiry.

Examples in context

Example 1. Systematic sampling on a beach transect. A coastal investigation testing whether pebble size changes along a beach uses systematic sampling: measuring the long axis of, say, ten pebbles at a fixed interval (every $50$ m) along a transect in the direction of longshore drift. Systematic sampling is justified because it gives even coverage of the gradient being tested, and an adequate number of sites and pebbles per site reduces anomalies. A risk assessment addresses tides, slippery surfaces and lone working. This is the classic Eduqas physical-geography sampling design and feeds directly into a Spearman's rank analysis.

Example 2. Stratified sampling for a place study. A human-geography investigation comparing deprivation or environmental quality across a town uses stratified sampling: dividing the town into recognised districts or land-use zones and sampling each in proportion to its size, so no area dominates the result. It combines primary data (environmental quality surveys, questionnaires with informed consent) and secondary data (the census, indices of deprivation), meeting the Changing Places requirement for both quantitative and qualitative evidence. Stratified sampling is justified because it ensures all parts of the town are represented, the standard Eduqas human-geography sampling design.

Try this

Q1. Define stratified sampling. [2 marks]

• Cue. Dividing the population into subgroups (strata) and sampling each in proportion, so the sample represents the whole population.

Q2. Explain why a justified sampling strategy improves an investigation's reliability. [3 marks]

• Cue. A sampling strategy matched to the question reduces bias and makes the data representative of the population, so the results and any statistical test are more reliable and the conclusions more defensible than from an arbitrary or convenience sample.