Health issues: Students should understand the principles of sampling as applied to scientific data, including epidemiological data.

The idea

Sampling means studying a manageable subset of a larger group to draw conclusions about the whole. The population is everyone of interest; the sample is the part actually measured. A representative sample must reflect the population's mix, so random sampling avoids bias: number every individual, then use a random number generator to pick participants. Sample size matters: larger samples reduce the effect of chance variation, giving more reliable estimates. In epidemiology, scientists compare disease patterns between groups, for example surveying 1000 adults to estimate smoking rates, then linking smoking to lung cancer incidence. Correlation does not prove causation; other variables such as age, diet or occupation may confound results.

What earns the marks

• Define the population as the whole group being studied and the sample as the subset selected for measurement.
• Explain that a representative sample must reflect the variation within the population, including relevant subgroups.
• Describe random sampling as giving every member of the population an equal chance of selection, reducing bias.
• State that increasing sample size reduces the influence of chance and improves the reliability of conclusions.
• Apply sampling ideas to epidemiological data, such as comparing disease frequency between exposed and unexposed groups.
• Recognise that correlation between a factor and a disease does not establish causation because confounding variables may exist.

Common mistakes and how to put them right

• Confusing the sample with the population.

  Put it right: The sample is the measured subset, while the population is the whole group about which conclusions are drawn.

• Assuming a bigger sample automatically removes bias.

  Put it right: A large biased sample, such as volunteers from one gym, is still unrepresentative.

• Treating a correlation as proof of cause.

  Put it right: An observed link may result from confounding variables, so further evidence is needed.

Exam tips

• Name the population, the sample and the sampling method explicitly when describing a study.
• Use the phrase 'representative of the population' and justify it with a feature of the sampling method.
• When interpreting epidemiological data, comment on sample size, possible bias and confounding variables before concluding.