A sample is the group actually studied; the target population is the wider group the sample should represent. AQA names four sampling methods, each with strengths and weaknesses for representativeness.
Random sampling
Every member of the target population has an equal chance of selection (e.g. drawing names from a hat, computer-generated list of all employees).
Strength: unbiased — no systematic preference for any subgroup, so likely to be representative if the population list is complete. Weakness: requires a complete list (often impossible at population scale); even random samples can be unrepresentative by chance, especially with small N.
Opportunity sampling
Select whoever is available and willing at the time (passers-by, your psychology class).
Strength: quick, cheap, practical for student projects. Weakness: highly biased — skewed by who happens to be there (e.g. only daytime shoppers, only Year 12s).
Systematic sampling
Use a fixed system to choose every nth person from a list (every 10th name on the school roll, every 5th customer through the door).
Strength: structured and easy to apply; gives wider coverage than opportunity sampling. Weakness: bias creeps in if the list itself is patterned (e.g. every 10th student happens to be a senior); not strictly random.
Stratified sampling
Split the population into strata (subgroups by relevant variable — age, gender, year group), and sample randomly within each stratum in proportion to its size in the population.
Strength: highest representativeness — every relevant subgroup is included in the right proportion. Weakness: requires you to know the population breakdown; more time-consuming.
Practical tips for a 4-mark question
A strong answer:
- Names the method.
- Describes how it would be carried out in the given scenario.
- Gives one strength and one weakness in that scenario.
E.g. "I would use stratified sampling because the school has 60% girls and 40% boys; sampling 6 girls and 4 boys ensures gender balance, addressing potential gender differences in the DV."
AI-generated · claude-opus-4-7 · v3-deep-psychology