Assess sample representativeness
Judge who was included, who was missed and whether a sample can support the stated population claim.
Section 28-5.1.2 · 2028–2030 · 5.1.2 · Assess samples and selection
Original StudyForA scenarios and formative model answers; no Cambridge marks, predicted grades or certified mastery. Alternative reasoned answers may be valid. Full-length unseen papers, a wider mixed-problem bank and calibrated specialist teacher marking remain outstanding.
Before you begin
- Read an English passage closely
- Distinguish a statement from its supporting reason
Work on paper. Try each question before revealing help, and explain why your method works.
Quick readiness check
Before evaluating support, what should you identify?
What you will learn
- Assess sample size, characteristics and selection
- Match a conclusion to the population actually observed
Start with the claimed population
A survey of library users can describe those respondents, but does not automatically describe all residents. Compare relevant characteristics such as access, age, working hours or language needs. A large sample can still be systematically biased if some groups cannot participate.
Fictional survey: 80 of 100 online respondents favour a digital-only booking system. It was shared only on an existing app. This is 80% of respondents, not proof that 80% of all users favour it; people unable to use the app were poorly represented.
Size and selection solve different problems
Small samples may be unstable, but enlarging the same selective sample does not remove its exclusion mechanism. Voluntary responses can overrepresent strong opinions. Incentives can affect who participates and what they say. Random selection helps only if the sampling frame covers the target population and response patterns are considered.
Suggest an improvement linked to the limitation: offer accessible offline participation, sample across relevant groups, record non-response and report the denominator. Avoid demanding every person be surveyed; representative sampling can support cautious inference.
Sources for this lesson
Pause and explain
Does adding more app-only respondents remove exclusion of people unable to use the app?
Worked example
Assess the app survey’s claim about all booking users.
Show the worked solution
- Compute 80/100=80% of respondents.
- Identify app-only recruitment.
- Explain why digitally excluded users may respond differently.
Answer The survey supports a statement about these respondents, but its app-only selection weakens the all-user generalisation. Recruiting offline users and assessing non-response would address this limitation more directly than adding only app respondents.
Try a topic test
Choose Easy, Normal or Difficult. Save your answers and open worked review after finishing.
Test this topic →From guided practice to a new situation
Write your answer and reasoning first. Use a hint only if you are stuck. The model solution is for self-checking; your written work is not automatically marked.
Guided: State the correct denominator for the 80% result.
Give me a hint
Use the group actually surveyed.
Compare my reasoning
- Eighty favourable responses were recorded.
- There were one hundred respondents.
- Do not replace respondents with all residents.
The denominator is the 100 respondents, so 80% of that responding sample favoured the proposal. No all-resident percentage is established by these data alone.
Look for these in your work
- Uses the recorded population
- Calculates the proportion
- Limits the claim
Independent: A daytime station survey concludes that commuters oppose late-night trains. Identify a selection weakness and an improvement.
Give me a hint
Who is absent during the recruitment period?
Compare my reasoning
- Late-night users may be absent by day.
- Their needs may differ.
- Sample across travel periods and record relevant groups.
Daytime recruitment may underrepresent late-night travellers, who are particularly relevant to the proposal. Surveying across daytime and evening periods would address that specific selection weakness; simply doubling daytime interviews would not.
Look for these in your work
- Identifies the missed group
- Explains relevance
- Proposes a targeted remedy
Application: A random list covers all students, but only 35% respond and responses mostly come from one year group. Assess the final sample.
Give me a hint
Separate initial random selection from actual response.
Compare my reasoning
- The initial frame may be broad.
- Differential non-response changes the final sample.
- Check whether year group affects the opinion being measured.
The random invitation list is a strength, but the final sample may be unrepresentative because non-response concentrates answers in one year group. The judgement depends on whether this characteristic affects the question; response follow-up and group comparisons are needed.
Look for these in your work
- Recognises the initial strength
- Assesses non-response
- Links characteristics to the claim
Common mistakes
- Equating a large sample with representativeness
- Changing the population in the conclusion
- Ignoring non-response
What can you explain now?
Does adding more app-only respondents remove exclusion of people unable to use the app?
Compare with the explanation
No, the selection mechanism still excludes them
The remedy must address how participants enter the sample.
After trying it yourself, choose your next review. This is your self-assessment.
Your review choice appears on Today. Sign in to sync it across devices.
Make it stick
Does adding more app-only respondents remove exclusion of people unable to use the app?
If you needed the solution, close it and solve the problem again from a blank page. A correct answer is stronger when you can explain the reason for each step.