Revision Library

revisionlibrary.co.uk

GCSE · Sociological research methods

GSOC.RM11 Reliability, Validity and Representativeness in Sociological Research

AQA 8192 · Calculators not allowed · about 60 minutes

Total Marks

Name: _______________________________ Date: ____ / ____ / ______

Answer ALL questions. Write full sentences for questions worth 3 marks or more. Spend about 60 minutes on the whole pack.

Which one of the following best describes the concept of reliability in sociological research methods, as used in GCSE sociology?

A) Whether the research method gives a deep and true picture of social life
B) Whether repeating the research using the same method would produce similar results
C) Whether the sample reflects the whole population so findings can be generalised
D) Whether the research follows ethical rules about consent

(Total for Question 1 is 1 mark)

Define what sociologists mean by 'validity' in the context of research methods used to study people and social behaviour.

(Total for Question 2 is 2 marks)

Define what sociologists mean by 'representativeness' when evaluating a sample used in social research.

(Total for Question 3 is 2 marks)

Explain one reason why structured questionnaires, used in a large school survey of pupil attitudes to homework, tend to score highly on reliability.

(Total for Question 4 is 4 marks)

Explain one reason why structured questionnaires, used to study attitudes to homework across many schools, may perform poorly on validity.

(Total for Question 5 is 4 marks)

Explain one reason why unstructured interviews, used to research young people's reasons for truancy, tend to score highly on validity.

(Total for Question 6 is 4 marks)

Explain one reason why unstructured interviews may score poorly on reliability when used in several different schools.

(Total for Question 7 is 2 marks)

Item B: A researcher sends a structured questionnaire to 500 pupils across ten schools chosen by the researcher's contacts, and receives 320 responses. The researcher reports the findings as if they represent all pupils in the city.

(a)Using Item B, identify two features of the sample or method that might make the sample unrepresentative of all pupils in the city.(2)

(b)Using Item B, explain one way the researcher could improve representativeness when sampling pupils in the city.(2)

(Total for Question 8 is 4 marks)

Explain one strength of using a large, random sample in survey research when the main concern is representativeness.

(Total for Question 9 is 4 marks)

Explain one practical or ethical problem that can reduce representativeness when using snowball sampling to reach hidden groups, for example young homeless people.

(Total for Question 10 is 2 marks)

State one way reliability can be improved when using structured interviews in research with school staff.

(Total for Question 11 is 2 marks)

State one reason why qualitative participant observation may be low in representativeness.

(Total for Question 12 is 2 marks)

Explain two differences between reliability and validity when evaluating sociological research methods.

(Total for Question 13 is 4 marks)

Explain one reason why a method chosen mainly for high reliability might be criticised by sociologists who prioritise validity.

(Total for Question 14 is 4 marks)

Discuss whether reliability or validity matters more to sociologists when evaluating research methods. In your answer you should explain both concepts, give arguments supporting the importance of reliability, give arguments supporting the importance of validity, and reach a reasoned conclusion about which matters more in typical sociological research.

Discuss whether reliability or validity matters more to sociologists when evaluating research methods. Use examples and reach a reasoned conclusion.

(Total for Question 15 is 8 marks)

Mark scheme · GSOC.RM11 Reliability, Validity and Representativeness in Sociological Research

Question 1

B1 B (whether repeating the research using the same method would produce similar results)
Answer: B

Question 2

B1 identifies that validity is about whether a method measures what it claims to measure or gives a true picture (oe)
B1 development, e.g. valid methods capture meanings, motives and in-depth understanding rather than only surface features
Answer: Validity is about whether a research method measures what it claims to measure and gives a true, in-depth picture; valid methods capture meanings, motives and depth rather than only surface features.

Question 3

B1 identifies that representativeness is how well a sample reflects the wider population it is drawn from (oe)
B1 development, e.g. a representative sample allows researchers to generalise findings to the wider population
Answer: Representativeness is how well a sample reflects the wider population it is drawn from; a representative sample allows researchers to generalise the findings to that wider population.

Question 4

B1 identifies a reason, e.g. questions are standardised and asked in the same way to every respondent
B1 development, e.g. standardisation reduces variation caused by different interviewers or different phrasing
B1 further development, e.g. this makes it likely that repeating the survey under similar conditions will produce similar results
B1 clear linking to reliability rather than validity or representativeness
Answer: Structured questionnaires are standardised so every pupil receives the same questions in the same order; this reduces variation from different interviewers or question wording and makes it likely that repeating the survey under similar conditions would produce similar results, so reliability is high.

Question 5

B1 identifies a reason, e.g. closed response options may not capture a respondent's true feelings or the complexity of attitudes
B1 development, e.g. respondents may choose the nearest available option or give socially desirable answers rather than express their real view
B1 further development, e.g. this reduces depth and the truthfulness of data, so questionnaires can lack validity
B1 clear link to validity rather than reliability or representativeness
Answer: Structured questionnaires with fixed response options can fail to capture respondents' true feelings or the complexity of their attitudes; pupils may pick the nearest option or give socially desirable answers, reducing depth and truthfulness and so lowering validity.

Question 6

B1 identifies a reason, e.g. unstructured interviews allow respondents to speak at length and explain meanings, motives and context
B1 development, e.g. the researcher can probe and follow up, producing rich, detailed qualitative data
B1 further development, e.g. this in-depth material gives a more accurate, true picture of why people behave as they do, increasing validity
B1 clear linking to validity rather than reliability or representativeness
Answer: Unstructured interviews let respondents talk at length and let the researcher probe for meanings and motives; the resulting rich, detailed qualitative data gives a truer, more in-depth picture of behaviour and therefore higher validity.

Question 7

B1 identifies a reason, e.g. interviews are not standardised so different interviewers may ask different questions or probe different issues
B1 brief development, e.g. this variation makes it unlikely the same results would be produced if the study was repeated
Answer: Unstructured interviews are not standardised, so different interviewers may ask different questions or probe different issues; this variation means repeating the study is unlikely to produce the same results, lowering reliability.

Question 8

(a) B1 a correct feature, e.g. schools were chosen by the researcher's contacts (convenience sampling) rather than randomly
(a) B1 a second correct feature, e.g. only 320 of 500 questionnaires were returned, so non-response could bias the sample
(a) Answer: The schools were chosen by the researcher's contacts rather than randomly (convenience sampling), and only 320 of 500 questionnaires were returned, so non-response bias may make the sample unrepresentative.
(b) B1 identifies an improvement, e.g. use random sampling or stratified sampling across all city schools
(b) B1 development, e.g. this would ensure the sample more closely matches the wider pupil population and reduce selection bias
(b) Answer: The researcher could use random or stratified sampling across all city schools, ensuring the sample better matches the wider pupil population and reducing selection bias.

Question 9

B1 identifies a strength, e.g. random sampling reduces selection bias and increases the chance the sample mirrors the population
B1 development, e.g. larger samples also reduce the impact of random variation, making estimates more stable
B1 further development, e.g. this allows safer generalisation from sample to population and supports external validity of quantitative results
B1 clear linking to representativeness and generalisation
Answer: Random sampling reduces selection bias and a large sample reduces random variation, so together they increase the chance the sample mirrors the population and allow safer generalisation from sample to population, improving representativeness.

Question 10

B1 identifies a problem, e.g. snowball sampling often reaches connected networks so the sample may be biased towards similar people
B1 brief development, e.g. this clustering limits how well the sample represents the broader hidden population
Answer: Snowball sampling tends to recruit people through existing networks, so the sample may be biased towards similar individuals and not represent the wider hidden population, limiting representativeness.

Question 11

B1 identifies a way, e.g. train interviewers and use a standardised interview schedule
B1 brief development, e.g. training and standard questions reduce interviewer variation and improve repeatability
Answer: Train interviewers and use a standardised interview schedule so all staff are asked the same questions in the same way, reducing interviewer variation and improving reliability.

Question 12

B1 identifies a reason, e.g. participant observation usually studies a small group or setting in depth
B1 brief development, e.g. the small, non-random sample makes it hard to generalise to a wider population
Answer: Participant observation typically focuses on a small group or setting in depth, so the small, non-random sample makes it difficult to generalise findings to a wider population, lowering representativeness.

Question 13

B1 identifies first difference, e.g. reliability is about consistency and repeatability
B1 develops first difference, e.g. a reliable method produces similar results if repeated
B1 identifies second difference, e.g. validity is about accuracy and whether the method measures what it claims
B1 develops second difference, e.g. a method can be reliable but not valid if it consistently measures the wrong thing
Answer: Reliability concerns consistency and repeatability, meaning a reliable method produces similar results if repeated; validity concerns accuracy and whether the method measures what it claims, so a method can be reliable but not valid if it consistently measures the wrong thing.

Question 14

B1 identifies a reason, e.g. highly standardised methods like closed questionnaires may miss meanings and motives
B1 development, e.g. critics prioritising validity argue social behaviour cannot be understood fully through superficial, repeatable measures
B1 further development, e.g. this can lead to misleading conclusions about attitudes or motives even if the data are consistent
B1 links the criticism to sociological concerns about understanding social reality rather than just producing neat, repeatable numbers
Answer: Methods chosen for high reliability, such as closed questionnaires, may miss meanings and motives, so sociologists who prioritise validity criticise them for producing superficial data; consistent but shallow measures can produce misleading conclusions about social behaviour despite being repeatable.

Question 15

Level 1 (1-3): Basic statements about reliability and validity with limited development, or a one-sided answer with little or no example and no clear conclusion.
Level 2 (4-6): Clear explanation of reliability and validity, with developed arguments for why each might be important, some use of examples, and a simple supported judgement or partial conclusion.
Level 3 (7-8): Well-developed, balanced discussion of both reliability and validity, uses appropriate examples (for example questionnaires versus unstructured interviews), shows understanding of trade-offs and practical issues, and reaches a clear, justified conclusion about which matters more in typical sociological research.
Indicative content:
- Define reliability: consistency and repeatability of results; illustrate with structured questionnaires and standardised interviews.
- Define validity: truthfulness and depth of understanding; illustrate with unstructured interviews and participant observation.
- Arguments for reliability mattering more: enables comparisons across time/groups, supports quantitative generalisation, easier to test and check, important for policy-related research needing stable measures.
- Arguments for validity mattering more: sociology studies meanings and motives, so understanding social reality may require valid, in-depth methods even if less repeatable; qualitative insight can explain why patterns exist.
- Trade-offs and practicalities: reliable methods may sacrifice depth, valid methods may be time-consuming and unrepresentative; mixed-methods approaches can combine strengths.
- Examples: school attitude surveys (reliability useful for tracking change) versus life-history interviews (validity essential to understand complex life choices).
- Reasoned conclusion: balanced answers might argue validity is crucial for core sociological understanding, but reliability is essential for measurement and comparison; the best answer judges which matters more depending on the research aim and may recommend a compromise or mixed methods.

Reliability, Validity and Representativeness in Sociological Research - Worksheets, Questions and Revision