SO23: Research Design & Ethics

Research Design & Ethics

Planning research: sampling, reliability, validity, operationalisation, and ethical principles including BSA guidelines.

📋 Key Vocabulary and Concepts

For Research Design & Ethics, you must know:

• Sampling
• Random sampling
• Stratified sampling
• Opportunity sampling
• Reliability
• Validity
• Ecological validity
• Operationalisation
• Informed consent
• Confidentiality
• Right to withdraw
• Deception
• BSA guidelines

❓ Practice Questions

✅ Answers

1. Random: most representative but hard. Stratified: ensures diversity. Opportunity: quick but biased. Choice depends on: resources, need for representativeness, access.
2. Reliability = consistency (repeat, same results). Validity = accuracy (measures what it claims). Both needed: reliable but invalid = consistently wrong; valid but unreliable = accidentally right.
3. Ethical issues: informed consent (victims may be too distressed), confidentiality (identifying abusers risks victims), protection from harm (discussing abuse may re-traumatise), right to withdraw, sensitivity.

🎯 Exam Tips

• Sampling: random (unbiased, hard) vs opportunity (easy, biased).
• Reliability = CONSISTENCY. Validity = ACCURACY. Both needed.
• BSA Ethics: consent + confidentiality + protection + withdrawal + avoid deception + debrief.

📝 Exam Technique

⚠️ Common Errors

✍️ Model Answer

📊 AO Deep Dive

📖 Detailed Notes: Research Design and Ethics

💡 Planning Sociological Research

Research design is the overall plan for a piece of sociological research. It begins with choosing a research topic and formulating a specific research question or hypothesis. A hypothesis is a testable prediction, for example: ‘Working-class students are more likely to be excluded from school than middle-class students.’ Positivist sociologists prefer hypothesis-led research because it produces clear, testable findings that can be used to build scientific theories. Interpretivist sociologists often prefer more open-ended research questions that allow unexpected findings to emerge, for example: ‘How do working-class students experience school discipline?’ The choice of research method follows from the research question: questionnaires and experiments suit hypothesis testing, while interviews and observations suit exploration of meanings and experiences.

Sampling is a crucial part of research design because sociologists cannot study entire populations. A representative sample allows findings to be generalised. Random sampling gives every member of the target population an equal chance of selection, producing the most representative sample. Stratified random sampling divides the population into subgroups (e.g. by age, gender, ethnicity) and then randomly selects from each, ensuring all groups are represented. Systematic sampling selects every nth person from a list. Quota sampling ensures the sample reflects the population’s characteristics but the researcher chooses who to include, introducing bias. Opportunity sampling uses whoever is conveniently available and is the least representative. For qualitative research, purposive sampling deliberately selects participants who have specific experiences relevant to the research question, and snowball sampling uses existing participants to recruit others, which is useful for reaching hidden populations.

🔍 Ethical Issues in Sociological Research

Ethical considerations are fundamental to sociological research. The British Sociological Association (BSA) publishes ethical guidelines that researchers must follow. The principle of informed consent means participants must understand what the research involves and voluntarily agree to take part. They should be told the purpose of the research, what they will be asked to do, and how their data will be used. Deception — misleading participants about the true purpose or nature of the research — violates informed consent and is only acceptable if the research question is socially important and cannot be addressed any other way. Even then, participants must be debriefed afterwards and given the right to withdraw their data.

Confidentiality means that participants’ identities must be protected; data should be anonymised so that individuals cannot be identified from the research report. This is especially important when researching sensitive topics such as domestic abuse or illegal activity. The right to withdraw means participants can leave the research at any time without giving a reason and without penalty. Protection from harm means researchers must not cause physical or psychological harm, including stress, embarrassment or anxiety. Vulnerable groups — children, people with learning difficulties, elderly people — require special safeguards, and research involving children under 16 requires parental consent. Covert research, where participants do not know they are being studied, is only justified in exceptional circumstances where the research question is of significant social importance and no other method is feasible.

⚖️ Theoretical Issues: Validity, Reliability and Representativeness

Validity refers to whether a research method genuinely measures what it claims to measure. A valid measure of social class would accurately reflect an individual’s economic and social position. Questionnaires may lack validity if respondents misunderstand questions or give socially desirable answers. Interviews may have greater validity because the researcher can clarify questions and probe responses, but the interviewer’s presence may also distort answers. Ecological validity refers to whether findings reflect real-life situations — observational methods generally have high ecological validity because they study people in their natural settings, while laboratory experiments typically have low ecological validity because the artificial environment changes behaviour.

Reliability means that if the research were repeated using the same method, the same results would be obtained. Questionnaires and structured interviews are highly reliable because they use standardised questions and procedures that can be replicated exactly. Unstructured interviews and observations are less reliable because each interview or observation is unique — different researchers may produce different findings even when studying the same phenomenon. Representativeness means the sample accurately reflects the characteristics of the target population, allowing generalisation. Large random samples are the most representative; small purposive samples are the least. The tension between validity, reliability and representativeness reflects the deeper divide between positivist and interpretivist approaches: positivist methods prioritise reliability and representativeness, while interpretivist methods prioritise validity and depth.

📊 Ethical Principles in Sociological Research

❓ Additional Practice Questions

✅ Additional Answers

1. Ethical guidelines protect participants from harm, exploitation and loss of privacy. The BSA guidelines ensure researchers respect participants’ autonomy through informed consent, protect their wellbeing through the principle of no harm, and maintain their dignity through confidentiality. Covert research violates informed consent because participants do not know they are being studied. However, some sociologists argue that covert research can be justified when the research question is socially important and no other method can access the group. For example, studying racist organisations or criminal gangs may require covert methods because members would not consent to being studied. A balanced view is that covert research should be a last resort, the potential social benefits must clearly outweigh the ethical costs, and participants should be debriefed where possible.
2. An appropriate strategy would use semi-structured interviews with a purposive sample of 20–30 homeless people, recruited through homeless shelters and support services. Snowball sampling could help reach rough sleepers who do not use services. Semi-structured interviews balance the need for comparability across participants with the flexibility to explore individual experiences in depth. Ethical safeguards include: obtaining informed consent with clear information sheets; ensuring confidentiality by using pseudonyms; being sensitive to the vulnerability of participants (many have mental health issues or addiction problems); offering participants the right to withdraw without penalty; and providing information about support services. The researcher should also consider whether offering payment or food could be seen as coercive given participants’ vulnerability.

📝 Exam Questions by Topic

• AQA Past Papers - Topic Questions -- AQA
• Edexcel Past Papers - Topic Questions -- Edexcel
• OCR Past Papers - Topic Questions -- OCR
• Eduqas Past Papers - Topic Questions -- Eduqas
• CCEA Past Papers - Topic Questions -- CCEA

🎬 Video Resources

• BBC Bitesize: GCSE Sociology
• Research Design (YouTube)
• Ethics in Sociology (YouTube)