Social Research and Applied Sociology: Methods, Sampling and Ethics
What this domain tests
This domain covers how sociological knowledge is produced and used: the steps of research, quantitative versus qualitative designs, methods and tools, sampling, ethics, and the applied uses of sociology in policy, industry and social work.
Core rules and concepts
- Steps: choosing the problem, reviewing literature, framing the hypothesis, designing the study, sampling, collecting data, analysing, and reporting.
- Types: pure versus applied; descriptive, exploratory and explanatory; quantitative versus qualitative.
- Methods: survey, observation (participant and non-participant), interview (structured and unstructured), questionnaire, case study, experiment, content analysis, ethnography, focus group.
- Tools: a schedule is filled in by the enumerator, a questionnaire by the respondent; the Likert scale measures attitudes; sociometry (Moreno) maps group relations.
- Sampling: probability designs - simple random, systematic, stratified, cluster - permit generalisation; non-probability designs - purposive, quota, convenience, snowball - do not.
- Ethics: informed consent, confidentiality, anonymity, no harm to respondents, voluntary withdrawal.
- Applied sociology: policy evaluation, social work, market research, and industry - the Hawthorne studies (1924–32, Elton Mayo) showed that being observed changes behaviour.
Worked example
Question: 'Do mid-day meals improve school attendance?' Reasoning: hypothesis - attendance is higher where meals are served; independent variable - provision of the meal; dependent variable - attendance rate; sample - stratified random selection of schools by district; data - attendance registers (secondary) plus interviews with teachers and parents (primary); comparison with schools lacking the scheme guards against spurious conclusions; parental consent satisfies ethics. Correlation alone would not prove causation.
Traps that cost marks
- Reversing schedule and questionnaire.
- Claiming non-probability samples support statistical generalisation.
- Confusing a sample with the population, or a hypothesis with a conclusion.
- Treating correlation as causation.
- Omitting ethics when the question mentions fieldwork.