Research Methods and Techniques: Observation, Interview, Questionnaire, Survey, Case Study
What this domain tests
How sociological knowledge is produced - the research process, sampling, tools of data collection - and the justification for choosing one method over another.
Core content
- Research process: define the problem, review literature, state a hypothesis, choose a design, select a sample, collect data, analyse, report. Key terms: variable (independent/dependent), concept, reliability (consistency) and validity (measuring what is intended), primary vs secondary data.
- Sampling: census vs sample; probability methods (simple random, systematic, stratified, cluster) and non-probability methods (purposive, quota, snowball, convenience).
- Survey: collection of standardised information from a large number of cases, usually by questionnaire or interview schedule.
- Observation: controlled/uncontrolled, participant/non-participant, overt/covert. Participant observation gives depth but risks the observer's bias (Whyte's Street Corner Society).
- Interview: structured, semi-structured, unstructured, focus group. A schedule is filled by the interviewer; a questionnaire is filled by the respondent.
- Questionnaire: open vs closed items, mailed or distributed, low response rate, no probing.
- Case study: intensive study of one case - a person, group, community or institution - for depth rather than generalisation (life history, single-case design).
- Ethics: informed consent, confidentiality, no harm, honesty about the researcher's identity. Scales: Bogardus social distance, Likert, Guttman.
Worked example
"Why do pupils drop out from a government high school in a district town?" A survey would measure dropout rates and correlate them with income and parental education, giving breadth. An unstructured interview with dropouts and teachers, or a case study of one school, gives depth and meanings but cannot be generalised. Reasoning: match the method to the research question, state the sampling procedure and the known limitation; this earns the reasoning marks.
Traps
Swapping questionnaire with schedule; treating a sample as the population; confusing reliability with validity; claiming correlation proves causation; ignoring respondent consent.