An interview in social research is a systematic method of collecting information by asking people questions about their experiences, beliefs, attitudes, knowledge, behaviour, identities, relationships, circumstances or interpretations of social life. Unlike an ordinary conversation, a research interview has an explicit research purpose, a defined population or participant group, a planned topic or question structure, procedures for recording and analysing responses, and ethical requirements concerning consent, confidentiality and the treatment of participants.
Interviews can generate information about both observable circumstances and subjective meanings that may not be directly accessible through observation or administrative records. A researcher might interview workers about employment conditions, parents about childcare, migrants about experiences of settlement, students about educational experiences, patients about healthcare, or community members about local change. The central strength of the method is that it allows participants to explain their own perspectives in their own words while enabling the researcher to probe meanings, clarify ambiguous statements and explore processes in considerable depth.
Research interviews vary according to how tightly the questions and responses are structured. A structured interview uses a predetermined sequence of questions and usually standardised wording and response categories, making answers easier to compare across participants and often producing data that can be coded quantitatively. A semi structured interview uses an interview guide containing planned themes and questions while allowing the interviewer to change the sequence, ask follow up questions and explore unexpected issues.
An unstructured interview is much more open ended and may resemble an extended guided conversation in which the participant has considerable influence over the direction of discussion. In qualitative sociology, semi structured and unstructured interviews are particularly valuable when the researcher wants to understand meanings, identities, biographies, experiences and social processes rather than simply measure predetermined variables. The choice of interview structure therefore depends on the research question, the required comparability of responses, the sensitivity of the topic, the characteristics of the participants and the analytical approach being used.
Interview questions can be open ended, closed ended, factual, attitudinal, behavioural, hypothetical or interpretive. Open ended questions invite participants to produce extended answers, whereas closed questions restrict responses to predefined alternatives. Good interview design avoids leading questions that implicitly encourage a particular answer, double barrelled questions that ask about two different issues simultaneously, confusing technical terminology and unnecessarily complicated wording.
Interviewers may use probes such as requests for examples, explanations or clarification, but probing must be sufficiently neutral to avoid manufacturing the participant's response. The sequence of questions also matters because early questions can establish rapport and later questions can be influenced by what has already been discussed. Sensitive questions may require careful placement, explanation and opportunities for participants to decline. Interview schedules are commonly piloted before full data collection so researchers can identify unclear questions, excessive length, problematic assumptions and missing topics.
The interviewer is not a neutral recording device, because the interaction between interviewer and participant can influence the data produced. Characteristics such as age, gender, ethnicity, social class, professional status, perceived authority, communication style and shared or different experiences can affect what participants are willing to disclose and how they interpret questions. This is often discussed through concepts such as interviewer effects, positionality, reflexivity and rapport.
Rapport refers to the development of sufficient trust and communicative comfort for meaningful discussion, while reflexivity requires researchers to consider how their own assumptions, identity and behaviour may have shaped the research encounter. Interviewers therefore need training in listening, probing, neutrality, managing sensitive disclosures, recognising ambiguity and avoiding unnecessary judgement. In qualitative research, the interview is often understood as an interaction through which accounts are produced rather than as a simple mechanism for extracting pre existing facts from participants.
Sampling determines whose experiences are represented in an interview study. Researchers may use probability sampling when the objective is statistical representativeness, although many qualitative interview studies instead use purposive sampling to recruit participants with particular experiences or characteristics relevant to the research question. Other approaches include theoretical sampling, quota sampling, snowball sampling and convenience sampling.
Sample size is not determined by a universal rule: qualitative researchers may collect interviews until additional interviews provide limited new conceptual information, often described using ideas such as saturation, although contemporary researchers distinguish different forms of saturation and question whether it can always be established mechanically. Recruitment procedures can substantially affect findings because people who volunteer for research may differ systematically from people who decline. Researchers therefore need to describe inclusion criteria, recruitment routes and relevant characteristics of participants so that readers can understand the scope and limitations of the evidence.
Ethical considerations are central because interviews involve people and can elicit private, sensitive or potentially harmful information. Participants normally need sufficient information to provide informed consent, including an explanation of the study's purpose, what participation involves, how recordings or transcripts will be handled, whether participation is voluntary and whether withdrawal is possible under the stated conditions. Confidentiality concerns how researchers protect information from unauthorised disclosure, while anonymity concerns whether an individual's identity can be concealed.
Complete anonymity can be difficult when participants discuss distinctive experiences or when researchers retain identifiable contact information. Data protection also applies to audio recordings, transcripts, demographic details and research notes. Sensitive interviews may involve risks involving trauma, stigma, employment, family relationships or legal circumstances, requiring appropriate safeguarding and referral procedures where relevant. Ethical review helps researchers identify and manage these risks before fieldwork begins.
Interview data can be analysed quantitatively or qualitatively depending on the design. Structured interviews can generate datasets in which answers are coded as variables and analysed using descriptive statistics or inferential methods. Qualitative interviews generally involve transcription followed by processes such as coding, thematic analysis, framework analysis, grounded theory analysis, narrative analysis or discourse analysis. Researchers may examine recurring themes, contrasting accounts, sequences of experience, categories of meaning and relationships between personal narratives and wider social structures.
Transcription itself involves analytical decisions concerning pauses, emphasis, interruptions, dialect and non verbal features. Researchers must distinguish participants' accounts from the researcher's interpretations and provide enough methodological information for readers to evaluate how conclusions were reached. Computer assisted qualitative data analysis software can organise large collections of coded material, but software does not independently determine the meaning of social data.
Interviews have major advantages but also important limitations. They can provide rich contextual information, reveal meanings and experiences that surveys cannot easily capture, permit clarification and probing, and allow researchers to study complex or sensitive processes. They can also be relatively adaptable when unexpected issues arise. However, interviews can be expensive and time consuming, interviewer effects can influence responses, participants may forget or misrepresent events, socially desirable answers may occur, and retrospective accounts may reconstruct rather than reproduce past experiences.
The presence of an interviewer can alter behaviour, while recruitment and non response can produce selection bias. Interviews therefore work particularly well when matched carefully to the research question and combined, where appropriate, with other evidence such as observation, documents, surveys or administrative data. Their value lies not simply in producing quotations but in systematically interpreting accounts within their social, institutional and historical contexts.
