Types of Data and Research Methods

Types of Data: Primary, Secondary, Quantitative and Qualitative

Before choosing a research method, sociologists must decide what kind of data they need. Primary data is original data collected directly by the researcher — through questionnaires, interviews or observation. Secondary data is data already gathered by someone else that the researcher re-uses — government statistics, newspaper archives, historical records, diaries and journals, or published research.

Quantitative data takes numerical form — percentages, rates, frequencies, scores. It allows comparisons across large samples and can be analysed statistically. Positivist sociologists, who argue sociology should follow the methods of natural science, prefer quantitative data because it is objective and replicable. Qualitative data takes verbal form — the words, accounts, feelings and interpretations of participants. It provides depth and texture. Interpretivist sociologists, associated with interactionism, prefer qualitative data, arguing that understanding social behaviour requires grasping the subjective meanings of the participants.

Secondary sources include official statistics (produced by the Office for National Statistics, the Home Office, and other government bodies) and non-official statistics (produced by pressure groups, charities or academic bodies). Official statistics have strengths: they are large-scale, nationally consistent, free to access, and allow trends to be tracked over time. However, sociologists must evaluate them critically: crime statistics, for example, only count crimes that are reported to and recorded by the police, so the true level of offending is almost certainly higher than official figures suggest. Personal documents such as diaries and letters can also serve as qualitative secondary sources, providing insight into private experience that no survey could access.

Questionnaires

A questionnaire is a standardised set of written questions distributed to a sample. Closed questions offer a fixed set of responses (e.g. yes/no, tick all that apply, a rating scale) and produce quantitative data. Open questions allow respondents to answer in their own words, producing qualitative data.

Strengths: questionnaires can reach large samples quickly and cheaply. They are standardised — all respondents answer the same questions — which improves reliability. Anonymity may encourage honest answers on sensitive topics. Weaknesses: response rates can be low, introducing bias. Closed questions impose the researcher’s categories on complex experiences. Questionnaires cannot probe or follow up unexpected answers, which reduces validity.

Interviews: Structured, Unstructured and Semi-Structured

A structured interview follows a fixed schedule of questions in the same order and wording for every participant, much like a spoken questionnaire. This produces quantitative data that is easy to compare. The weakness is limited depth — no follow-up is possible.

An unstructured interview begins with an opening topic and allows the conversation to develop naturally. It can reveal unexpected themes and rich qualitative data. The weakness is that it is time-consuming and harder to compare across participants. The interviewer effect — respondents giving socially desirable answers because of the interviewer’s characteristics — is a risk in all interview types. A semi-structured interview uses a list of key themes but allows probing; it combines comparability with depth.

Observation: Participant and Non-Participant

Observation involves directly watching people in their natural social setting. Non-participant observation means the researcher watches without taking part. The presence of an observer may alter participants’ behaviour (the Hawthorne effect).

Participant observation means the researcher joins the group and participates in their daily life, either overtly (with the group’s knowledge) or covertly (without it). Willis’s 1977 study Learning to Labour is a classic example: Willis spent time with twelve working-class boys in their school and community, observing how they actively resisted school culture and produced their own counter-school subculture. This gave qualitative depth that no survey could produce. Participant observation is high in validity — it captures real behaviour in natural settings. Its weaknesses are that it is time-consuming, the researcher may go native (over-identify with the group and lose objectivity), and covert observation raises serious ethical concerns about deception.

Mixed Methods and Triangulation

Rather than choosing a single method, many sociologists use a mixed methods approach. For example, a study of attainment gaps might begin with a large-scale questionnaire to identify patterns, then follow up with in-depth interviews to explore the meanings behind those patterns. This strategy is sometimes called triangulation — using multiple methods so that the weaknesses of one approach are compensated for by the strengths of another.