How we know what we claim to know about society is as important as the claim itself. Sociological research has long been organised around two great traditions — the quantitative, which counts and measures in search of patterns, and the qualitative, which listens and observes in search of meaning. Neither is simply right; each answers different questions and rests on a different image of what social reality is. This note sets out both traditions, the techniques they use, and the standards — sampling, reliability, validity and objectivity — by which their findings are judged.

Quantitative methods

Quantitative research treats social phenomena as measurable. It converts aspects of social life into numbers — rates, frequencies, scores — so that they can be compared, correlated and analysed statistically. Its characteristic tools are the large-scale survey, the structured questionnaire and the analysis of official statistics, and its logic is to identify variables and test relationships between them across many cases.

The tradition draws its philosophy from positivism. If society is governed by law-like regularities, then the researcher's task is to observe them from the outside and state them precisely. Durkheim's use of suicide rates is the classic demonstration — patterns invisible in any single case become clear once thousands are aggregated. The strengths of the approach are breadth, comparability and the ability to generalise from a sample to a population. Its limits are equally clear — numbers can flatten the meanings behind behaviour, and what is easy to measure is not always what matters most.

Qualitative methods

Qualitative research begins from the opposite premise — that social life is made of meanings, and meanings must be understood rather than counted. Its aim is depth, texture and the actor's point of view. The researcher gets close to the people studied, often over long periods, using ethnography, participant observation and unstructured or in-depth interviews to see the world as they see it.

The interpretive roots of this tradition lie in Weber's principle of Verstehen — the idea that explaining social action requires grasping the subjective meanings actors give to it. Where the survey asks how many, the ethnography asks what it means and why. Its strength is richness and validity — a close, faithful account of lived experience. Its limits are the mirror image of the quantitative — small samples make generalisation difficult, and the researcher's own presence and interpretation can shape what is found.

Techniques of data collection

Most methods are combinations of a few basic techniques, each with characteristic strengths and limits.

  • Questionnaire — a set of written questions, often closed-ended, administered to many respondents. It is cheap, fast and easily quantified, but it captures only what its questions anticipate and suffers from low response rates and superficial answers.
  • Interview — structured interviews standardise every question for comparability; unstructured and semi-structured interviews trade that comparability for depth and flexibility, letting respondents raise what matters to them. Interviews risk interviewer bias and the tendency of people to give socially approved answers.
  • Observation — in participant observation the researcher joins the group studied, gaining insight unavailable to outsiders, at the cost of possibly influencing behaviour and of difficulty in recording data systematically. Non-participant observation keeps more distance and objectivity but less understanding.
  • Case study — an intensive study of a single unit, such as a community, organisation or event. It offers holistic depth and is valuable for generating hypotheses, but its findings cannot be safely generalised to other cases.

Sampling

Because researchers rarely study a whole population, they study a sample and infer from it. In probability sampling — random, systematic, stratified or cluster — every unit has a known chance of selection, which allows the researcher to estimate how far the sample can be generalised. In non-probability sampling — quota, snowball or purposive — cases are chosen by convenience or judgement; this is common in qualitative work but weakens claims to representativeness. A good sample is both large enough and unbiased, so that it mirrors the population it stands for.

Reliability and validity

These are the twin tests of good measurement. Reliability is consistency — a reliable instrument yields the same result when the study is repeated under the same conditions. Structured, standardised methods such as questionnaires tend to score highly here. Validity is truth — a valid instrument actually measures what it claims to measure. Qualitative methods often score highly on validity because their closeness to the subject yields authentic data.

The two frequently pull in opposite directions. A rigid questionnaire is reliable but may miss the real meaning of an answer; a rich ethnography is valid but hard to replicate. Recognising this trade-off is central to designing sound research — and to explaining, in an exam, why so many sociologists now combine methods.

Objectivity and value-neutrality

Positivists held that the researcher should stand outside their subject, keeping personal values from contaminating the findings, much as Weber urged value-neutrality in the conduct of enquiry. Critics reply that objectivity is an ideal never fully reached — the choice of topic, the framing of questions and the interpretation of results all carry values. The realistic goal is not the absence of values but their disciplined control, through transparency about method and reflexivity about the researcher's own position.

How to use this in the exam

Present the two traditions as complementary rather than rival, and let the positivism-versus-interpretivism debate organise your answer. Illustrate techniques with the reliability-validity trade-off, and reach for triangulation — combining methods so that the strengths of one offset the weaknesses of another — as your synthesising point. Concrete examples, a survey set against an ethnography, will lift an answer above a mere list of definitions.