You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
Sociological research is a systematic process through which sociologists gather evidence to test theories, support or challenge existing knowledge, and generate new understanding about the social world. Far from being a haphazard exercise, research follows a series of deliberate decisions — each of which has consequences for the validity, reliability, and representativeness of the eventual findings. Understanding the research process is essential for A-Level Sociology because it underpins every study you will encounter and is the foundation on which all evaluation rests: you cannot judge whether Durkheim's conclusions about suicide are trustworthy, or whether Willis's account of 'the lads' is generalisable, unless you understand the methodological journey each researcher took to reach them. This lesson maps that journey stage by stage, and — crucially — shows how each stage is shaped by the deeper theoretical battle between positivism and interpretivism.
This material sits within the Theory and Methods component of the AQA A-Level Sociology specification (7192). Theory and Methods is assessed on Paper 1 (7192/1) — Education with Theory and Methods (where it appears as a 10-mark methods question and a 10-mark theory/methods question) and on Paper 3 (7192/3) — Crime and Deviance with Theory and Methods (where it can be examined as a 10-mark "Outline and explain" question and a substantial 30-mark essay). The research process also underpins the Methods in Context question on Paper 1, a 20-mark applied question requiring you to evaluate a method for studying a specific educational issue. Understanding how a research design is assembled — operationalisation, sampling, pilot studies, choice of method — is therefore directly examinable across the whole qualification, not just in stand-alone methods questions.
The research process is not an abstract checklist; every stage connects to the substantive topics you study elsewhere.
Key Definition: Empirical evidence — data and information acquired through observation, experience, or experiment, rather than through logic, theory, or belief alone.
Sociologists conduct research for several interconnected reasons:
| Purpose | Explanation |
|---|---|
| To test hypotheses | Research allows sociologists to gather empirical evidence to support or refute theoretical claims about society |
| To inform social policy | Governments and organisations rely on sociological research to shape laws, policies, and interventions |
| To challenge common sense | Everyday assumptions about social life are often incorrect; research provides evidence-based alternatives |
| To give a voice to the powerless | Research can highlight the experiences of marginalised groups who are otherwise ignored in public discourse |
| To build and refine theory | Research findings contribute to the ongoing development of sociological theories and perspectives |
A vivid illustration of research challenging common sense is Eileen Barker's The Making of a Moonie (1984): the popular assumption was that members of the Unification Church had been 'brainwashed', but Barker's longitudinal research showed that most people who attended recruitment workshops did not join, and many who joined left voluntarily — an evidence-based finding that overturned a moral panic. This is the core value of systematic research: it disciplines our intuitions against evidence.
The research process typically follows a logical sequence, although in practice researchers move back and forth between stages — qualitative research in particular is often iterative, with the research question evolving as fieldwork progresses. The flow diagram below presents the idealised linear model for clarity, but you should remember that real research is rarely so tidy.
flowchart TD
A[Identify research topic] --> B[Review existing literature]
B --> C[Formulate hypothesis or research question]
C --> D[Operationalise concepts]
D --> E[Select research method - PET considerations]
E --> F[Conduct pilot study]
F --> G[Collect data - primary or secondary]
G --> H[Analyse data]
H --> I[Draw conclusions and report findings]
I -. new questions raised .-> A
F -. refine instrument .-> D
The starting point is choosing an area of social life to investigate. This choice may be influenced by:
Before collecting new data, researchers conduct a thorough literature review — a survey of existing research, theories, and debates in their chosen area.
| Purpose of a Literature Review | Detail |
|---|---|
| Identify what is already known | Avoids duplicating existing research |
| Find gaps in knowledge | Highlights questions that have not yet been answered |
| Refine the research question | Helps the researcher focus on a specific, manageable question |
| Identify appropriate methods | Shows which methods have been used successfully (or unsuccessfully) in similar studies |
| Establish a theoretical framework | Connects the new research to existing sociological theories |
Key Definition: Literature review — a comprehensive survey and critical analysis of existing research and theoretical writing relevant to a particular topic or research question.
A hypothesis is a testable statement that predicts a relationship between two or more variables. Crucially, whether a researcher uses a hypothesis at all is itself a theoretical choice: positivists typically begin with a precise hypothesis to be confirmed or falsified, whereas interpretivists often prefer an open-ended research question because imposing a hypothesis in advance risks pre-judging the meanings they want to discover.
Key Definition: Hypothesis — a precise, testable prediction about the relationship between two or more variables. It is usually derived from a theory and is tested through empirical research.
Operationalisation is the process of turning abstract sociological concepts into measurable indicators. This is essential because many sociological concepts — 'social class', 'educational achievement', 'poverty', 'integration' — are too vague to measure directly.
| Abstract Concept | Possible Operationalisation |
|---|---|
| Social class | Occupation (using the NS-SEC classification), income level, or parental education |
| Educational achievement | GCSE grades, A-Level results, or degree classification |
| Poverty | Household income below 60% of the national median (relative poverty measure) |
| Religiosity | Frequency of church attendance, self-reported belief in God, or membership of a religious organisation |
| Crime | Number of recorded offences, self-report survey data, or victim survey data |
The choice of operational definition has dramatic consequences. If 'poverty' is operationalised as household income below 60% of the median, a different proportion of the population counts as poor than if an absolute, subsistence-based measure is used — so the apparent "scale" of poverty is partly an artefact of measurement. Interpretivists go further, arguing that operationalising rich human experiences into numerical indicators inevitably strips out the meaning that matters most.
Exam Tip: Operationalisation appears frequently in exam questions. Be ready to explain why it is necessary, to give examples, and to evaluate it — note that the way a concept is operationalised can significantly affect findings, and that interpretivists see operationalisation as a threat to validity.
The choice of method is best organised using the PET framework — Practical, Ethical, and Theoretical factors. This framework is the analytical spine of every methods answer at A-Level.
| Factor | Key Questions |
|---|---|
| Practical | How much time and money is available? Can the researcher gain access? Is the topic hard to reach (e.g. criminal behaviour)? How large is the research team? |
| Ethical | Will participants give informed consent? Is there deception, harm, or invasion of privacy? Are vulnerable groups (e.g. children) involved? |
| Theoretical | Does the researcher prioritise reliability and representativeness (positivist) or validity and Verstehen (interpretivist)? |
A pilot study is a small-scale trial run of the research, conducted before the main study begins.
| Purpose of a Pilot Study | Detail |
|---|---|
| Testing the research instrument | Checks whether questionnaires, interview schedules, or observation frameworks work effectively |
| Identifying practical problems | Reveals issues such as difficulty accessing participants, ambiguous questions, or procedures that take too long |
| Refining wording | Ensures questions are clear and understood as the researcher intended |
| Checking ethical issues | Identifies unforeseen ethical problems |
| Estimating time and cost | Helps the researcher plan resources for the full study |
| Training the researcher | Provides practice in administering the instrument, important for interviews and observations |
Key Definition: Pilot study — a small-scale preliminary study conducted before the main research in order to test the feasibility, timing, cost, and effectiveness of the research design and methods.
This is the main fieldwork stage. Data may be:
The researcher interprets the analysed data in relation to the original hypothesis or research question. Findings are reported in academic journals, books, or policy reports, and should include a clear account of methods so that others can evaluate and — for positivists especially — replicate the research.
Every stage above ultimately serves four overarching evaluative concepts. These are the criteria you deploy to judge any study.
| Concept | Meaning | Prioritised by |
|---|---|---|
| Validity | The extent to which a method measures what it claims to and gives a true, authentic picture | Interpretivists |
| Reliability | The consistency of a method — whether repetition produces the same results (replicability) | Positivists |
| Representativeness | Whether the sample is typical of the wider population, allowing generalisation | Positivists |
| Objectivity | Whether research is free from the researcher's personal values and bias | Positivists (and contested by Marxists/feminists) |
The crucial AO3 insight is that these criteria often trade off against one another. A large postal survey maximises representativeness and reliability but tends to sacrifice validity; an unstructured interview maximises validity but undermines reliability and representativeness. There is no perfect method — only methods that are more or less appropriate to a given research question and theoretical stance. This trade-off is why triangulation (combining methods to offset their respective weaknesses) is so often recommended.
Exam Tip: When evaluating any study, work through validity, reliability, representativeness, and objectivity in turn, and always explain why the method's features strengthen or weaken each one — assertion alone scores poorly.
A recurring exam theme is that the choice of method is not a neutral, purely technical decision — it flows from the researcher's theoretical position about what society is and how it can be known. This is sometimes summarised as the link between ontology (what social reality is), epistemology (how it can be known), and methodology (which methods follow). You do not need the technical vocabulary, but you do need to grasp the logic.
| Theoretical position | View of social reality | Preferred methods | Priority |
|---|---|---|---|
| Positivism | Society is an external, objective reality of measurable "social facts" that shape behaviour | Questionnaires, structured interviews, experiments, official statistics | Reliability, representativeness, objectivity, cause and effect |
| Interpretivism | Society is constructed through the subjective meanings people give to action | Unstructured interviews, participant observation, personal documents | Validity, depth, Verstehen |
| Realism / critical realism | Observable patterns are produced by underlying, often unobservable structures and mechanisms | Mixed methods; willing to use both quantitative and qualitative tools | Explaining underlying causes |
Durkheim exemplifies the positivist logic: because he saw suicide rates as a social fact, he reached for official statistics and sought law-like correlations. Weber and later interpretivists exemplify the opposite logic: because they saw social action as meaningful, they reached for methods that recover meaning. The single most reliable way to lift a methods essay into the top bands is to show that you understand this connection — that arguing about methods is arguing about the nature of social reality.
Because each method embodies a trade-off, the strongest conclusions are conditional: a method is good for a particular research aim. A study seeking to measure the scale of a social problem across a whole population needs reliability and representativeness, so a survey is appropriate; a study seeking to understand how a small group experiences something needs validity and depth, so observation or unstructured interviews are appropriate. Examiners reward candidates who reach this judgement explicitly rather than declaring one method simply "best".
The PET framework introduced above can be applied to almost any scenario question. Below it is worked through for a concrete example — investigating why some pupils truant — to model the applied reasoning examiners reward.
| Factor | Applied to studying truancy |
|---|---|
| Practical | Truants are by definition absent and hard to reach; a school roll provides a sampling frame but lists pupils who should be present, not where they actually are; funding and time limit how long a researcher can spend tracking them down |
| Ethical | Pupils are minors, so parental and school consent is needed; truancy may involve other deviant activity, raising questions about confidentiality and whether to report it; promising anonymity may conflict with safeguarding duties |
| Theoretical | A positivist might use official attendance statistics and a structured survey to find correlations (e.g. with free-school-meal status); an interpretivist would prefer unstructured interviews or observation to understand how truants themselves make sense of school |
This kind of applied breakdown — naming the factor, then tying it to the specific group and topic — is exactly what the Methods in Context question demands. Notice how the characteristics of the research population (here, that truants are absent and are children) drive the practical and ethical analysis. The same discipline applies to studying any hard-to-reach group: offenders, gang members, abuse survivors, or members of secretive religious movements.
Three further sets of factors shape method choice and are worth deploying in essays:
| Type | Description | Example |
|---|---|---|
| Quantitative data | Numerical data measurable and analysed statistically | Census data, crime statistics, survey results |
| Qualitative data | Non-numerical data giving rich, detailed descriptions | Interview transcripts, diary entries, ethnographic field notes |
| Primary data | Data collected first-hand for the researcher's own study | Data from a questionnaire the researcher designed |
| Secondary data | Data that already exists, produced for another purpose | Official statistics, newspaper articles, historical documents |
A key debate concerns whether research can — or should — be value-free:
Key Definition: Value-freedom — the idea that research should be conducted objectively, without the researcher's personal values influencing the process or findings.
Ethics is not a single stage but a thread running through every decision, and it forms the "E" of PET. Professional bodies such as the British Sociological Association (BSA) publish guidelines that researchers are expected to follow. The principal ethical considerations are:
| Ethical principle | What it requires | Example of tension |
|---|---|---|
| Informed consent | Participants should agree to take part knowing the study's nature and purpose | Covert observation (e.g. Humphreys, Patrick) makes consent impossible by design |
| Confidentiality and anonymity | Identities and data should be protected | Patrick published under a pseudonym and concealed the gang's identity to protect everyone |
| Avoiding harm | Participants (and researchers) should not be physically or psychologically harmed | Studying violent gangs endangered Patrick; sensitive interviews may distress respondents |
| Right to withdraw | Participants may leave at any time | Hard to honour in covert work where participants do not know they are being studied |
| Vulnerable groups | Extra protection for children and others who cannot fully consent | Research in schools requires parental and institutional consent and safeguarding awareness |
These principles frequently conflict with practical and theoretical goals. Covert participant observation often produces the most valid data on deviant groups precisely because it abandons informed consent — the clearest illustration that PET factors pull against one another. A researcher must weigh the potential knowledge gained against the ethical cost, and examiners reward candidates who show they understand this is a genuine dilemma, not a checklist. Note too that ethical and theoretical questions intertwine with the value-freedom debate: feminists such as Oakley argue that a supposedly "detached", value-free stance is itself an ethical failing because it treats participants as mere data-sources rather than as people.
(a) Outline and explain two reasons why sociologists carry out a pilot study before their main research. [10 marks]
AO breakdown: This question is pure AO1 (knowledge, 4 marks) and AO2 (application, 6 marks). There is no AO3 requirement, so do not waste time evaluating — develop two clear, separate reasons with applied detail.
(b) Item: Sociologists must make a series of decisions before they begin collecting data, including how to turn abstract concepts into something they can actually measure. The way they do this can shape their findings, and different sociologists disagree about whether this process strengthens or weakens research.
Applying material from the Item and your knowledge, evaluate the view that operationalising sociological concepts improves the quality of research. [20 marks]
AO breakdown: AO1 (knowledge) ~6, AO2 (application of the Item to the question) ~6, AO3 (analysis and evaluation) ~8. You must use the Item hook ("shape their findings") and reach a reasoned judgement.
Mid-band response (to part b): "Operationalising concepts means turning ideas like social class into something measurable, such as job type. This is good because it lets sociologists collect data and make comparisons. Positivists like it because it gives them numbers. However, interpretivists say it loses meaning. So operationalisation has strengths and weaknesses." This answer states relevant points but stays general. It names positivists and interpretivists but does not develop why the disagreement matters, gives no studies, and the evaluation is a bald list rather than a sustained argument.
Stronger response: "Operationalisation improves research by making abstract concepts measurable, which is essential for the comparison and replication positivists value. For example, operationalising social class through the NS-SEC scale allowed large-scale surveys to demonstrate the persistent class gap in attainment. As the Item notes, however, the way concepts are operationalised can shape findings: defining poverty as below 60% of median income produces a different 'poverty rate' than an absolute measure would, so the choice is not neutral. Interpretivists argue this is a weakness because reducing rich experience to indicators sacrifices validity." This develops both sides, uses the Item explicitly, and supplies concrete examples. To reach the top band it needs a clearer theoretical framing and a reasoned overall judgement.
Top-band response: "Whether operationalisation improves research cannot be answered without reference to the positivism/interpretivism debate, because the two traditions hold incompatible views of what 'quality' means. For positivists, quality means reliability and representativeness: operationalising 'integration' into measurable indicators is exactly what enabled Durkheim to compare suicide rates across groups and infer social causes, and operationalising class via NS-SEC underpins the entire evidence base on the attainment gap. On this view operationalisation is the precondition of scientific sociology. Yet, as the Item signals, the process 'shapes findings' — the apparent scale of poverty shifts with the chosen definition, so operational decisions smuggle in value judgements that undermine the very objectivity positivists claim. Interpretivists press a deeper objection: operationalisation imposes the researcher's categories on respondents and strips out the subjective meanings that constitute social reality, fatally damaging validity. The most defensible judgement is that operationalisation improves reliability at the cost of validity; its value therefore depends on the research aim. For testing patterns across large populations it is indispensable; for understanding lived meaning it can be counterproductive — which is why triangulation, combining operationalised measures with qualitative depth, is often the strongest design." This sustains methodological evaluation throughout, frames the question theoretically, uses studies precisely, engages the Item, and reaches a nuanced, justified conclusion.
Examiner-style commentary: The decisive difference between the Stronger and Top-band scripts is sustained evaluation tied to a conceptual framework. The Top-band answer never simply lists pros and cons; it explains why the disagreement exists (rival definitions of quality), anchors each claim in a real study (Durkheim, NS-SEC), exploits the Item, and resolves the debate with a conditional judgement. Candidates who memorise studies but cannot connect them to the validity–reliability trade-off plateau in the middle bands.
The classic research-process model was developed for an era of surveys, interviews and paper documents. Big data and digital research now complicate it. When a sociologist analyses millions of social-media posts, the "sampling frame" is opaque, consent is rarely obtained, and the data was generated for commercial — not research — purposes, echoing the interpretivist critique of official statistics in a new form. Digital and online research ethics raise fresh questions: is a public tweet "published" or "private"? Can a researcher covertly observe an online forum? The boundary between observation and surveillance blurs. There is also growing interest in participatory and co-produced research, in which the studied community helps frame the research question — a direct descendant of Oakley's feminist critique of hierarchical research relationships. These debates show that the research process is not a fixed recipe but an evolving set of practices continually renegotiated as technology and ethics change.
The research process provides a structured framework for sociological inquiry, but each stage — topic, literature review, hypothesis or question, operationalisation, method choice via PET, pilot study, data collection, analysis, and reporting — involves decisions that flow from the researcher's theoretical commitments and shape the validity, reliability, and representativeness of the findings. Mastering the process means more than memorising the stages: it means understanding the trade-offs between them and recognising the positivism/interpretivism debate that runs underneath every choice.
Exam Tip: In scenario questions you may be asked to design a study for a given issue. Work through the stages systematically, justify each choice using PET, and show awareness of the validity–reliability trade-off. Examiners reward applied, evaluative reasoning over recited definitions.
This content is aligned with the AQA A-Level Sociology (7192) specification.