A research method should help answer the question being investigated. A questionnaire can collect comparable answers, an interview can explore meanings and observation can examine behaviour in a setting. None is automatically best for every topic. The useful comparison considers the evidence needed, the participants, practical constraints and ethical requirements.
Start by defining the research question and the relevant population. Then explain why a method fits and what it may miss. This guide uses a fictional study of participation in an after-school club. It is an educational illustration rather than a proposal to collect real student data. Follow the current syllabus and any school guidance when discussing actual research activities.
Make the research question specific
“Study education” is too broad to guide a method. “How many students in a named year group attend the club each week?” asks for a different kind of evidence from “How do students describe their reasons for joining or avoiding the club?” One focuses on a participation pattern; the other explores meanings and experiences.
A project can ask both, but the methods need to serve those distinct aims. Write the target clearly before selecting a favourite technique. If the question concerns behaviour during meetings, asking about attendance alone may not provide the required detail. Method choice begins with the claim you want to investigate, not with a memorised list of advantages.
Distinguish quantitative and qualitative evidence
Quantitative evidence is expressed numerically, while qualitative evidence concerns meanings, descriptions or other non-numerical detail. Methods do not always fit exclusively into one category. A questionnaire can contain closed numerical items and open-ended responses. An observation can record counts as well as descriptive field notes.
Explain the evidence produced by the particular design rather than labelling every questionnaire quantitative and every observation qualitative without qualification. In the fictional club study, a weekly attendance count provides numerical evidence. A student's account of feeling welcome provides a different kind of evidence. Both can be relevant if the research question requires them.
Use a questionnaire for comparable responses
A questionnaire can ask several participants the same questions in a consistent form. Closed questions can make responses easier to compare and summarise. In the club example, a carefully defined item about attendance during a specified week may help describe participation across the selected group.
However, the meaning and accuracy of answers depend on the questions and response process. “Do you attend regularly?” is ambiguous because participants may interpret “regularly” differently. A clear time period and suitable response options improve the design. Do not assume that identical wording automatically guarantees valid answers; respondents may misunderstand, forget or answer in a socially desirable way.
Recognise what closed questions can miss
Predetermined response options can omit reasons participants consider important. If the club questionnaire offers only “too busy” or “not interested”, it may miss transport, accessibility or other relevant experiences. Those are possible factors in the fictional design, not established findings about a real group.
An open response or an earlier exploratory stage can help identify suitable categories. Explain the trade-off: more detail can improve understanding but take longer to interpret consistently. A method evaluation should connect that limitation to the study's aim. If the aim is to understand unfamiliar reasons, a narrow list of options may be a particular weakness.
Use interviews to explore meanings
An interview can let participants explain an answer and allow appropriate follow-up questions. A less structured approach may reveal issues the researcher did not anticipate. In the club example, a participant might describe how different parts of the meeting affect their experience, offering detail beyond a simple attendance count.
The interaction can also shape responses. Participants may try to please the interviewer or feel uncomfortable discussing a topic. Differences in follow-up questions can complicate comparison. Evaluate the actual design: a structured interview and an unstructured interview do not have identical strengths and limitations. Explain how the format affects the evidence required.
Observe behaviour when the question needs it
Observation can examine what happens in a setting rather than relying only on retrospective reports. A study of who speaks during club activities may require evidence about interaction. The researcher could define the behaviours to record and observe appropriate sessions under the relevant permissions and ethical arrangements.
Observation does not automatically reveal what participants think or why they act. A quiet member may have several possible reasons for speaking less. The researcher's presence may also affect behaviour. Distinguish the observable action from an inferred motive. A useful evaluation explains both the access to behaviour and the limits of interpreting it.
Consider secondary sources
Existing attendance records or documents can provide information without creating every item of data from the beginning. Check why the records were produced, what their categories mean and whether they cover the period and population required. A register may record attendance but say little about why someone joined or stopped attending.
Definitions can change over time, and missing records can limit comparison. Do not assume that an official-looking document answers every research question accurately. Explain its relevance and limitations. If different sources use different meanings of participation, recognise that mismatch before combining their numbers into a single conclusion.
Define the population and sampling frame
The population is the group the study concerns. The sampling frame is the list or source used to select a sample where the design requires one. If the question concerns all students in a year group, interviewing only current club members excludes non-members whose reasons may be central to the aim.
Check who can enter the sample and who is missing. A convenient group can provide useful exploratory accounts but may not support broad generalisation. A larger sample does not automatically correct a biased selection process. The sociology concept guide explains why precise definitions matter before applying a label to a study.
Explain the sampling approach in context
Random selection requires a suitable process and sampling frame; it does not mean choosing whoever happens to be nearby. A stratified approach can select from relevant subgroups according to a defined plan. Other approaches may serve different practical or exploratory purposes. Use the current course's definitions and explain why the selected approach fits the research question.
Avoid saying any one sampling label guarantees perfect representativeness. Non-response, incomplete frames and other factors still matter. In the fictional club study, selecting only students available after a meeting may overrepresent participants with time to stay. The limitation comes from who is included, not simply whether the final sample number sounds large.
Pilot the questions and procedures
A pilot is a small preliminary check of the research design. It can reveal unclear wording, unsuitable response options or a procedure that takes longer than expected. For example, participants may interpret “last week” differently if the question does not identify the dates. Revising that wording can improve the later collection process.
A pilot does not prove that every later response will be accurate or that the sample is representative. State what the check can realistically reveal. If discussing a questionnaire, examine leading or double-barrelled wording as well. “Is the club enjoyable and useful?” combines two judgements that a participant might answer differently.
Include ethical considerations as part of design
Relevant issues can include informed consent, privacy, confidentiality and avoiding harm. In school-based research, permissions and participant understanding require attention to the setting and participants. Explain the issue that matters for the proposed method rather than adding an unrelated generic statement at the end.
For example, named answers about feeling excluded may reveal sensitive information to other club members. A design should consider how responses are recorded, accessed and reported under the appropriate guidance. Do not assume that removing names automatically prevents identification in a small group. Ethical reasoning affects what data can appropriately be collected and how the findings are presented.
Combine methods for a reason
Using more than one source can provide different perspectives on a question. Attendance records might describe a pattern, while interviews explore participants' accounts of it. The combination is useful when the evidence addresses complementary parts of the aim. Adding methods merely to make a project appear larger does not guarantee stronger conclusions.
Differences between sources need interpretation. If reported attendance and a register disagree, investigate definitions, dates or recording processes rather than automatically choosing whichever supports the preferred conclusion. The mark-scheme review guide offers related advice on separating visible evidence from unsupported assumptions in an examination response.
Write a contextual method recommendation
Name the method, explain the evidence it would produce and connect that evidence to the research question. Add a relevant practical, ethical or quality limitation. If a different method would address that limitation, explain the trade-off. This is stronger than writing the same advantages and disadvantages regardless of the topic.
Students seeking subject guidance can explore the O Level Sociology course, course catalogue and admissions FAQ. Bring the exact research question and your proposed design. Feedback can then focus on whether the method, sample and evidence actually fit together.
Frequently asked questions
Is a questionnaire always the best way to study a large group?
It can collect comparable responses, but suitability depends on the question, design and participants. Closed items may miss important meanings, and non-response or misunderstanding can limit the results. Explain the fit instead of relying on sample size alone.
Are interviews always qualitative?
Interview formats can produce different kinds of evidence. Structured questions can generate comparable responses, while open discussion can provide detailed accounts. Describe the particular design and data rather than assuming every use of a method has identical characteristics.
Does observation show why people behave as they do?
It can show actions and interactions, but motives may need additional evidence. Avoid treating an observed behaviour as proof of one explanation. Researcher presence and interpretation can also affect the evidence.
Does a large sample guarantee representativeness?
No. Selection, the sampling frame and non-response can matter as well as size. Identify who is included and excluded. A large convenient group may still differ from the population the study claims to describe.
What makes a methods answer more specific?
Connect the technique to the exact research aim, participants and evidence needed. Explain one relevant strength and limitation through that context. Avoid a generic list that could be copied unchanged into an unrelated study.
Ask about O and A Level courses
Share your subjects, current study stage and the support you need. The Abeel Eschool team can help you check the current course options.