Research methods explained
What each approach can and cannot answer, how to choose between them, and why your methodology chapter is not a description of what you did.
Method selection feels like a preference and is not. Your research question already determines what kind of evidence would answer it, and the job is to recognise which. Students who choose a method first and fit a question to it produce dissertations that never quite work.
The two families
| Quantitative | Qualitative | |
|---|---|---|
| Answers | How much, how many, is there a difference | How, why, what does it mean |
| Data | Numbers | Words, images, observations |
| Sample | Large, aiming to represent | Small, aiming for depth |
| Analysis | Statistical | Thematic or interpretive |
| Reasoning | Deductive: test a hypothesis | Inductive: build from the data |
| Strength | Generalisability | Depth of understanding |
| Weakness | Misses meaning and context | Cannot generalise |
Match the method to the question
Quantitative question
To what extent does Instagram use predict body image concerns among first-year undergraduates?
Needs measurement and a statistical test.
Qualitative question
How do first-year undergraduates describe the relationship between Instagram use and how they feel about their bodies?
Needs interviews and thematic analysis.
Same topic, different evidence. The words in your question tell you which: to what extent measures, how do explores.
Quantitative methods
Survey
Standardised questions to a large sample. Efficient and generalisable when the sample is properly drawn, but limited to what you thought to ask, and self-report is not behaviour.
Experiment
Manipulate one variable, control others, measure the effect. The only design that supports causal claims. Often impractical or unethical outside a laboratory.
Quasi-experiment
Comparison without random allocation, typically because groups already exist. Common in education and health, where randomising would be unacceptable. Weaker causal inference, but frequently the only realistic option.
Secondary data analysis
Analysing existing datasets such as the Labour Force Survey or Understanding Society. Large samples, no collection burden, no ethics application for the collection itself. You are limited to variables someone else chose to measure.
Qualitative methods
Semi-structured interviews
A topic guide with freedom to follow what emerges. The workhorse of qualitative student research. Ten to fifteen interviews is a typical dissertation scope. Our guide to data collection methods covers sample sizes and piloting.
Focus groups
Six to eight participants discussing together. Useful when the interaction between views is itself the data. Dominant voices are the usual risk.
Observation
Watching behaviour in its setting, either participating or not. Captures what people do rather than what they say they do, which frequently differs. Time-consuming and ethically delicate.
Document and content analysis
Systematic analysis of existing texts: policy documents, media coverage, social media, archives. Often the most realistic option for a student with no access to participants and no time for ethics approval.
Mixed methods
Three common designs, and each needs a reason.
| Design | Sequence | Used when |
|---|---|---|
| Explanatory | Quantitative, then qualitative | You have a pattern and need to understand why |
| Exploratory | Qualitative, then quantitative | You need to know what to measure before measuring |
| Convergent | Both at once, compared | You want two angles on the same question |
Mixed methods is roughly twice the work. Markers reward it when the question genuinely requires both and penalise it when it looks like indecision. Say explicitly what each strand contributes that the other cannot.
Methods versus methodology
This distinction costs more marks than any other in methodology chapters. Methods are what you did. Methodology is the reasoning that makes those methods appropriate.
Methods only
Twelve semi-structured interviews were conducted and analysed thematically.
Methodology
Because the study sought to understand how participants themselves made sense of adherence, rather than to measure its frequency, an interpretive approach was appropriate. Semi-structured interviews were chosen over a survey because they allow participants to raise factors the researcher had not anticipated. Twelve were conducted, consistent with the point at which saturation is typically reached in homogeneous samples (Guest et al., 2006).
The second version justifies rather than reports, and names a rejected alternative. Both moves are what "critical" means in a methodology chapter.
Sampling
| Approach | Type | Note |
|---|---|---|
| Random | Quantitative | Gold standard, rarely achievable in student work |
| Stratified | Quantitative | Random within subgroups, preserves proportions |
| Convenience | Either | Most student research. Acknowledge the limitation explicitly. |
| Purposive | Qualitative | Chosen for relevance, not representativeness |
| Snowball | Qualitative | Participants recruit others. For hard-to-reach groups. |
Convenience sampling is not a flaw to hide. It is a constraint to state, along with what it means for your claims.
What your methodology chapter must contain
- Your philosophical position, if your discipline expects one
- Why this approach suits this question, with an alternative considered and rejected
- Sampling: who, how many, how recruited, and why that number
- Data collection, in enough detail to be repeated
- Analysis approach, named and referenced
- Ethical approval and how consent and confidentiality were handled
- Limitations, stated before the marker finds them
Item two is where the marks are. Everything else can be accurate and the chapter still reads as descriptive if you never explain why you chose what you chose.
Last updated: 30 July 2026 · Written by the AssignWise editorial team ·Spotted something wrong?
Frequently asked questions
The techniques you use to collect and analyse data: interviews, surveys, experiments, observation, document analysis. Methodology is the layer above, meaning the reasoning for why those techniques suit your question.
Quantitative research measures, producing numbers you can test statistically, and answers how much or how many. Qualitative research explores meaning through words, and answers how and why. The question decides which you need, not preference.
Combining both in one study, usually because the question has a measurable component and an interpretive one. It is more work than either alone and needs justifying: doing both is not automatically better than doing one well.
Methods are what you did. Methodology is why those methods suit your question, including your assumptions about what counts as evidence. Markers ask for methodology and receive methods more often than any other confusion in dissertation writing.
Work backwards from the question. If it contains "how many" or "to what extent", you need quantitative methods. If it contains "how do" or "why do", you need qualitative. If it contains both, you have either a mixed-methods study or two questions.
Want a second pair of eyes on it?
Send us your document for a fixed, no-obligation quote. We reply within one working hour.
Get Your Free Quote