What Is Research Methodology?

Research methodology is where a lot of students first feel out of their depth. The terms pile up fast (methodology, methods, design, paradigm, sampling strategy), and your advisor (supervisor, in the UK) expects you to use them correctly, often without anyone having explained them to you. Mix them up in your proposal, and your committee may assume the thinking behind them is muddled too.

We’ve coached 10,000+ students over the last 10 years, and we’ve found the jargon is usually the hard part, not the thinking. In this post, I’ll explain what research methodology means in plain language, how it differs from methods and design, and the four decisions it covers, then show you practical examples from two successful dissertations.

---

What is research methodology?

Research methodology is your study’s plan and the reasoning behind it: the set of decisions about how you’ll collect and analyze data to answer your research aims, objectives and research questions, along with the reasoning behind each one. In a dissertation or thesis, it comes down to four decisions:

• What data to collect (e.g., qualitative or quantitative data)
• Who to collect it from (i.e., the sampling strategy)
• How to collect it (i.e., the data collection method)
• How to analyze it (i.e., the data analysis methods)

These four aren’t separate boxes to tick. Your research questions decide what kind of data you need, the data decides who can give it to you, and both narrow the ways you can sensibly collect and analyze it. Change one, and the others usually have to move too.

Within any formal piece of academic research (a dissertation, a thesis or a journal article), you’ll find a methodology chapter or section that covers these decisions. A good one explains why each choice was made, showing that each method is the best fit for the study’s research questions. That justification is what your committee (your examiners, in the UK) is really reading for.

Saunders’ research onion is a popular way to see where these decisions fit among the others a methodology chapter makes, peeling back layer by layer from philosophy on the outside to techniques at the core:

You don’t need to use the onion to write a good methodology chapter, but it’s a useful map of what the chapter has to cover. Don’t worry about every term on it yet; the philosophy layer comes up again below. Our explainer on Saunders’ research onion walks through each layer.

Methodology vs. methods vs. design

A question we often get from students:

“So, is methodology just a fancy word for methods?”

Not quite. Methods are the specific techniques you use to collect and analyze data: an interview, a survey, a t-test, thematic analysis. Methodology is the strategy behind them. As one widely cited paper on research paradigms puts it, drawing on Michael Crotty, methodology is “the strategy or plan of action which lies behind the choice and use of particular methods”, and so it’s concerned with “why, what, from where, when and how data is collected and analyzed”.

Think of methods as the tools in a toolbox and methodology as the plan for the job: which tools you’ll use, in what order, and why those ones. Two studies can use the same method (interviews, say) and still have very different methodologies, because they use it for different reasons and pair it with different sampling and analysis.

Research design sits close to methodology, and the two words overlap. Design is more about the overall strategy of the study: for example, whether you’ll run an experiment in which you manipulate one variable while controlling others, follow people over time, or study a single case in depth. Above both sits your research philosophy, your assumptions about reality and knowledge, which shapes which designs and methods make sense in the first place.

The onion shows the order of the decisions. This figure shows how the four terms people mix up relate to each other:

Some programs simply call the chapter “Methods”. Don’t lose sleep over the labels. What matters is that your chapter covers the decisions and justifies each one, in the terms your program uses.

---

What are qualitative, quantitative and mixed-methods?

Qualitative, quantitative and mixed methods are the three broad methodological approaches, distinguished, roughly speaking, by their focus on words, numbers or both.

Qualitative research collects and analyzes words (written or spoken) and other non-numerical data, such as images, video or observed behavior. Quantitative research focuses on measurement and testing using numerical data.

A qualitative methodology is often used when the research aims and questions are exploratory. For example, you might use one to understand how people perceived an event, or how voters made sense of a political candidate. A quantitative methodology is often used when the aims are confirmatory: measuring the relationship between two variables (say, sleep and exam performance) or testing a set of hypotheses.

That pairing is a tendency, not a rule. One education researcher argues that the exploratory-confirmatory distinction “cuts across the qualitative-quantitative divide”: a survey can explore, and interviews can test whether earlier findings hold in a wider group. So ask first whether your study is exploring or testing, then pick the kind of data that does that job best.

The mixed methods approach combines qualitative and quantitative data in one study, so that each answers the part of the question it’s best suited to. It can be powerful, but running two strands of data and joining them up takes considerably more work, so it needs a clear reason rather than a wish to cover all bases.

Choosing between the three is where many of the students we work with get stuck, so this walkthrough shows you how to decide:

What is sampling strategy?

Sampling is about deciding who (or where) you’re going to collect your data from. It’s rarely possible to collect data from every single person in your group of interest (the “population”), so you’ll work with a smaller portion of that group that’s accessible and manageable (the “sample”).

How you select that sample (your sampling strategy) has a big effect on what your findings can claim. There are many sampling methods to choose from, but they fall into two broad camps: probability sampling and non-probability sampling.

Probability sampling selects participants at random, so that everyone in the population has a known chance of being picked. It’s like putting every name in a hat, shaking it up and drawing out the “winners”. Random selection reduces the risk of selection bias, and it’s what lets you generalize your results statistically to the whole population.

Non-probability sampling doesn’t rely on random selection. A convenience sample, for example, means you’d only interview or survey people you can easily reach (perhaps colleagues or members of a group you belong to). In purposive sampling, you choose participants deliberately because they have the experience your questions need. As one nursing research paper puts it, the point is “the better matching of the sample to the aims and objectives of the research”. The trade-off is that you can’t generalize the results statistically.

Here’s a sociology PhD from Cornell, a study of conflicts over honey bee health, describing its 66 interviewees in its methods appendix. It reads as a mostly purposive sample, so watch for what it says about who is, and isn’t, represented:

Notice the last sentence. Andrews says plainly that the sample isn’t representative and explains why: the respondents were chosen for the attributes each part of the research needed. A page later, the appendix also names the cost, listing selection bias among the limitations and noting that recruiting through clubs oversampled club members. Do both in your own write-up: say what kind of sample you have and why it fits your questions, then say what it can’t tell you.

This explainer walks through the main probability and non-probability sampling methods:

What are data collection methods?

Data collection methods are the ways you gather the data for your study. Some of the most common are:

• Interviews (which can be unstructured, semi-structured or structured)
• Focus groups and group interviews
• Surveys (online or on paper)
• Observations (watching and recording activities)
• Experiments and tests (e.g., a memory task run in a lab)
• Biophysical measurements (e.g., blood pressure or heart rate)
• Documents and records (e.g., financial reports or court records)

Which one you use depends on your research questions first, and on practicalities (time, access, money) second. If your research is exploratory, interviews and focus groups are often a good fit. If it aims to measure specific variables or test hypotheses, a survey or an experiment that produces numerical data is usually the better choice.

Get that order right. One mistake that comes up again and again is a method chosen for ease rather than fit: a survey because it seems quicker, or interviews because statistics feel scary. Surveys aren’t automatically easier (a good questionnaire takes real work to design and validate), and a method that doesn’t fit your questions will cost you far more time when your committee asks why you chose it.

---

What are data analysis methods?

Data analysis methods are the techniques you use to make sense of your data. They split along the same line as the data itself: qualitative (words-based) or quantitative (numbers-based).

Popular data analysis methods in qualitative research include:

• Qualitative content analysis
• Thematic analysis
• Discourse analysis
• Narrative analysis
• Interpretative phenomenological analysis (IPA)
• Grounded theory
• Framework analysis

Most qualitative methods involve data coding: labeling chunks of your data by what they’re about. The analysis method you choose decides how you code and what you build from the codes, and some studies use more than one, depending on their research questions.

On the quantitative side, the methods fall into two main groups:

• Descriptive statistics, which summarize your sample (e.g., means, medians, modes)
• Inferential statistics, which test whether patterns hold beyond your sample (e.g., correlation, regression, t-tests)

As with data collection, the right analysis method depends on your research aims and objectives, the type of data you have, and practicalities such as the software and skills available to you.

This Cornell master’s thesis in human development, a two-session psychology experiment on food nostalgia and memory, sets out both groups of methods and the software early in its analysis section:

The two sentences do the two jobs in order: describe the data first, then test it. They also name the tools (the R packages) and cite them, which is part of what lets someone else repeat the analysis. The section then sets out each model in turn and what it tested. Follow the same order in your own write-up, and say which test answers which research question, and why it suits your data.

How do I choose a research methodology?

Your research questions have a major influence on your methodology, so start with them, before any methods decisions. The first question to ask is whether your research is exploratory or confirmatory in nature.

If your aims are mainly exploratory, you might consider qualitative data collection methods (e.g., interviews) and analysis methods (e.g., qualitative content analysis). If they’re confirmatory, looking to measure or test something, quantitative methods (e.g., surveys, experiments and statistical analysis) are the more likely fit.

Then work through the four decisions in order, checking each one against your research questions as you go:

Practical example

Here are the same two dissertations from earlier, laid out against the four decisions. One is a quantitative experiment, the other a sociology PhD that leans mostly on interviews and observation. Read across each row to see how differently two real studies answered the same four questions:

Decision	Human development master’s (Zhou, 2024)	Sociology PhD (Andrews, 2020)
What data	Numbers: image ratings, memory confidence scores, recognition accuracy and reaction times	Mostly words: interview transcripts and observation notes, plus archival documents and a statewide survey
Who from	66 participants recruited, with exclusion criteria stated; 46 analyzed after invalid responses were filtered	66 interviewees chosen for particular attributes, from novice beekeepers to educators, mostly in New York State; 291 survey respondents
How collected	Two lab sessions a week apart: rating food images and recalling memories, then a recognition test	Participant observation (the author is a beekeeper and club officer), semi-structured interviews, observation of 22 organizations’ meetings, archival research and a survey
How analyzed	Descriptive statistics, then mixed-effects regression models (lme4) in R, plus ANOVAs	Transcripts and interview notes coded according to the research aims

Notice how each column hangs together. Both start from 66 people, and that’s where the resemblance ends. Zhou’s study asks whether food nostalgia predicts how confident and accurate people’s memories are, so everything points to numbers and statistics. Andrews studies how beekeepers clash over what responsible management of pests and disease looks like, so nearly everything points to people’s accounts. Neither is better. Each fits its own questions.

One gap is worth spotting, too. Andrews’ methods appendix says very little about analysis, only that the material was “coded according to different research aims”. That can be enough in an ethnographic, theory-led study like this one, but your committee will usually expect you to name your analysis method and justify it. Check what yours asks for.

If you’d like to see how real methodology chapters handle this, our research methodology examples annotate extracts from three master’s theses, one for each approach. And once your four decisions are made, our guide to writing the methodology chapter shows you how to write them up.

Still have questions?

What are the types of research methodology?

The three main types are qualitative, quantitative and mixed methods. Qualitative research works mainly with words, such as interview transcripts, quantitative research works with numbers, and mixed methods combines the two in one study. Lists that name more types usually add research designs, such as experimental, descriptive or case study research, which describe a study’s overall strategy rather than its kind of data.

Why is research methodology important?

Your methodology is what makes your findings believable. If the data, sample and analysis don’t fit your research questions, the conclusions built on them won’t hold up, however interesting they are. A clearly explained methodology also lets other researchers judge your study and, for quantitative work, repeat it, which is why committees read the methodology chapter so closely.

Which research methodology is best?

No methodology is best in general. The best one for your study is the one that answers your research questions within the time, access and skills you have. A study asking whether one variable predicts another usually needs quantitative data, while a study asking how people experience something usually needs qualitative data. Your program’s conventions matter too, so look at a few recent dissertations from your department.

How do I start writing my methodology?

Start with your research questions, not with a list of methods. Decide what kind of data would answer them, who could provide it, how you’ll collect it and how you’ll analyze it, and note a reason for each choice as you go. Then check your program’s guidelines for how the chapter should be structured.

Can’t find your answer here? Ask a Grad Coach directly – the initial chat is free.

Speak with a friendly coach →

Cite-worthy sources

Cite this article

Jansen, D. & Warren, K. (2020). What Is Research Methodology?. Grad Coach. https://gradcoach.com/what-is-research-methodology/

Copy citationMore citation formats

• Andrews, E. (2020). Beekeeping in the Anthropocene: Theorizing conflicts over honey bee health [Doctoral dissertation, Cornell University].
• Campbell, S., Greenwood, M., Prior, S., Shearer, T., Walkem, K., Young, S., Bywaters, D., & Walker, K. (2020). Purposive sampling: Complex or simple? Research case examples. Journal of Research in Nursing, 25(8), 652-661.
• Foster, C. (2024). Methodological pragmatism in educational research: From qualitative-quantitative to exploratory-confirmatory distinctions. International Journal of Research & Method in Education, 47(1), 4-19.
• Saunders, M. N. K., Lewis, P., & Thornhill, A. (2019). Research methods for business students (8th ed.). Pearson.
• Scotland, J. (2012). Exploring the philosophical underpinnings of research: Relating ontology and epistemology to the methodology and methods of the scientific, interpretive, and critical research paradigms. English Language Teaching, 5(9), 9-16.
• Zhou, S. (2024). Effect of food nostalgia on memory accuracy and memory confidence [Master’s thesis, Cornell University].