Most methodology chapters open with a philosophy section, and most of them are written defensively - a paragraph of borrowed terminology that could be deleted without changing anything that follows. Examiners notice, because the point of the section is to justify your method, and a justification that does not connect to the method is decoration.
This post explains the two underlying questions, the four positions you are most likely to adopt, how to choose honestly, and how to write the section so it earns its place.
A methodology chapter that hangs together
ThesisAI drafts full academic documents with consistent structure and inline citations verified against the papers they came from, so your philosophy, method and analysis stay aligned.
Empezar a escribirThe Two Questions Underneath
Strip the vocabulary away and research philosophy asks two things.
Ontology: what is there to know? Does the thing you are studying exist independently of the people observing it, or is it produced by how people interpret and enact it? Gravity does not need a believer. "Job satisfaction" arguably does. Most social phenomena sit somewhere in between, and where you place yours is an ontological claim.
Epistemology: how can we know it? If the thing exists, what counts as legitimate evidence about it? Measurement and statistical inference? Detailed accounts from the people involved? Both, in some order?
Your answers constrain your method. If you hold that organisational culture is constructed through the meanings staff attach to daily practice, a Likert questionnaire is a strange instrument for studying it. If you hold that reaction time is a stable measurable property, a set of six interviews is a strange way to establish it. The philosophy section exists to make that fit explicit.
The Four Positions You Will Meet
| Paradigm | Ontology | Epistemology | Typical methods |
|---|---|---|---|
| Positivism | One reality exists independently of observers | Knowledge comes from observable, measurable fact; the researcher stays detached | Experiments, surveys, statistical modelling, hypothesis testing |
| Interpretivism | Reality is socially constructed and multiple | Knowledge comes from understanding meanings; the researcher is part of the process | Interviews, ethnography, case studies, thematic and narrative analysis |
| Critical realism | A real world exists, but our access to it is always mediated and fallible | Look past observed events to the underlying mechanisms that generate them | Mixed methods, in-depth case studies, retroductive reasoning |
| Pragmatism | Sets the question aside as unresolvable in practice | Whatever produces useful, credible answers to the research question | Mixed methods, action research, applied and policy work |
Two more you may need to name. Postpositivism is positivism that has accepted its limits: a real world exists, but measurement is imperfect and theories are provisional rather than proven. Much quantitative social science is postpositivist even when it calls itself positivist. Constructivism is often used interchangeably with interpretivism; where authors distinguish them, constructivism emphasises how knowledge is built, interpretivism how meaning is understood.
Choosing Honestly
The common failure is choosing the paradigm that sounds most sophisticated and then describing a method that contradicts it. Work in the other direction: start from your question and your data, and identify the position you are already occupying.
| If your question is ... | You are probably ... |
|---|---|
| Does X cause Y? How much does X change Y? | Positivist or postpositivist |
| How do people experience X? What does X mean to them? | Interpretivist |
| Why does X produce Y here but not there? What mechanism explains it? | Critical realist |
| What works, for whom, and what should we do about it? | Pragmatist |
Three honest checks before you commit:
- Does your method match? If you claim interpretivism and then run a regression on 400 questionnaires, one of the two is wrong.
- Does your analysis match? Counting theme frequencies and reporting them as prevalence is a positivist move inside a qualitative design. That can be defensible, but it needs saying.
- Does your field accept it? Paradigm conventions are disciplinary. Read three recent theses from your own department before deciding, because the acceptable range in experimental psychology is not the range in critical management studies.
Writing the Section
Aim for three or four paragraphs in a master's thesis, more in a doctorate. The structure that works:
- State your ontological position and why your phenomenon warrants it
- State your epistemological position and what will therefore count as evidence
- Name the paradigm and cite the methodological literature you are drawing on
- Draw the consequence explicitly: because of the above, the study uses these methods
Step four is the one students skip, and it is the only one that makes the section load-bearing.
Weak: "This research adopts an interpretivist philosophy. Interpretivism holds that reality is socially constructed and that there are multiple realities. Saunders et al. (2019) describe interpretivism as focusing on meanings. A qualitative approach was therefore adopted."
Textbook definitions with nothing about this study. "Therefore" does no work, because nothing specific preceded it. This paragraph would sit equally well in any interpretivist thesis ever written, which is the problem.
Strong: "This study treats 'inclusive leadership' as a category that exists through the meanings staff attach to their managers' behaviour rather than as a fixed property of individuals, since the same action is read as inclusive in one team and performative in another. That is an interpretivist ontological commitment, and it has a direct methodological consequence: a standardised scale would record whether staff endorse a researcher's definition, not how they construct the category themselves. The study therefore uses semi-structured interviews and reflexive thematic analysis, and treats divergence between accounts as data rather than as measurement error."
The position is tied to this phenomenon, the alternative is considered and rejected for a stated reason, and the method follows from it. Note the final clause: the philosophy changes how disagreement in the data is interpreted, which is a real consequence.
Mixed Methods and the Paradigm Problem
If you are combining qualitative and quantitative work, you will meet the objection that the two rest on incompatible assumptions. Two respectable answers exist.
Pragmatism declines the argument: the research question dictates the method, and the philosophical dispute is set aside as unproductive for applied work. This is the most common position in mixed methods theses and is well supported in the literature.
Critical realism answers it directly: a real world generates observable events, quantitative work can establish patterns in those events, and qualitative work can investigate the mechanisms producing them. The two are complementary because they address different layers rather than competing accounts of the same layer.
What does not work is claiming both positivism and interpretivism in adjacent paragraphs without addressing the tension. Our guide to qualitative vs quantitative research covers the practical side of combining them.
FAQs About Research Philosophy
Do I really need a philosophy section?
In most UK, European and Australian social science programmes, yes, and it is often explicitly marked. In many North American and STEM programmes it is not expected at all. Check your department's thesis template and recent submissions rather than assuming.
What is the difference between ontology and epistemology?
Ontology asks what exists; epistemology asks how we can know about it. Ontology comes first logically, because what you think exists constrains what would count as evidence about it.
Is positivism outdated?
No, though strict positivism is rare in contemporary social science. Most quantitative work is more accurately postpositivist: it assumes a real world while accepting that measurement is imperfect and findings are provisional. Describing yourself as postpositivist is often the more accurate claim.
What is the difference between methodology and method?
Methodology is the reasoning that justifies your approach, including philosophy, design and rationale. Methods are the specific procedures - the interviews, the instrument, the statistical test. Our research methodology guide covers the full chapter.
Can I use critical realism without a heavy theory background?
You can, but be careful. Critical realism has a demanding vocabulary and examiners who know it will expect you to use it precisely. If you cannot explain the difference between the empirical, the actual and the real, pragmatism is usually the more honest choice.
How long should the section be?
Three or four paragraphs in a master's thesis, two to four pages in a doctorate. Length is not the point - connection is. A short section that changes how you analysed your data beats three pages of definitions.
What is an axiology?
Your position on the role of values in research: whether you claim detachment, or acknowledge that your values shape the questions and interpretation. Some templates ask for it alongside ontology and epistemology. Interpretivist and critical work usually pairs it with a reflexivity statement about your own position relative to participants.
The section is short and it is not really about philosophy. It is about whether you can explain why your method is the right instrument for the thing you are studying, and why the obvious alternative is not. Write the method first if it helps, then write the philosophy that honestly describes what you did. Our guide to writing a methodology section covers everything that follows it.