Should You Choose a Paradigm Before Choosing Your Research Methods?

Does Research Philosophy Have to Come Before Methods?

A familiar research-methodology diagram begins at the top with ontology and epistemology, moves downward to paradigm or research philosophy, then to methodology, research design, methods, and finally data analysis.

The implication seems straightforward: settle your philosophy first, and everything else should follow.

There is considerable logic behind this sequence. If your assumptions about knowledge influence what counts as evidence, then choosing methods without considering those assumptions can produce an incoherent study. Crotty's influential framework, for example, connects epistemology, theoretical perspective, methodology, and methods while emphasizing that these elements should be logically related.

But conceptual dependence is not necessarily chronological order.

Researchers do not always begin with a fully articulated philosophical position and then derive every later decision from it. A study may begin with a practical problem, theoretical puzzle, research question, methodological tradition, available dataset, community concern, or unexpected observation. As the study develops, researchers may become more explicit about the philosophical assumptions their questions and methodological choices entail.

The goal is therefore not to obey one universal sequence. It is to reach a coherent relationship among the research problem, question, philosophical assumptions, methodology, methods, evidence, and intended claims.

No, Paradigm Selection Does Not Have to Be the First Step

Research Planning Is Often More Iterative Than Methodology Diagrams Suggest

Why Does the Paradigm-First Model Make Sense?

The paradigm-first model begins from a reasonable premise: research methods are not intellectually isolated procedures.

If a researcher believes that knowledge of a phenomenon can be developed through standardized measurement and controlled observation, some methodological choices become more plausible than others. If understanding contextual meanings is central to what counts as knowledge, methods capable of preserving those meanings become important.

Philosophical assumptions can therefore constrain methodological possibilities.

This is why research-methodology frameworks often depict connections among epistemology, theoretical perspective, methodology, and methods. Crotty's framework explicitly asks researchers to articulate how these elements relate rather than treating methods as freestanding techniques.

For researchers working within a strongly established philosophical tradition, beginning with the paradigm may therefore be entirely appropriate.

But Logical Order Is Not the Same as Discovery Order

A methodology chapter presents research as a coherent argument. The process through which the researcher arrived at that argument may have been considerably messier.

You might begin by noticing that teachers respond very differently to generative AI policies. That observation leads to a question about how teachers interpret professional responsibility. The question makes meaning and context central. You then recognize that an interpretive methodology and interpretivist philosophical assumptions fit the inquiry.

In the final thesis, you may explain the paradigm before describing the methodology.

Chronologically, however, the research question helped you recognize the philosophical position.

There is nothing inherently defective about that sequence.

Research Questions Can Lead You Toward Philosophical Assumptions

Consider these two questions:

"Does access to an AI tutor improve students' examination performance?"

"How do students understand the role of AI tutoring in becoming independent learners?"

The first question invites evidence about effects and comparison. The second makes participants' meanings central.

You do not need to choose positivism or interpretivism before noticing that these questions seek different forms of knowledge.

Indeed, beginning with the substantive question can prevent philosophy from becoming an exercise in selecting a preferred "-ism" and then searching for a research problem that fits it.

Sometimes Methodology Comes With Philosophical Commitments

Certain methodological traditions have developed alongside recognizable philosophical positions.

If you decide that phenomenology is appropriate because your research genuinely concerns lived experience, philosophical questions associated with phenomenological inquiry become difficult to avoid. If you conduct constructivist grounded theory, the methodological tradition already carries assumptions that should inform how you understand data generation and analysis.

Likewise, critical realist research brings particular commitments concerning reality, causal mechanisms, and explanation.

In these situations, choosing a methodology can prompt or refine your philosophical position rather than merely follow it.

The relevant requirement is that you understand the commitments embedded in the methodological tradition you adopt.

Methods Can Also Reveal Philosophical Questions You Had Not Considered

Suppose you initially decide to use a questionnaire because a validated instrument already exists.

As you examine the scale, you discover that it treats "student engagement" as a stable construct represented by a single numerical score.

You then need to ask whether that conceptualization fits your research question. Do you regard engagement as sufficiently stable and comparable across participants? Are there contextual meanings the scale cannot represent? What kind of knowledge does the score provide?

The method has not dictated your paradigm. It has exposed philosophical assumptions that require examination.

This is another reason research planning can be iterative.

Practical Problems Often Come Before Philosophy

Applied research frequently begins because someone needs to know something.

A university has declining retention. A hospital wants to evaluate a new training program. A government agency needs to understand why a policy is failing. A community identifies an inequity requiring investigation.

Researchers may initially begin with the problem rather than with ontology.

Pragmatism is especially relevant here because it conceptualizes research as inquiry directed toward problematic situations, experience, action, and consequences.

This does not mean philosophy disappears. It means the philosophical reasoning can emerge through examining what kinds of inquiry the problem requires.

Transformative Research May Begin With Ethical and Political Commitments

A transformative research paradigm provides another sequence.

A researcher may begin from concern about an injustice affecting a marginalized community. Questions about power, human rights, participation, cultural responsiveness, and whose knowledge counts may then shape the research problem and methodology.

Here, axiological and ethical commitments can precede detailed decisions about methods.

The research does not necessarily begin by resolving abstract ontology before identifying the social problem. Yet its philosophical orientation can still be explicit and consequential throughout the inquiry.

Critical Realist Research May Begin With an Explanatory Puzzle

A researcher observes that the same educational intervention produces different outcomes across schools.

The puzzle is explanatory: why does it work here but not there?

That question may lead the researcher toward critical realism because its concepts of generative mechanisms, context, open systems, and stratified reality provide a useful explanatory framework.

Again, the substantive research problem may lead toward the philosophy rather than the philosophy appearing first chronologically.

There Is No Universal Research-Planning Sequence

Mertens directly challenges simplistic sequencing in mixed-methods debates by asking whether the paradigm or approach comes first. Her broader argument emphasizes that paradigms involve assumptions about ethics, reality, knowledge, and systematic inquiry and that researchers should understand how those assumptions relate to methodological decisions rather than treating quantitative, qualitative, or mixed methods themselves as paradigms.

The practical lesson is that research development can proceed through reciprocal adjustment.

The arrows between these elements should often run in both directions.

Paradigm First Can Be Appropriate

There are situations in which beginning explicitly with philosophy is sensible.

A researcher committed to a feminist epistemology may develop questions specifically concerned with how gendered power shapes knowledge and institutions. A critical theorist may begin from an explicit commitment to examining domination and ideology. A critical realist may design a research program around mechanism-based causal explanation.

In such cases, philosophy is not a label attached to an otherwise independent project. It helps generate the questions themselves.

Paradigm-first planning can therefore be intellectually coherent when the philosophical commitments are substantive rather than ceremonial.

Question First Can Also Be Appropriate

Many researchers, particularly those entering a new topic, begin with a substantive question.

This can be an advantage.

Asking what kind of answer the question requires can reveal the relevant philosophical assumptions more naturally than selecting a paradigm from a textbook table.

If your question asks what proportion of a population experiences something, you need a defensible account of measurement and population inference.

If it asks how people understand an experience, you need an account of meaning and interpretation.

If it asks what mechanisms generate an outcome, you need an account of causation.

The philosophical work emerges from examining what your knowledge claim requires.

Method First Is Riskier but Not Automatically Wrong

Sometimes researchers begin with a method because circumstances make it available.

You inherit a longitudinal dataset. You gain unusual access to an organization for ethnographic observation. A natural experiment occurs. A large-scale survey has already been administered.

Methodological opportunity can legitimately generate research questions.

The risk arises when researchers force questions onto the available method without considering what the evidence can actually establish.

If you begin with data or methods, philosophical reflection becomes especially important because you need to identify the assumptions and inferential limits already embedded in what is available.

Do Not Choose a Paradigm Merely to Justify a Method You Already Wanted

This is the opposite problem.

A researcher decides to use interviews because they seem easier to conduct, then chooses interpretivism because interviews are supposedly interpretivist.

Another decides to administer a survey, then declares positivism because surveys contain numbers.

A third decides to combine both and selects pragmatism because that completes the familiar methods table.

These are not strong philosophical arguments.

The paradigm should explain assumptions that genuinely inform the inquiry, not serve as retrospective paperwork for a preferred technique.

Paradigms Do Not Mechanically Determine Methods

Even if you choose your paradigm first, it will not necessarily tell you exactly which method to use.

The same research method can often be used under different paradigms, and a single paradigm can support several methodological possibilities.

An interpretivist study might use interviews, observations, documents, focus groups, or combinations of these. A pragmatic study can be quantitative, qualitative, or mixed. A critical realist study can likewise draw on multiple forms of evidence.

Philosophy constrains and informs methodological reasoning. It rarely functions as a method-selection algorithm.

The Research Question Does Not Mechanically Determine the Paradigm Either

Two researchers can sometimes ask closely related questions from different philosophical positions.

One may investigate institutional inequality through critical realism, seeking mechanisms that reproduce unequal outcomes. Another may approach the same problem through a transformative framework emphasizing power, marginalized knowledge, and social change.

Both positions might be defensible.

This is consistent with the broader principle that different research designs can sometimes answer the same research question. Research planning involves judgment among defensible alternatives rather than automatic deduction from one element to another.

The Goal Is Eventual Alignment

The fact that research planning is iterative does not mean everything can remain provisional forever.

By the time you justify the study, the relationships among question, philosophy, methodology, methods, analysis, and claims should be coherent.

This is the distinction between process and product.

The process may involve changing your mind, trying different formulations, revising methods, discovering philosophical tensions, and returning to the research question.

The final methodological argument should explain the position you reached and why it is defensible.

Think of Alignment as a Network, Not a Waterfall

No single row necessarily has to be completed permanently before the next begins.

How a Research Question Can Lead You Toward a Paradigm

The paradigm did not appear first chronologically. Yet the eventual relationship among the question, philosophy, and methodology is coherent.

Had the researcher retained the original predictive question, a postpositivist quantitative approach might also have been defensible. The research problem did not contain one philosophically predetermined solution waiting to be discovered.

Common Misunderstandings About When to Choose a Paradigm

Where Should You Start?

For many researchers, particularly those who are not working from an established philosophical tradition, the research problem and question provide the most useful starting point.

Then move repeatedly between the question, philosophical assumptions, methodology, and methods until the pieces make sense together.

The important endpoint is paradigm-methodology alignment. How you reached that endpoint can be iterative.

Your methodology chapter is an argument for the research you ultimately conducted. It does not need to pretend that the research design descended perfectly formed from ontology on Tuesday morning.

Before Finalizing Your Paradigm and Methods

Frequently Asked Questions About Choosing a Paradigm and Methods

Research Philosophy Should Guide Your Logic, Not Dictate a Chronological Ritual

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.