Choosing Your Research Philosophy

Your research philosophy section often feels like the most abstract part of a proposal — but it's not a formality. It's the logical foundation your design, methods, and even your research questions are supposed to follow from, and committees can tell when it's been added after the fact rather than actually driving the study.

Positivism

Assumes an objective reality that can be measured and studied independently of the researcher. Associated with quantitative methods, hypothesis testing, and the goal of generalizable, replicable findings. If your study measures variables, tests hypotheses statistically, and aims to generalize to a wider population, positivism (or post-positivism, its more nuanced variant acknowledging that measurement is never perfectly objective) is usually the natural fit.

Interpretivism (constructivism)

Assumes reality is socially constructed and best understood through the subjective meanings people assign to their experiences. Associated with qualitative methods — interviews, focus groups, ethnography — where the goal is depth of understanding in context, not generalization. If your study explores lived experience, meaning-making, or a phenomenon best understood through participants' own words, interpretivism fits.

Pragmatism

Rejects the idea that you must choose one worldview and instead focuses on what approach best answers the actual research question — commonly the philosophical basis cited for mixed-methods designs. If your study combines quantitative and qualitative components (e.g., a survey followed by interviews to explain the survey results), pragmatism is usually the philosophy your committee will expect you to name and justify.

How to actually justify your choice

Don't just define the philosophy — connect it explicitly to your research question and design. A weak version states "this study adopts a positivist philosophy." A strong version explains that because the research question asks whether an intervention produces a measurable, generalizable effect, a positivist assumption of an objective, measurable reality is the appropriate foundation for the quantitative, hypothesis-testing design that follows.