Your Methodology Choice Is Not a Formality. It’s an Argument.

Research design choices are one of the most consequential decisions you’ll make in any study, and yet most researchers treat them like a box to tick. Pick a method, justify it in two sentences, move on. However, your methodology is not background noise. It’s a claim about how knowledge works, and reviewers can tell when you haven’t really thought it through. Syengo & Wanyama, writing in the Journal of Research and Academic Writing (2026), identify three persistent problems in social science research: inadequate methodological training, weak integration in mixed methods studies, and what they call philosophical incoherence in research design. That last one is the quiet killer of otherwise solid papers.

Making Research Design Choices That Actually Hold Up

Philosophical incoherence sounds alarming, but in practice it just means your method and your worldview don’t match. For example, you might claim a constructivist stance and then run a survey with a five-point Likert scale, treating the results as objective facts. That’s a contradiction. Reviewers in qualitative-heavy fields will catch it immediately. Therefore, before you finalise your design, write one paragraph that connects your ontological position to your method. Not for the paper itself, necessarily, but for you. If you can’t write that paragraph, your methodology section will feel hollow.

Here’s a concrete thing to do right now. Open Creswell’s Research Design (any edition) and read the table that maps worldviews to methods. It takes twenty minutes. After that, ask yourself honestly: does my chosen method fit the claims I’m actually making? If the answer is no, or even maybe, you have a problem worth solving before submission, not after.

Also, stop choosing your method because it’s what your supervisor used, or because it’s what your department does. That’s how you end up defending a regression model when your research question is fundamentally about lived experience.

When Mixed Methods Is the Right Call (and When It Isn’t)

Mixed methods research is fashionable right now, and for good reason. However, Syengo & Wanyama flag that weak integration is endemic. Many studies collect qualitative and quantitative data separately, report them separately, and then write a discussion that gestures vaguely at both. That’s not integration. That’s two papers stapled together.

If you’re going to use a mixed methods design, decide upfront which strand is dominant and why. Is the qualitative component explaining the quantitative findings? Is the survey validating themes from interviews? Your integration point should appear in your research questions, not just your discussion. In addition, be specific about the sequence. Sequential explanatory, sequential exploratory, convergent parallel design: these aren’t just jargon. They tell the reader exactly how your data sources will talk to each other.

If you can’t explain the integration in one clear sentence, you probably don’t have a mixed methods study. You have a muddle.

The Peer Review Test You Should Run Before Submission

Here’s an opinion that might sting a little. Most methodology sections are written to satisfy a reviewer, not to genuinely justify a design. As a result, they’re vague, defensive, and full of citations that the author hasn’t really engaged with.

Try this instead. Before submission, ask a colleague from a different methodological tradition to read only your methodology section. A quantitative researcher reading your ethnographic design, or vice versa. Their discomfort will show you exactly where your justification is thin. This is more useful than any checklist.

Also, if you’re using a less common approach, such as a partially randomised preference trial or a qualitative longitudinal design, name it precisely and cite a methodological paper that defends it. Reviewers are more confident when you show you know the literature on your own method, not just the literature in your field.

Research design choices shape everything downstream: your data, your analysis, your claims, your contribution. Get them right at the start and the rest of the paper has a fighting chance.

So here’s the challenge: write one sentence that describes your methodology and one sentence that explains why no other approach would have answered your research question as well. If you can’t write the second sentence, you’re not done yet.

Image: Photo by Wonderlane on Unsplash

Explore more insights at Research Partners.