You have spent months (possibly years) building something you believe in. The literature review is thorough. The data is rich. The argument holds together. And yet, the moment a committee member leans forward and asks, “Can you walk me through your methodological rationale?”, something shifts. A quiet uncertainty creeps in. Because brilliant research and defensible research are not always the same thing and the gap between the two is almost always found in the methodology.
Methodology mistakes that can damage your thesis rarely look like disasters at the drafting stage. They look like reasonable choices, convenient shortcuts, or perfectly logical decisions made under pressure. That is precisely why they are so dangerous. They survive the writing process and surface only when someone with expertise sits across from you and starts asking questions you did not prepare for.
What follows is not a checklist. It is an honest, practitioner-level look at three methodological patterns I have seen repeatedly across hundreds of students in education, psychology, social sciences, and medicine, patterns that quietly undermine otherwise strong theses. If any of them sound familiar, that recognition alone is worth something.
Before you read on: if you are heading into submission in the next six months, consider running the free ‘Is Your Method Actually Defensible?’ diagnostic at the end of this article. It takes under five minutes and will tell you exactly where your risks sit.
Mistake One: Choosing a Method Because It Felt Right, Not Because It Was Justified
There is a version of this conversation I have had more times than I can count. A student tells me they chose thematic analysis because it seemed like a good fit for their data. Or they opted for SEM because their supervisor used it in a previous study. Or they went with a mixed-methods design because it felt more comprehensive. In each case, the method was selected before the research question had fully crystallised and that sequence is where the problem begins.
Method selection in research is not an aesthetic preference. It is a logical consequence of your ontological and epistemological positioning. Your philosophical framework determines what kind of knowledge you believe it is possible to produce. That, in turn, determines what kind of inquiry is appropriate. And that determines your method. When the method is chosen first and the justification is constructed backwards, the seams show, not to you, but to an experienced examiner.
The most common version of this mistake in qualitative work is the researcher who selects reflexive thematic analysis (correctly, by the way, it is a versatile and rigorous approach) but then applies it in a way that is closer to content analysis. They count themes. They treat them as frequencies. They move through the data looking for patterns to report rather than meanings to interpret. The method label is correct; the epistemological practice is not. Braun and Clarke have written extensively about this (the method requires a constructivist orientation that cannot be bolted on at the write-up stage). In quantitative work, the equivalent error is applying Structural Equation Modelling to a dataset and a research question that simply do not warrant it. SEM is a powerful tool for testing theoretically-derived models with latent variables and complex causal pathways. It is not a sophisticated form of regression that you use to appear more rigorous. When the model-data fit is poor and the sample size is borderline and the theoretical grounding is thin, no amount of presentational confidence will make the analysis defensible under scrutiny.
The fix is not to change your method. It is to go back to your epistemology and make the justification explicit, traceable, and coherent, so that every methodological decision flows visibly from something deeper than preference.
Mistake Two: Treating Rigour as a Section Rather Than a Practice
Almost every methodology chapter contains a section on rigour. In qualitative research it might be framed around Lincoln and Guba’s trustworthiness criteria (credibility, transferability, dependability, confirmability). In quantitative work it typically addresses reliability, validity, and generalisability. In both cases, the section is almost always written as a retrospective declaration: here is what I did to ensure quality.
The problem is not with the section. The problem is when rigour lives only in that section and nowhere else in the thesis. Because rigour is not a paragraph you write. It is a set of practices you embed throughout the entire research process and when it has not been embedded, the paragraph describing it reads as precisely what it is: a claim without evidence.
In qualitative work, I see this most acutely with researcher reflexivity. A student will write a thoughtful positionality statement in their methodology chapter and then disappear entirely from their own analysis. There is no acknowledgement of how their background shaped what they noticed. There is no audit trail of how their interpretations evolved. There is no evidence of the analytical rigour that reflexive thematic analysis, IPA, or grounded theory actually demands. The statement says, ‘I am aware of my positionality.’ The analysis says, ‘I proceeded as though I were invisible.’ In quantitative work, the equivalent issue appears in measurement. A researcher builds a survey instrument, runs a factor analysis, reports acceptable Cronbach’s alpha values, and then treats the instrument as validated. But internal consistency is not validity. A scale can be internally consistent and still be measuring the wrong construct entirely. Without evidence of construct validity (convergent, discriminant, or criterion-related) the measurement model is assumed rather than demonstrated. And assumptions are precisely what examiners are paid to interrogate.
Rigour has to be visible in your decisions, not just declared in your prose. If you can point to a specific methodological choice and explain how it protected the integrity of your findings, that is rigour. If you can only point to a paragraph that says you took it seriously, that is not enough.
Is your methodology chapter doing this? The Research Strategy Audit will map your methodological architecture and identify exactly where your rigour claims are unsupported. Find it at Payhip.
Mistake Three: Confusing Limitations with Weaknesses
Every research methodology has limitations. This is not a problem, it is a feature of honest scientific inquiry. The difficulty arises when students treat their limitations section as a form of preemptive apology rather than a demonstration of methodological awareness. There is an enormous difference between a researcher who understands why their design has inherent constraints and one who is simply listing things they hope the examiners will overlook.
The distinction matters because limitations and weaknesses are not the same thing. A limitation is a boundary condition of your design, something that is inherent to the methodological approach you chose and that you can account for theoretically. A weakness is a flaw that could have been addressed and was not. When students conflate the two, they either underplay genuine flaws by framing them as unavoidable limitations, or they overplay legitimate boundaries by apologising for design features that are entirely appropriate to their context.
A purposive sample in qualitative research is not a weakness. It is a methodological decision that reflects the epistemological assumption that meaning is context-specific and that depth is more valuable than breadth for your particular research question. It becomes a weakness only if you have claimed generalisability you cannot support, or if your sampling strategy is inconsistent with your stated approach. Framing it as “a limitation due to small sample size” misrepresents the design and signals to the examiner that you do not fully understand why you made the choices you made. The same logic applies on the quantitative side. Cross-sectional design as a limitation is fine if you are studying a construct at a specific point in time and your claims are appropriately bounded. It becomes a weakness if you have used causal language throughout your discussion chapter (words like “effect”, “impact”, “lead to”) without the longitudinal or experimental design that causal inference actually requires. Examiners do not miss this. They flag it, they ask about it, and a student who has not thought it through carefully will struggle to respond with the confidence the moment requires.
The measure of a sophisticated researcher is not whether their study has limitations (it always does). It is whether they can articulate those limitations precisely, explain why the design was appropriate despite them, and describe clearly what future research would need to look like to address the boundaries they have identified.
Before You Submit: Run the Diagnostic
If you have read this far and found yourself mentally flagging sections of your own methodology, that instinct is worth following. The good news is that methodological problems identified before submission are fixable. The same issues identified by an examiner during your viva are significantly harder to recover from (not impossible), but harder, and more stressful than they need to be.
The “Is Your Method Actually Defensible?” diagnostic is a short, structured set of questions designed to surface exactly the kinds of risks described in this article. It takes under five minutes, it is free, and it will give you an honest picture of where your methodology stands before the people who matter most have a chance to weigh in.
If the diagnostic reveals deeper structural concerns (or if you already know that your methodology chapter needs more than a quick review) the Research Strategy Audit is the next step. It is a comprehensive, personalised analysis of your research design, delivered asynchronously, with specific recommendations for every area of concern. No generic feedback. No templates. Just a clear, expert assessment of your specific study and what it needs.
Access the Research Strategy Audit in Payhip
One More Thing
Methodology is not the most glamorous part of a research. Nobody starts a doctorate, a bachelor or masters because they are passionate about epistemological positioning or measurement invariance. But it is the part of your thesis that determines whether everything else you have built can be trusted, by your examiners, by your field, and ultimately by you.
When you get to the other side of your viva, when you can say with complete confidence that your methodology held up under the hardest questions your committee could put to it, that moment is worth something specific. It is worth celebrating in a way that matches the weight of what you have accomplished. And the good news is that you do not have to figure that out alone. GetYourGuide connects you to thousands of curated experiences worldwide (the kind that turn a milestone into a memory). A cooking class where a local chef walks you through ingredients you have never heard of. A guided food tour through a neighbourhood market, led by someone who actually grew up there. An after-hours private tour of a world-class museum, one hour before anyone else walks through the door. A desert safari followed by a night under the stars in a luxury glass dome, watching the sky the way you never had time to when you were writing chapter four. A hot air balloon ride over a landscape you chose purely because it was beautiful and you had earned the right to go somewhere just for that reason. Whatever form it takes, you have spent years producing knowledge. Now go and collect some experiences.
Disclosure: As an Affiliate, I earn from qualifying purchases. Thanks for supporting this blog and helping me stay caffeinated in the post-grad world.
Leave a Reply