How to Align a Dissertation Question, Methodology and Analysis
A clear way to connect aims, research questions, philosophy, design, sampling, data collection and analysis without creating a methods chapter full of disconnected labels.
Alignment is the main quality test
A strong methodology chapter is not a list of research terms. Each decision should follow logically from the problem and research question. If the question asks about measurable relationships, a quantitative design may fit. If it asks how people understand or experience a phenomenon, qualitative methods may be more appropriate. Mixed methods is justified when both forms of evidence are genuinely needed—not because using two methods sounds more advanced.
Move from aim to data requirements
Write the aim and each research question, then ask what evidence would answer it. This reveals whether you need numerical measures, interview accounts, documents, observations, secondary datasets or a combination. The type of evidence then constrains the design, sampling and analysis choices.
Use philosophy to explain assumptions, not decorate the chapter
Terms such as positivism, interpretivism and pragmatism should explain assumptions about knowledge and inquiry. They should not be inserted as isolated definitions. A quantitative hypothesis-testing study may be consistent with a post-positivist or positivist orientation, while an interpretive interview study may emphasise socially constructed meaning. The wording should match the actual design.
Sampling must fit the population and purpose
Probability sampling supports statistical inference when a suitable sampling frame exists. Purposive sampling may be appropriate when participants need particular experience or characteristics. Convenience sampling is often practical but has limitations. Describe who is eligible, how they are approached, why the sample is suitable and what biases may remain.
Analysis should answer the research questions
The analysis plan belongs in the methodology because it is part of the design. Quantitative research should identify variables, coding, descriptive statistics and inferential methods. Qualitative research should explain coding, theme development and how interpretation will be documented. Avoid promising sophisticated analysis that the data cannot support.
Ethics and quality run through the whole method
Consent, confidentiality, data protection, researcher reflexivity, validity, reliability, trustworthiness and bias should be addressed where relevant. They are not an appendix to the design. The final methodology should show that the research question, evidence, procedure and analysis form one coherent chain.