ACADEMIC WRITING GUIDE

How to Write a Dissertation Methodology Chapter (With Worked Examples)

By Acadelyra Editorial TeamPublished

A dissertation methodology chapter explains how a study was designed and carried out—or, in a proposal, how it will be carried out—and why its methods are appropriate for the research questions. It is more than a list of interviews, questionnaires, or statistical tests. A useful methodology chapter enables readers to understand how evidence was obtained, evaluate the strength of the design, and judge which conclusions the study can reasonably support.

The difficult part is often alignment. Your research question, overall approach, sampling decisions, data collection procedures, analysis, and ethical safeguards must fit together. This guide walks through those decisions and provides three substantial, hypothetical worked examples: a qualitative interview study, a quantitative correlational study, and an explanatory sequential mixed-methods study.

Important: The examples are teaching illustrations, not accounts of completed studies. Their participants, procedures, and planned analyses must not be copied into a dissertation as if they describe research that actually occurred.

What is a dissertation methodology chapter?

The methodology chapter documents the logic and practical procedures of an investigation. It typically explains the study's research approach and design, setting or sources of evidence, population and sample (if applicable), data collection, analysis, measures to support research quality, ethical safeguards, and methodological limitations.

A useful distinction is that methodology concerns the broader rationale and organising approach, whereas methods are particular techniques used to generate or analyse evidence. For example, semi-structured interviews are a method. Explaining why interviews are appropriate for understanding students' experiences, how participants are selected, and how interview accounts will be interpreted is methodological reasoning.

Requirements differ across disciplines. A laboratory dissertation may need precise materials and experimental procedures; a historical dissertation may discuss archival selection and source criticism rather than participant recruitment. Follow your department's chapter requirements rather than treating any template as universal.

Dissertation methodology chapter structure

Dissertation methodology chapter structure
SectionWhat a reader needs to know
IntroductionWhat research question and methodological choices will this chapter address?
Approach and designWhy is this overall approach suited to the study?
Setting, sources, and samplingWho or what provides evidence, and why were those cases selected?
Instruments and collectionWhat was collected, using which procedures and measures?
AnalysisHow were the data prepared, examined, and interpreted?
Quality and ethicsHow were important threats to research quality and participant protections addressed?
LimitationsWhat are the design's meaningful boundaries?
Chapter summaryHow do the decisions work together to address the research question?

Depending on the institution, research philosophy or positionality may appear as a separate section; mixed-methods studies should explicitly describe the integration of quantitative and qualitative components.

Step 1: Start with the research question, not the tool

A frequent mistake is deciding to use an online survey or interview first, then attempting to make a research question fit the chosen tool. Reverse that process. Identify what the question asks and what evidence could answer it.

Consider two hypothetical questions about student learning:

  • Quantitative question: Is self-reported weekly independent study time associated with examination scores among first-year undergraduates?
  • Qualitative question: How do first-year undergraduates describe the challenges of establishing independent study habits?

The first asks about a measurable relationship. The second asks for accounts of experiences and meaning. Even though the topic is similar, the required evidence and possible conclusions are different.

State your research question and objectives early in the chapter, then map each objective to its relevant data source and analysis. If your question is still too broad, our guide on developing a strong research question explains how to narrow it.

Step 2: Justify the research approach and design

Three broad approaches frequently appear in empirical dissertations:

  • Quantitative: Uses numerical measures to examine distributions, relationships, differences, or effects, as permitted by the design.
  • Qualitative: Explores meanings, perspectives, interactions, experiences, or processes through detailed contextual evidence.
  • Mixed methods: Combines quantitative and qualitative components for a defined reason and explains how their evidence will be integrated.

Then name the research design—the organising plan for the investigation. Possibilities include cross-sectional surveys, experiments, case studies, phenomenological research, or explanatory sequential mixed methods. A design is not justified merely by saying that it is "suitable."

For example, compare these statements:

Too vague: "A correlational design was used because it was appropriate for the topic."

Better: "A cross-sectional correlational design was selected to investigate whether reported weekly independent study hours were associated with examination performance during one academic period. The study observes existing differences rather than assigning study-time conditions; therefore, the analysis can estimate an association but cannot demonstrate that additional study time caused a change in examination scores."

The stronger version makes the relationship between the question, timing, design, and limitation explicit. For additional help selecting an approach, read how to choose the right research methodology.

Some departments require a research paradigm or philosophical position. Where relevant, explain how assumptions about knowledge, evidence, and interpretation informed your design. Avoid adding lengthy generic definitions of philosophical labels that do not affect the actual investigation.

Step 3: Explain the setting, population, sample, and recruitment

Define the people, organisations, documents, observations, or datasets from which evidence will be drawn. For participant-based studies, distinguish the target population from the accessible population or sampling frame. Students at one institution, for example, do not automatically represent students in every institution.

Explain eligibility, exclusions, recruitment, selection strategy, planned or actual sample size, and any likely selection bias. For probability-based quantitative samples, sample size may involve precision, power, and anticipated nonresponse. For qualitative studies, sample adequacy instead depends on the research aims, design, heterogeneity, and depth of information. Neither a large sample nor a familiar numerical rule guarantees credibility.

Illustration: A researcher studying experiences of digital lecturer feedback could purposively recruit students who recently received that feedback. That is a defensible choice for exploring experiences, but it does not make the group statistically representative of all university students. See our sampling method and sample size guide for a fuller explanation.

Step 4: Describe instruments and collection procedures precisely

Readers should be able to follow the sequence of data collection without having to guess what happened. Include the instrument or source, variables or topics covered, administration procedure, timing, permissions, and relevant safeguards.

For a survey, specify how the main variables are operationalised, the response scales, any adapted instruments, and how the questionnaire was administered. For interviews, describe the interview format, major topics, consent to recording, and approximate duration where relevant. For documentary or secondary-data research, identify source repositories, inclusion rules, access permissions, time span, and extraction methods.

Weak: "Data were collected using questionnaires."

Stronger proposed procedure: "After institutional permissions and ethics approval are obtained, eligible students will receive an invitation containing study information and a voluntary consent process. Consenting participants will complete an online questionnaire measuring their reported weekly independent study time during a defined reference period. Responses will be assigned study identifiers, reviewed for incomplete records, and stored in a restricted-access location before analysis."

In a completed dissertation, replace future-tense plans with an accurate account of what took place. Record any approved or justified deviations from the plan instead of silently rewriting history.

Step 5: Explain the analysis—not just the software

The analysis section must say how evidence was turned into findings that answer the research questions. A software name alone is not an analysis plan.

For a quantitative study, describe variable construction, screening, handling of missing values, descriptive summaries, the intended statistical model or test, assumptions, covariates where justified, and how uncertainty and effect magnitude will be reported. An observational association should not be described as a causal effect.

For a qualitative study, specify the analytic tradition and how it will be applied. For instance, reflexive thematic analysis may involve close familiarisation with data, coding, development and review of interpretive themes, and reflective engagement with the researcher's assumptions. Do not mechanically claim that every qualitative approach requires inter-coder agreement or member checking; explain why your quality practices suit your approach.

For a mixed-methods study, specify when the components occur, why one informs the other, and where evidence is brought together—such as interview sampling based on survey results or a joint display comparing numerical patterns with narrative explanations.

Our guide to writing a dissertation data analysis plan explores these decisions with more examples.

Step 6: Explain research quality and ethics

The words validity, reliability, credibility, and trustworthiness are not interchangeable guarantees. Identify the criteria appropriate to your design and what you did—or propose to do—to address them.

Quantitative studies may need to discuss measurement error, relevant validity evidence, missing data, selection effects, confounding, and model assumptions. Qualitative studies may need to explain reflexivity, transparent analytic decisions, contextual description, and coherent links between extracts and interpretations. For mixed methods, explain the quality of each component and the reasoning behind their integration.

Ethical safeguards might include required institutional review, informed consent, voluntary participation, safeguarding, confidentiality, data minimisation, secure storage, controlled access, and lawful or institutionally approved retention and disposal. Be precise about what consent and confidentiality can realistically provide. Do not claim approval was obtained unless it was, and do not imply that anonymisation can always be guaranteed.

Step 7: State limitations in terms of the conclusions they affect

A limitation matters because it changes what readers should infer. A convenience sample from one university restricts claims about a wider population; self-reported study hours may be misremembered; a cross-sectional correlation does not establish the direction of causality; and interview participants' accounts cannot estimate how common a view is among all students.

Describe important limitations, relevant mitigation, and the boundaries of your conclusions. Avoid both exaggerating a weakness and dismissing it with an unsupported claim that it "did not affect the study."


Worked example 1: Qualitative methodology chapter passage

Hypothetical question: How do first-year university students describe using lecturers' written feedback to improve subsequent assignments?

Model passage (proposed study)

This study will employ a qualitative design using semi-structured individual interviews to explore first-year undergraduates' interpretations of lecturers' written assignment feedback. A qualitative approach is justified because the research question asks how students understand feedback and describe their responses to it, rather than seeking a population estimate of feedback effectiveness.

Participants will be purposively recruited from first-year undergraduates at the selected institution who have received written feedback on at least one graded assignment. An initial recruitment range of approximately 12–18 interviews is proposed as a planning estimate, not a fixed rule; the researcher will review sample adequacy in light of the analytic purpose and richness and variation of accounts, within the approved study protocol. The recruitment method, eligibility criteria, and participant characteristics will be documented.

After appropriate approval and informed consent, interviews will follow a topic guide covering participants' understanding of comments, examples of changes attempted in later assignments, and obstacles to applying feedback. Where participants consent, interviews will be audio-recorded and transcribed. Direct identifiers will be removed when feasible and data access restricted.

Transcripts will be analysed using reflexive thematic analysis. The researcher will become familiar with each account, code material relevant to the research question, develop and refine themes across interviews, and use reflexive notes to examine interpretive assumptions. Selected de-identified quotations will support the presentation of themes. Accounts of improvement will be interpreted as participants' reported experiences, not independent proof that feedback improved grades.

Required ethics permissions will be obtained before recruitment. Participation will be voluntary and will not influence students' academic assessment. The study will explain practical limits to confidentiality and will follow the institution's approved data-retention arrangements.

Why it works

The approach matches a question about interpretations and experiences. Sampling deliberately identifies students who have encountered the relevant phenomenon. The analytic method explains how interview material will become a reasoned interpretation, while the final sentences prevent an unjustified causal conclusion. The proposed sample range is explicitly illustrative and requires contextual justification.

Worked example 2: Quantitative methodology chapter passage

Hypothetical question: Is self-reported weekly independent study time associated with examination performance among first-year undergraduates at one university?

Model passage (proposed study)

A quantitative cross-sectional correlational design will be used to examine the relationship between self-reported weekly independent study time and examination scores among first-year undergraduates. The design permits estimation of an association within the study population but does not involve assigning students to different amounts of study time.

Eligible participants will be recruited through institutionally approved channels. The accessible population, sampling frame, response rate, and departures from planned recruitment will be documented. The intended sample size will be justified through an appropriate power or precision calculation using explicitly stated assumptions and realistic expectations about response, rather than choosing an arbitrary number.

Students who consent will report their typical independent study hours for a defined period in a structured questionnaire. Subject to separate institutional authorisation and participant permissions where required, examination outcomes will be linked from administrative records using controlled study identifiers. The questionnaire's study-time measure, data source for scores, and rules for unusable or missing records will be documented.

The analysis will screen for duplicate or implausible records, summarise the distributions of the variables, and estimate the association using a method appropriate to the observed measurements and assumptions. Any adjustment variables will be justified in relation to the research question and plausible confounding. Results will report estimated effects and uncertainty, not only statistical significance. Missing-data decisions and deviations from the analysis plan will be disclosed.

Because study time is self-reported and the design is observational, recall error, selection effects, and unmeasured confounding may influence the association. Findings will not be interpreted as evidence that increasing study hours necessarily causes higher scores.

Why it works

The passage names the measurable relationship, clarifies the observational design, shows how variables would be obtained, and describes a transparent analysis strategy. It does not pretend that a specific sample size, statistical result, or institutional approval already exists. The limitation is directly tied to the conclusions the researcher can make.

Worked example 3: Mixed-methods methodology chapter passage

Hypothetical question: How satisfied are first-year university students with digital academic feedback, and how do they explain their satisfaction or dissatisfaction?

Model passage (proposed study)

This study will adopt an explanatory sequential mixed-methods design. First, a quantitative survey will describe patterns of satisfaction with digital academic feedback. Second, semi-structured interviews will investigate how selected students explain experiences associated with those patterns. The sequential design is chosen because the interview phase can provide context for survey findings that numerical responses alone cannot explain.

In the quantitative phase, eligible first-year students will complete a questionnaire with clearly defined satisfaction measures and relevant characteristics of their feedback experiences. Recruitment, instrument quality, missing-data handling, and planned descriptive and inferential analyses, where warranted, will be specified in the protocol.

After preliminary survey analysis, consenting respondents will be purposively invited for interviews to represent contrasting satisfaction experiences—for example, relatively high and low ratings. Interview questions will build on survey findings while allowing participants to describe issues not anticipated by the questionnaire. The researcher will explain any sampling restrictions introduced by participation in both phases.

Survey data will be analysed according to the stated measurement levels and research questions. Interview transcripts will be analysed through a defined qualitative approach. Integration will occur when the researcher compares themes with survey patterns, including a joint display noting where accounts elaborate, challenge, or complicate the quantitative findings. Differences between components will be reported rather than forced into agreement.

Required institutional permissions and ethical approval will be secured before recruitment. Participants will consent separately to relevant activities and any linkage between survey and interview records. Because neither the observational survey nor the interviews experimentally manipulate feedback conditions, the integrated findings will be interpreted as descriptive and explanatory insights rather than proof of causation.

Why it works

Both components have a job: the survey identifies patterns, while the interviews investigate possible explanations. The second phase is connected to the first through participant selection, and the integration step explains what makes this genuinely mixed-methods research rather than two disconnected data collections. The passage also recognises the risks of over-interpreting observational findings.

A practical formula for writing methodology paragraphs

A strong methodology paragraph often moves through decision → reason → procedure → safeguard or limitation.

Example: "A cross-sectional survey design was selected to examine the association between students' study habits and examination scores. This design was suitable because the question concerns measured differences within a defined academic period rather than the effect of an intervention. Eligible participants completed a structured study-habits questionnaire, and authorised examination scores were analysed using a method appropriate to the observed variables. Because the data were observational, the findings were interpreted as associations rather than evidence of causation."

The paragraph answers not only what was done, but why, how, and what can be inferred.

Common mistakes to avoid

Listing methods without explaining the choices. A design is not justified by repeating that it was "appropriate." Connect it to the question and evidence needed.

Reporting plans as completed facts. Proposal-stage writing should say what will happen. Completed research should describe what actually happened, including relevant changes from the plan.

Calling a convenience sample representative. A bigger sample cannot by itself repair selection bias or establish population generalisability.

Treating software as an analytic method. State the actual statistical or qualitative analysis; identify software only as a supporting tool where relevant.

Assuming all qualitative research follows the same quality checklist. Justify techniques according to the selected qualitative tradition.

Combining quantitative and qualitative data without integrating them. A mixed-methods design needs a reason and a point of connection between components.

Making stronger claims than the design supports. Correlation does not demonstrate causation, and participant experiences are not population prevalence estimates.

Claiming approvals or procedures that did not occur. Accurate reporting is essential to academic integrity.

Checklist before submitting your chapter

  • Does the design clearly follow the research questions and objectives?
  • Are approach, design, and methods distinguished and justified?
  • Are setting, sampling, recruitment, and eligibility adequately explained?
  • Can the reader follow how data were measured or generated?
  • Is the analysis appropriate for each question and type of evidence?
  • Are quality checks and relevant ethics safeguards concrete rather than generic?
  • Do limitations specify what cannot confidently be concluded?
  • Are hypothetical plans, proposed procedures, and completed work clearly distinguished?
  • Does the chapter meet the institution's specific dissertation requirements?

Frequently asked questions

How long should a dissertation methodology chapter be?

There is no universal length. A complex mixed-methods project may require much more procedural explanation than a small secondary-data study. Prioritise the detail necessary to understand and evaluate the design, and follow your university's expectations.

Can I write the methodology before collecting data?

Yes. In a proposal, explain the planned methods and why they are suitable. In the completed dissertation, revise the account to reflect the procedures actually used and disclose meaningful deviations from the original plan.

Does a methodology chapter need references?

Usually, yes. Cite authoritative methodological sources when explaining established designs, instruments, analytic approaches, or reporting standards. Citations support your rationale but cannot replace a precise account of your own procedures.

Is methodology the same as results?

No. The methodology chapter explains how evidence was collected and analysed; the results or findings chapter presents what that process revealed. Some disciplines organise these elements differently, so check institutional conventions.

Do I need a research philosophy section?

Only where required or substantively relevant. If included, connect philosophical assumptions to actual decisions about evidence, interpretation, and design instead of offering unrelated definitions.

Final thoughts

A convincing dissertation methodology chapter is a connected explanation of research choices—not a catalogue of technical terminology. Start with your question, explain why each major method fits, document what happened or what is proposed, and make the limits of the evidence clear.

If you need help reviewing your chapter's structure, clarity, methodological explanations, or consistency with your assignment brief, explore Acadelyra's academic-support services. Your research decisions, data, and academic responsibilities remain your own.

Authoritative reporting resources

These are useful reporting resources, not replacements for your dissertation handbook or supervisor's instructions.