Design from the question outward
- State the approach and research design in field-appropriate terms.
- Define the population, cases, texts, system or materials and justify the selection.
- Describe the sample or corpus and how it was obtained; give a reason for the chosen size or scope.
- Explain instruments, data sources, apparatus, software or procedures in enough detail to understand what was done.
- Set out the analysis steps and show how each answers an objective or question.
- Report ethics, consent, privacy, permissions, safety and data management requirements that apply.
Give the reason for each major choice, not just its name. “A questionnaire was used” does not explain why it fits the question, what it measured, how it was administered or how missing responses were handled.
Worked example: justify a feasible method
The study adopted a survey research design. A questionnaire was used and data were analysed with SPSS.
The question asks how a specified group rates access to a defined service, so a cross-sectional survey can describe reported experiences during the chosen period. The target group and inclusion rules are stated; recruitment and consent procedures are explained; each questionnaire section is mapped to an objective; pilot changes and missing-response handling are recorded; the analysis describes the summaries used and the limits of self-reported data. The researcher still needs to supply real study details rather than copy this example.
Match analysis to evidence
For qualitative work, explain how material was selected, prepared, coded or interpreted and how you kept an audit trail. For quantitative work, define variables and measures, identify the planned statistics, and check assumptions relevant to the chosen test. For build or design projects, describe requirements, architecture, implementation decisions, test cases and evaluation criteria. Mixed methods need a clear explanation of how the strands connect.
Frequent methodology errors
- Naming a design without explaining why it answers the question.
- Claiming random sampling when participants were selected by availability.
- Reporting a sample-size formula without its inputs or assumptions.
- Choosing statistical tests before defining variables and checking assumptions.
- Omitting permissions, consent, exclusion rules, instrument limitations or missing data.
- Describing software as if it were the analytical method.
Methodology review checklist
- Could a reader trace each objective to a source of evidence and an analysis step?
- Are setting, population, selection and study period explicit?
- Are instruments or procedures described accurately enough to judge?
- Are ethics and safety requirements addressed before data collection?
- Do limitations match the method's actual reach?
Sources and further reading
These university writing resources inform the general advice here. They do not replace your department's handbook or supervisor's guidance.