A useful methodology review tests whether the question, design, sampling, measurement and analysis form one coherent plan. It should produce actionable revision notes—not replace the researcher's authorship or promise approval.
1. Check the chain from problem to analysis
The problem statement should support a focused question; the objectives should be answerable; variables should represent the concepts; and each planned analysis should map back to an objective. Gaps in this chain often become expensive after data collection.
A reviewer should identify whether the claims implied by the objectives are compatible with the design. Cross-sectional data cannot automatically support causal language or temporal conclusions.
2. Review population, sampling and sample size
Define the source population, study population, inclusion and exclusion rules, sampling frame and selection steps. The sample-size method must match the primary objective, anticipated analysis and sampling design.
A formula alone is insufficient. Document confidence, precision, effect or prevalence source, design effect, cluster count, non-response and practical recruitment path as applicable.
3. Test variables and the instrument
Create an objectives-to-variables table with operational definitions, units, response options, timing and source. Review questionnaire wording, skip logic, burden and whether sensitive fields are necessary.
Validated instruments still need context, licensing and scoring reviewed. A pilot should test comprehension and workflow, not merely confirm that the form opens.
4. Make the analysis plan specific enough to audit
State how variables will be coded, summarised and compared, how missing data will be handled and which assumptions will be checked. Name the method because it fits the outcome and design, not because software offers it.
Include effect estimates and confidence intervals alongside hypothesis tests. Pre-specify key analyses and distinguish exploratory work to reduce outcome-driven method changes.
5. Close feasibility, ethics and revision ownership
Review consent, privacy, data minimisation, access, transfer, retention and dissemination. Feasibility includes recruitment time, staff, platform, language, budget and the ability to analyse what will be collected.
The deliverable should be a prioritised comment matrix or reviewed document explaining each concern and recommended change. The researcher remains responsible for authorship, institutional requirements and ethics approval.
- Question-design alignment
- Defensible sampling path
- Operational variable table
- Specific analysis plan
- Ethics and feasibility
Research proposal and study-design support
We can review your actual form, dataset and reporting requirement, then scope the smallest service that solves the risk.
Explore the service ↗