03 · What You Need to Know
The Core Components of a Research Protocol
Start with clear study identification and document control
A protocol should be identifiable as a specific document for a specific study. At minimum, this generally means a clear study title and enough identifying information to distinguish the protocol from other projects and later versions.
Where version control matters, include a version identifier and date. This becomes particularly important when a study undergoes review or amendments because collaborators and reviewers need to know which document is current. SPIRIT 2025, which applies specifically to randomized trial protocols, includes protocol version date and identifier among its recommended items.
Version information may seem administrative until two team members begin working from different protocols. Establishing a protocol version and change-tracking system early can prevent that surprisingly unglamorous research problem.
Explain why the study is being conducted
A protocol needs enough background to establish the scientific rationale for the research. This does not necessarily require reproducing an entire literature review. The relevant question is what readers need to understand the problem, the existing knowledge, the uncertainty being addressed, and the justification for the planned study.
The World Health Organization's recommended format for research protocols describes the rationale as explaining why the research needs to be done and its relevance in light of current knowledge. The protocol should connect that rationale directly to the study's goals and objectives.
State the research questions, objectives, and hypotheses clearly
The protocol should make clear what the study is designed to answer. Depending on the research tradition, this may be expressed through research questions, objectives, hypotheses, aims, or a combination of these.
Distinguish primary and secondary objectives where that distinction is methodologically meaningful. Avoid objectives so broad that several incompatible designs could supposedly satisfy them.
More importantly, these statements cannot remain isolated at the beginning of the document. The rest of the protocol should show how the planned design and methods will actually address them. Checking whether the protocol aligns with the research questions, objectives, and hypotheses can expose mismatches before the study begins.
Specify the study design and setting
Name and describe the design accurately enough that readers understand the basic structure of the study. Depending on the research, this might involve an experimental, quasi-experimental, observational, qualitative, mixed-methods, diagnostic, longitudinal, cross-sectional, case-control, cohort, or other design.
Labels alone are insufficient. If the design includes groups, conditions, repeated observations, intervention phases, comparison periods, multiple sites, or other structural features, describe how they operate.
The setting may also matter. Explain where the study will occur and any contextual characteristics necessary to interpret or reproduce the procedures.
Define who or what will be studied
For studies involving participants, specify the target population and relevant eligibility criteria. Explain how potential participants will be identified, approached, recruited, selected, or assigned as appropriate to the design.
For research that does not recruit human participants, the equivalent issue remains: what units, records, documents, specimens, observations, datasets, cases, sites, or other sources will enter the study, and according to what criteria?
The sampling plan should state how the sample will be obtained and, where applicable, how its intended size was determined. Quantitative studies may require a formal sample-size calculation or another justification. Qualitative studies may instead justify sampling in relation to the research purpose, information needs, and analytical approach.
Define what will be measured, observed, manipulated, or explored
Terms such as "engagement," "achievement," "AI use," "stress," or "research productivity" can sound precise while representing many different things. The protocol should explain how important constructs, variables, exposures, outcomes, interventions, or phenomena will be operationalized.
Identify the instruments, measures, data sources, observation procedures, interview guides, laboratory procedures, or other tools that will produce the evidence. Where relevant, document their origin, scoring, administration, timing, interpretation, and evidence supporting their use.
For intervention research, describe the intervention and comparator in enough detail to understand what participants or units will actually receive. For qualitative research, explain how data generation procedures such as interviews, focus groups, observations, or document collection will address the research questions.
Describe the data collection procedure chronologically
The protocol should allow readers to understand what happens from the beginning of participation or observation through the completion of data collection. Who performs each procedure? In what sequence? At what time points? Using which instruments? Under what conditions?
A chronological description can reveal gaps that are less obvious when methods are organized only by topic. For example, a protocol may identify a questionnaire without stating whether it is administered before or after an intervention, or describe several measurements without specifying whether every participant receives all of them.
This is where sampling, measurement, data collection, and analysis need to fit together. Each stage produces constraints for the next.
Explain how the data will be managed
Data management is easy to postpone because it feels like something that happens after collection. In reality, many data-management decisions affect collection itself.
Depending on the study, the protocol or an associated data-management document may need to address coding, data entry, validation, storage, access, confidentiality, identifiers, quality checks, handling of source records, retention, backup, and eventual sharing or disposal.
WHO guidance specifically expects research protocols to describe data handling and coding, monitoring and verification, as relevant to the study. More specialized requirements may apply in regulated or sensitive research environments.
Specify how the data will be analyzed
"The data will be analyzed statistically" is not an analysis plan. Neither is "thematic analysis will be used" sufficient if important analytical choices remain unexplained.
The protocol should connect each major research question or objective to an appropriate analytical strategy. For quantitative research, relevant details might include outcome definitions, analysis populations, statistical models or tests, covariates, handling of missing data, significance criteria, sensitivity analyses, or other design-specific procedures. Not every study requires all of these.
For qualitative research, specify the analytical approach with enough detail to explain how raw material will be transformed into findings. Depending on the methodology, this might involve coding procedures, category or theme development, case comparison, interpretive processes, researcher involvement, or procedures used to support analytical rigour.
Complex studies may use a separate statistical analysis plan or other specialized document. SPIRIT 2025 explicitly recognizes that related documents can provide additional detail while being referenced from the main protocol.
Address ethical issues rather than merely writing "ethics approval will be obtained"
For research involving humans or other ethically sensitive materials or data, the protocol should identify the ethical issues created by the study and explain how they will be addressed.
Depending on the research, this may include informed consent, privacy and confidentiality, potential benefits and harms, vulnerable populations, compensation, withdrawal, safety monitoring, handling of incidental findings, data protection, or post-study considerations. Applicable requirements depend on the study and jurisdiction.
The Council for International Organizations of Medical Sciences, in collaboration with WHO, provides a detailed list of protocol items for health-related research involving humans. Its guidance makes clear that ethical considerations are integrated throughout the study plan rather than confined to an approval statement.
Document roles, oversight, quality, and study management when relevant
A protocol may need to identify investigators, participating institutions, sponsors, funders, study sites, and the responsibilities of different members of the research team. Larger or higher-risk studies can require substantially more information about monitoring, quality assurance, committees, safety oversight, and decision authority.
For a small student project, some of these arrangements may be simple. For a multicentre trial, they can be extensive. The principle is proportionality: document the governance and operational arrangements needed for the actual study rather than reproducing infrastructure that does not exist.
Include dissemination and other end-of-study plans when they matter
A protocol should look beyond data collection when later decisions are part of responsible study conduct. WHO guidance includes expected outcomes, dissemination of results, publication policy, duration, anticipated problems, project management, and related matters in its recommended protocol format.
Depending on the study, researchers may also need to specify how participants or communities will receive findings, where data or materials may be shared, how authorship or publication will be managed, and what happens to records or specimens after the project ends.
Watch Out
Do not copy every item from a clinical trial protocol checklist into an unrelated study. SPIRIT 2025, for example, provides 34 minimum items for randomized trial protocols. It is authoritative for that purpose, but it is not a universal template for every qualitative, observational, laboratory, educational, or other research design.
The protocol may point to associated documents rather than contain everything itself
A complete study record can extend beyond one file. Questionnaires, interview guides, consent forms, recruitment materials, statistical analysis plans, data-management plans, standard operating procedures, intervention manuals, and other supporting documents may contain details that would be cumbersome or inappropriate to reproduce in full within the main protocol.
The protocol should identify and reference important associated documents where appropriate. The aim is not to force every detail into a single enormous document. It is to ensure that important methodological information is documented, accessible to the people who need it, and consistent across the study materials.
