ACADEMIC WRITING GUIDE
How to Write a Systematic Review Protocol for a Dissertation (With a PRISMA-P Worked Example)
A dissertation systematic review can look rigorous at the end and still raise an uncomfortable question: Were the methods decided before the researcher knew which results they would produce? Without a recorded plan, seemingly small choices—excluding a borderline study, changing a primary outcome or switching from meta-analysis to narrative synthesis—can be difficult to distinguish from decisions made after seeing the evidence.
A systematic review protocol for a dissertation is that recorded plan. It sets out the question, eligibility criteria, search strategy, study-selection procedure, data extraction, appraisal and synthesis approach before the review's substantive decisions are made. It gives supervisors something concrete to review, helps research teams work consistently and leaves an audit trail if plans legitimately change. It is not the completed dissertation Methods chapter, a PRISMA flow diagram or simply a form to upload to a registry.
This guide explains what belongs in a protocol, how to use the PRISMA-P reporting checklist, where registration may be appropriate and how to write a realistic worked example without inventing research findings. The methodological starting points are the PRISMA-P 2015 guidance and Cochrane Handbook Chapter 1.
Teaching-example disclaimer: The topic, imaginary review team, prospective methods, illustrative search strings and hypothetical scenarios below are fictional. No literature search, screening, extraction or analysis has actually been undertaken for the example. Do not present its proposed methods as activities you performed, or its hypothetical records as real studies.
What is a systematic review protocol?
A protocol is a dated methodological specification for a planned evidence synthesis. It explains why the review matters and states how the researchers intend to identify, select, appraise and synthesize eligible evidence. A strong protocol is sufficiently specific that another qualified researcher could understand the planned decisions, while allowing transparent, documented amendments when unexpected methodological issues arise.
Protocols matter because systematic reviews are vulnerable to researcher discretion. Imagine discovering several contradictory studies halfway through screening. Without advance criteria, a reviewer might expand one population definition but not another, change outcome time points or adopt an analysis that happens to favor a preferred conclusion. A dated protocol provides a basis for asking whether a later change was methodologically justified and fully reported.
The PRISMA-P statement is a reporting guide for the protocol, not a certificate of methodological quality. Following its items makes planned methods more visible; it does not, by itself, guarantee that the chosen methods are suitable or that the review was carried out correctly.
A protocol also differs from a literature-review proposal. A general dissertation proposal may explain research questions, theory, feasibility and an overall timetable. A systematic review protocol must go further by predefining operational decisions about eligible studies, searches, screening, data items, risk of bias and synthesis. Institutions may combine the two documents, but neither should misrepresent the methods that will actually be used.
PRISMA-P versus PRISMA 2020: which one do you use?
These frameworks serve different stages of a project. PRISMA-P 2015 helps authors describe the planned systematic review. PRISMA 2020 guides transparent reporting of the completed review, including what was actually searched, screened, included and synthesised. https://pmc.ncbi.nlm.nih.gov/articles/PMC8005924/ provides the completed-review guideline.
| Question | Protocol / PRISMA-P | Completed review / PRISMA 2020 |
|---|---|---|
| When is it written? | Preferably before substantive study selection and data analysis | After review procedures and analyses are conducted |
| What does it describe? | Intended scope, methods, decisions and contingencies | Actual methods, results, deviations and limitations |
| Are findings presented? | No invented screening counts or effect estimates | Verified study counts, characteristics and findings |
| What happens if plans change? | Amend the protocol with dates and reasons | Disclose amendments and explain effects on interpretation |
| Main purpose | Make the planned methodology transparent and assessable | Make the completed review reproducible and interpretable |
For an example of the completed-review stage, see Acadelyra's guide to creating a PRISMA 2020 flow diagram for a dissertation. A PRISMA diagram reports selection results; it cannot replace a protocol written beforehand.
Before writing: confirm the question and review type
A protocol cannot repair an unfocused review question. Decide first whether the dissertation is an intervention-effect systematic review, qualitative evidence synthesis, scoping review, diagnostic review or another design. Each has different eligibility and synthesis requirements. Do not automatically apply trial-oriented risk-of-bias tools or effect-size methods to qualitative or scoping evidence.
For an intervention-effect review, PICO—Population, Intervention, Comparator, Outcomes—can help specify the question. Other frameworks may be more appropriate for questions about experience, prevalence or mapping a field. Explain the choice rather than naming a framework merely to appear systematic.
Consider this fictional review question:
Among first-year university students in credit-bearing undergraduate courses, how does structured written formative feedback from instructors, compared with usual feedback, affect course completion during the first academic term?
This question is intentionally narrow enough to support operational decisions. It does not assert that structured feedback works. The protocol will specify what counts as the intervention, how completion is measured and which designs can answer the question.
Before developing the final protocol, look for existing or ongoing syntheses so you can justify the review's contribution rather than unknowingly repeating work. Similar reviews do not automatically make a dissertation invalid: differences in population, intervention, question, date coverage or method may create a legitimate rationale. Record the actual search undertaken to identify prior reviews; do not invent one.
The essential sections of a dissertation review protocol
1. Title, authors, version and review status
Choose a title that identifies the design and topic. For example: Structured Written Formative Feedback and First-Term Course Completion in Undergraduate Education: Protocol for a Systematic Review. Identify authors, affiliations or roles as required; record version number, date, supervisor approval where relevant and registration information if registration actually occurs. A proposed registration is not a registration number.
Keep a stable filename and version history. A useful pattern is protocol-v1-YYYY-MM-DD, followed by separately dated amendments. Distinguish a supervisor's approval of the dissertation plan from approval or registration by an external registry.
2. Background, rationale and objectives
The background should explain the evidence problem, not attempt a full Results section. Define the intervention and its plausible mechanism, identify the uncertainty the review will address and explain why the target population and outcome matter. Cite real sources that support the rationale; do not claim that “no review exists” without checking.
An objective might read:
To systematically identify and appraise comparative studies of structured instructor-written formative feedback for first-year undergraduates, and to synthesize its association with first-term course completion relative to usual feedback practices.
Whether the review can support a causal conclusion depends on the eligible designs, their validity and the synthesis. If non-randomized studies are included, avoid promising causal certainty in advance.
3. Eligibility criteria specified in advance
Define populations, interventions, comparators, outcomes, eligible study designs, settings and any justified restrictions on language, reporting status or date. Make rules observable enough that two readers could apply them consistently. Explain what happens when reports are ambiguous, multiple reports belong to one study or a study lacks usable statistics for a specific synthesis.
| Domain | Proposed inclusion | Proposed exclusion or special handling |
|---|---|---|
| Population | First-year undergraduate students in credit-bearing courses | Non-degree or postgraduate-only populations; mixed cohorts require separable first-year data if the protocol specifies this |
| Intervention | Instructor-authored, structured written formative feedback on assessed work | Grade notification alone or entirely oral feedback |
| Comparator | Usual feedback practice or a prespecified relevant alternative | No eligible comparison for the planned comparative-effect question |
| Outcome | First-term course completion as prespecified and defined | Clarify outcomes that are measured but inadequately reported; do not select studies because results are favorable |
| Design | Randomized and specified non-randomized comparative designs | Editorials, commentary and uncontrolled descriptions for an effect question |
| Setting | Credit-bearing undergraduate modules | Informal workshops outside the target setting |
| Language / dates | To be explicitly declared and justified before searching | No retrospective restriction based on observed study findings |
A subtle point: study eligibility and eligibility for a particular outcome synthesis are not identical. A study that meets review inclusion rules but omits extractable results may still warrant inclusion and transparent reporting, depending on the prespecified criteria. Our inclusion and exclusion criteria guide develops this distinction with screening examples. Cochrane's Chapter 3 discusses planned grouping of studies and results.
4. Information sources and a reproducible search plan
List the databases and other sources you intend to search, appropriate to the topic. For education research, a plan might consider ERIC and other institutionally available multidisciplinary or social-science databases; database selection must reflect coverage, access and the question. Do not claim comprehensive coverage just because several platforms are listed.
For each electronic source, plan to preserve its name, platform, coverage or relevant restrictions, exact final search string and date searched. A good protocol provides a draft search strategy for at least one source and describes how it will be translated and pilot-tested for others.
For example, an illustrative search-building sketch is:
("formative feedback" OR "written feedback" OR "instructor feedback")
AND
("first-year" OR "first year" OR undergraduate*)
AND
("course completion" OR retention OR persistence)This is not a tested database-ready search. Field tags, controlled vocabulary, phrase handling, truncation and syntax differ by platform. A broad search may also deliberately omit an outcome term to improve sensitivity; document the final strategy and why it was chosen. Keep search development records rather than presenting this teaching string as a completed search.
Plan any citation searching, trial-register searching, grey-literature searches or author contact in advance when justified. Record the method and extent of these supplementary procedures later; never state that they were conducted merely because they appear in the protocol.
5. Records management, deduplication and study selection
Describe how citations will be imported, deduplicated and screened. Specify software only if it will actually be used; the procedure matters more than the brand name. Explain title/abstract screening, full-text retrieval, full-text exclusion reasons and how disagreements or unclear cases will be resolved.
If two independent screeners are planned, define independence and reconciliation. If only one dissertation researcher will screen because of practical constraints, state that honestly and discuss methodological limitations with the supervisor. Do not imply independent duplicate screening occurred when it did not.
The protocol should distinguish a record found by a search, a report such as a journal article, and the underlying study. Multiple reports may describe the same study. Our PRISMA flow-diagram guide explains why those units matter in the finished report.
6. Data extraction and data management
Define a planned extraction form and the data fields before looking at the final pattern of results. For the fictional feedback review, fields might include study identifiers, region, course type, design, participant eligibility, sample sizes, intervention components, comparison, completion definition, measurement time, reported results, missing outcome information and funding or conflicts of interest.
Specify who will extract, how fields will be piloted, how disagreements will be handled and how multiple publications of one study will be linked. If contacting authors for missing information is feasible, specify the approach. Do not promise author contact if the dissertation's resources cannot support it.
Our step-by-step guide to creating a systematic review data extraction form includes a field-level example and illustrates the distinction between studies and reports.
7. Planned risk-of-bias appraisal
Select an appraisal method suited to the eligible study designs and target results. For instance, randomization-related biases differ from confounding in non-randomized comparisons; a generic checklist should not be presented as interchangeable with design-specific risk-of-bias assessments.
State the proposed tool, unit of appraisal (study, outcome or result as appropriate), reviewer roles, disagreement resolution and intended use of appraisal findings in synthesis. Explain how you will handle high-risk evidence without quietly excluding unfavorable studies. Refer to Acadelyra's risk-of-bias assessment guide for a worked reporting approach.
8. Planned synthesis, heterogeneity and sensitivity analyses
Do not simply write “data will be meta-analysed where possible.” Explain the conditions under which studies would be sufficiently compatible to pool; specify which outcomes, comparisons and time points will be grouped; identify anticipated effect measures and models where relevant; and describe how inconsistency, missing data and study-level bias will affect interpretation.
If no defensible quantitative pooling is possible, plan a structured synthesis without meta-analysis rather than an unstructured summary. Prespecify meaningful groupings, how study characteristics will be tabulated, how direction and magnitude of effects will be reported and how conflicting findings will be examined. See our narrative synthesis guide.
Sensitivity analyses should test the dependence of a conclusion on defensible alternative assumptions—when the available data support them—not serve as a way to select the most attractive result. If subgroup analyses are planned, state their rationale and limits before inspecting subgroup patterns. The protocol may state contingencies, but it should not fabricate fixed effect estimates or thresholds unsupported by the question.
9. Assessing certainty and reporting the final review
For an intervention-effect review, consider whether an appropriate certainty-of-evidence framework, such as GRADE, will be used for the main outcomes. Specify the planned method and how it will inform interpretation. Certainty of a body of evidence is different from risk of bias within an individual study.
If appropriate, plan a Summary of Findings presentation with relevant comparisons and outcomes. Our GRADE and Summary of Findings guide explains how to communicate these judgments carefully.
Do not label the completed review “PRISMA-compliant” merely because a checklist was attached. Instead, describe your actual methods and use the relevant reporting checklist to ensure transparency about what was and was not done.
10. Timeline, roles, support and disclosures
Record achievable tasks: protocol development and approval, search strategy pilot, searching, screening, extraction, appraisal, synthesis and write-up. The timeline should reflect available supervision, database access and the number of reviewers; do not invent dates for activities that have already supposedly occurred.
State support or funding as applicable, known conflicts of interest and roles in decision making. If you intend to involve a librarian or second screener, specify what that person will actually do and whether participation is confirmed. Clarify data-management procedures and the institutional ethics or approval expectations for the dissertation; systematic reviews of published literature may not require the same ethics review as human-participant studies, but local requirements vary.
A worked example: turn a research idea into an auditable protocol
The following is a fictional planning exercise, not a registered protocol or a completed review. It shows how the parts of a protocol fit together.
Proposed title: Structured Instructor-Written Feedback and First-Term Completion Among First-Year Undergraduates: A Systematic Review Protocol.
Proposed review question: Does structured written formative feedback delivered by instructors, compared with usual feedback, improve course completion during the first term in first-year undergraduate courses?
Rationale: Course completion may be shaped by several teaching and student factors. Structured written feedback is one plausible intervention, but studies may differ in feedback content, setting, timing and design. A review would need to make those differences visible rather than assuming one common effect.
Eligibility: Comparative studies involving first-year undergraduates in credit-bearing courses, with eligible written-feedback interventions and relevant comparators; full inclusion and exclusion criteria would be approved before screening. Outcomes would distinguish course completion from satisfaction and perceived helpfulness.
Searching: Draft and pilot database-specific searches, include suitable education sources, document supplementary searches and preserve exact strings and search dates.
Selection: Deduplicate records; apply prespecified title/abstract and full-text screening rules; preserve full-text exclusion reasons and link multiple reports to the same underlying study. Report the actual number of reviewers and any limitations.
Extraction: Pilot a form recording study design, setting, participants, intervention components, comparator, outcome definition, assessment time, effect results and missing information.
Appraisal: Apply design-appropriate risk-of-bias methods; explain reviewer involvement and how concerns will be reflected in synthesis.
Synthesis: First group studies by compatible comparison, outcome definitions and time point. Quantitative pooling will be considered only when meaningful; otherwise use a prespecified structured synthesis. Any meta-analysis methods must be elaborated before application in a real protocol.
Transparency: Record protocol versions, legitimate amendments and any registration or repository posting accurately. Report discrepancies between the planned and completed review in the dissertation.
Notice what this example does not do: it does not invent a database search date, included study, screening count, pooled effect, registration ID or finding. A protocol describes intended actions; the completed dissertation must use the researcher's documented actions and verified evidence.
A practical PRISMA-P mapping checklist
PRISMA-P contains items relating to administrative information, introduction and planned methods. Use the official checklist for its full wording and any institutional adaptations. The mapping below is a study aid, not a reproduction of the entire checklist.
| Reporting area | What the dissertation protocol should make explicit | Common omission |
|---|---|---|
| Administrative information | Title, team roles, version, contacts, support and registration status if applicable | Claiming registration before receiving confirmation |
| Rationale and objectives | Uncertainty addressed and an answerable review question | Broad topic without a specific question |
| Eligibility | Explicit study and report characteristics, with justifications | Deciding eligibility after seeing results |
| Information sources and search | Planned sources, draft reproducible strategy, search documentation | Listing databases but no searchable strategy |
| Study records and selection | Deduplication, screening roles, exclusions, data management | No rule for ambiguous or duplicate reports |
| Data items and outcomes | Variables and priority outcomes, definitions and timing | Extracting only the outcomes that look interesting |
| Risk of bias | Design-appropriate tools, unit and roles | Unspecified “quality scoring” |
| Synthesis | Grouping, pooling conditions, alternatives, planned analyses | Assuming all studies can be meta-analysed |
| Meta-bias and certainty | Planned approach to selective reporting, missing evidence and confidence in findings where applicable | Treating study appraisal as identical to evidence certainty |
| Amendments | Method for dating, explaining and disclosing changes | Quietly revising the protocol after screening |
A checklist helps identify omissions. It cannot decide whether a restriction is justified, whether a chosen tool suits the review design or whether the planned sample and resources are feasible. These require substantive methodological judgment.
Should you register your dissertation systematic review?
Registration can improve transparency, but not every dissertation review is eligible for every registry. A protocol should exist before substantive review decisions regardless of whether public registration is possible or required.
PROSPERO
PROSPERO prospectively registers eligible systematic reviews involving outcomes of direct relevance to human health. Its current eligibility rules also address review stage and review type. A general education dissertation on course completion would not automatically qualify merely because it is systematic. Check the current official rules before applying, especially for training or classwork reviews. Do not claim a PROSPERO number unless the review has actually been registered.
OSF Registries
OSF Registries offers templates including Generalized Systematic Review, which can suit projects outside PROSPERO's health-related remit. Check the template, institutional policy and visibility or embargo settings before submission. A dated registration does not substitute for a detailed, well-designed protocol, and depositing something after the review is already complete is not prospective preregistration.
Institutional repositories and protocol publication
Depending on the discipline and university, a dated protocol may also be deposited in an institutional repository, shared as a method document or published in a suitable journal. Ask a supervisor what is permissible and what confidentiality, authorship or ethics constraints apply. Preserve the archived or submitted version and link it accurately when reporting the completed review.
What if registration is impossible? Complete, date and retain the protocol anyway. In the dissertation, disclose that the review was not prospectively registered and give the actual reason if relevant. Do not fabricate a registration number or imply that an unpublished document was publicly registered.
How to make legitimate protocol amendments without hiding them
A protocol is a plan, not a promise that no unexpected problem will arise. Screening may reveal a previously unanticipated ambiguity; a database may change its interface; outcome definitions may need clarification. What matters is when, why and how a change was made—and whether it could have been influenced by known results.
The Cochrane Handbook's introduction stresses that changes should not be made because of their effect on study results. Document amendments at the time, preserve earlier versions, note their implications for bias and update any registration or repository record as its rules permit.
| Version / stage | Proposed change | Reason | Transparency action |
|---|---|---|---|
| v1, before screening | Specify that “structured written feedback” requires documented instructor comments beyond a numerical grade | Clarify ambiguous intervention definition | Amend eligibility wording, date decision and retain original |
| v2, after search pilot, before selection | Add a relevant database to broaden education-research coverage | Pilot identified a coverage gap | Preserve original and revised search plans and justification |
| v3, after screening began | Clarify a rule for mixed-year cohorts where first-year results are not separately reported | Ambiguous reports encountered | Log timing, affected records, re-screening decisions and possible influence of observed information |
| Final dissertation | Explain all implemented deviations, including any that could affect interpretation | Allow readers to compare plans and actions | Report amendments and limitations openly |
These are illustrations of recordkeeping, not blanket permission to make any change. An amendment that changes the primary outcome after results are known requires particularly careful scrutiny and transparent disclosure. A vague statement that “the protocol was adapted where needed” is inadequate.
Example Methods wording: planned versus completed work
A protocol paragraph should use future-oriented or explicitly planned language:
This review will use prespecified eligibility criteria addressing first-year undergraduate populations, structured instructor-written formative feedback, eligible comparators and first-term completion. Searches will be developed and documented for relevant education and multidisciplinary databases. Records will be deduplicated and screened according to an approved selection procedure; data extraction and risk-of-bias assessments will follow planned, piloted procedures. Quantitative pooling will be considered only where comparisons and outcomes are sufficiently compatible. Protocol amendments will be dated and explained.
That is appropriate only if it accurately describes the actual plan. In the completed dissertation, the Methods section must be rewritten to describe what really occurred:
The review followed protocol version [actual identifier], approved on [verified date]. Searches were conducted in [actual sources] on [actual dates]. Screening, extraction and appraisal were undertaken by [actual people and procedures]. The review differed from the protocol in [actual documented respects], for [actual reasons].
The bracketed fields above are placeholders, not facts to invent. If an activity was not performed, do not claim it was. Reporting an imperfect but honest process is better than inventing independent screening, inaccessible databases or a registry entry.
Common protocol mistakes—and how to avoid them
Writing the protocol after the answers are known
A retrospective Methods description can be useful, but it is not evidence that methods were prespecified. Date the protocol before screening and analysis when possible; disclose if it was written or revised later.
Confusing a protocol with a PRISMA checklist
Ticking checklist items is not a substitute for operational decisions. The checklist prompts full reporting; the protocol must spell out criteria, methods, responsibilities and contingencies tailored to the research question.
Selecting only readily available articles
Convenience-based access rules can silently narrow the evidence base. Plan how to seek full texts, resolve uncertain cases and report inaccessible records. If practical limitations prevent full coverage, justify and disclose them rather than hiding them.
Overpromising two independent reviewers
An attractive protocol can be unrealistic. If a second reviewer has not agreed to participate, do not claim that independent duplicate review is guaranteed. Define a feasible plan, seek supervision and describe any deviations honestly.
Treating the absence of statistics as an exclusion rule
A study can meet the review's eligibility criteria even if an effect measure is not extractable for a particular synthesis. Prespecify the distinction between review inclusion, outcome availability and synthesis inclusion.
Planning meta-analysis without a pooling rationale
Pooling is not a prize for finding enough papers. Plan the comparisons, outcomes, effect metrics and criteria for clinical or methodological compatibility; be prepared to conduct a transparent structured synthesis instead.
Hiding amendments in the final dissertation
An amendment record is useful only if the final report compares planned and actual methods. Report changes, their timing and their possible implications—not merely the latest version of the protocol.
Frequently asked questions
Do I need a protocol for an undergraduate or master's dissertation systematic review?
If your project is genuinely a systematic review, a prespecified protocol is strong practice even when the department does not require public registration. Follow institutional guidance on scope, format, supervision and any applicable registration expectations.
Is PRISMA-P mandatory for every literature review?
No. PRISMA-P is intended for systematic review and meta-analysis protocol reporting. An ordinary narrative literature review, scoping review or another type of synthesis may require a different approach and reporting framework. Choose based on the actual method.
Can I write a protocol without knowing how many studies I will include?
Yes. You should not know or invent final inclusion counts beforehand. The protocol defines how eligible studies will be located and selected; the completed review reports the actual results.
Is a systematic review protocol the same as a research proposal?
Not exactly. A dissertation proposal can contain the rationale and proposed review methods, but a protocol needs operational detail sufficient for transparent systematic searching, selection, appraisal and synthesis. One document can serve both roles if all relevant requirements are met.
Should I register in PROSPERO or OSF?
Check the specific review's eligibility and the university's requirements. PROSPERO focuses on qualifying human-health-related reviews; OSF has wider registration templates. Neither should be presented as obligatory for every dissertation, and registration does not certify review quality.
What if my supervisor asks me to change an outcome after screening begins?
Discuss the methodological justification and whether knowledge of the evidence could influence the decision. Preserve the original protocol, record the amendment's timing and reasons, and disclose any consequences for interpretation in the final review.
Can I change the search strategy after the first database pilot?
Yes, a pilot may appropriately improve the strategy, particularly before substantive selection begins. Keep the pilot and final strategies distinguishable, record why terms or sources changed, and disclose material amendments.
Do I have to use two screeners if I am working alone?
Review standards often favor independent duplicate processes, but students may face constraints. Describe the actual planned procedure, consult the supervisor about acceptable safeguards and do not claim independent selection when it did not occur.
Does publishing a protocol mean my review is unbiased?
No. A protocol limits some opportunities for post hoc decisions, but biases can arise from inadequate searches, selection errors, poor appraisal, missing evidence and unsuitable synthesis. The final review still needs careful execution and transparent reporting.
Final protocol checklist before approval
The title, authors or roles, version and date are accurate.
The question and review design are explicit and justified.
Eligible populations, interventions or phenomena, outcomes and designs are operationally defined.
Search sources, draft strategies, coverage and documentation plans are described.
Deduplication, screening, full-text exclusions and duplicate reports are addressed.
Reviewer roles are feasible and described honestly.
Extraction fields, missing information and study identifiers are planned.
Appraisal methods suit the eligible designs and results.
Synthesis groups, pooling conditions and alternatives are prespecified.
Planned certainty assessments and reporting outputs are identified where appropriate.
Registration eligibility is checked and registration status is not overstated.
Version history and a procedure for documenting amendments are established.
Ethics, support, disclosure and institutional requirements have been considered.
All examples and placeholders are distinguishable from real research actions.
Final takeaway
A strong dissertation systematic review protocol makes methodological choices visible before the evidence can tempt researchers to revise them. It connects a clear question to eligibility, searching, selection, extraction, appraisal and synthesis; makes practical limitations explicit; and preserves an honest record when legitimate changes occur. A protocol does not guarantee a sound review, but a clear, dated protocol makes the review easier to evaluate and defend.
If you would value professional human support refining a review question, organising a protocol, checking references or improving the clarity of a dissertation methodology chapter, explore Acadelyra's Research Support and Dissertation & Thesis Support services. Acadelyra supports academic integrity and evidence-based writing; the research decisions, source evaluation and submitted work remain the student's responsibility.
Authoritative references and further reading
Moher, D., et al. (2015). Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement.
Cochrane. Cochrane Handbook for Systematic Reviews of Interventions, Chapter 1: Starting a review.
University of York, Centre for Reviews and Dissemination. PROSPERO eligibility requirements.
Center for Open Science. OSF Registries: Registration templates and guidance.
Page, M. J., et al. (2021). PRISMA 2020 statement for reporting completed systematic reviews.