Introduction
Your criteria are the decision engine of your entire review
Of all the methodological decisions you make in a systematic literature review, your inclusion and exclusion criteria are the most consequential. They determine which papers enter your synthesis and which do not. They define the boundaries of your evidence base. They are the first thing a peer reviewer of a management journal will scrutinise, and the first thing a viva examiner will ask you to justify.
And yet they are consistently the least carefully designed component of management SLRs. Snyder (2019), writing in the Journal of Business Research, notes that deciding on eligibility criteria is one of the most practically challenging steps in business research reviews, precisely because the conceptual diversity of management literature does not map neatly onto the health science templates that dominate SLR methodology guidance. Williams et al. (2021), reviewing SLR practice in the European Management Journal, are explicit: if the inclusion criteria are not appropriate or sound, the SLR may incur selection bias or include studies that do not address the focal questions.
This guide addresses that gap. It walks you through what criteria are, which framework to use for management research, how to write criteria that are specific enough to apply consistently and flexible enough to capture the conceptual diversity of the discipline, and exactly what PRISMA 2020 requires you to report.
A note on terminology
PRISMA 2020 uses the term "eligibility criteria" to refer to the combined set of inclusion and exclusion conditions. In practice, most management SLR papers present these as two separate lists: inclusion criteria (what a study must satisfy) and exclusion criteria (what disqualifies it). Both terms are used throughout this guide.
01 · Definitions
What inclusion and exclusion criteria actually are
Inclusion criteria specify the conditions a study must satisfy to be considered for your review. They translate your research question into observable, testable characteristics. A study that meets all inclusion criteria passes to the next stage of screening; a study that fails any inclusion criterion is excluded.
Exclusion criteria specify conditions that disqualify a study that might otherwise appear relevant. This is a distinction that matters: exclusion criteria are not simply the inverse of inclusion criteria. They address specific disqualifying features, a study may appear topically relevant but be excluded because it uses a methodology incompatible with your synthesis approach, or because it reports data at an incompatible unit of analysis.
PRISMA 2020 Item 5 (Page et al., 2021) requires that you specify all study characteristics used to decide eligibility, including components from your chosen research question framework, eligible study designs and settings, and report characteristics such as publication year, language, and publication status. Critically, PRISMA 2020 notes that vague exclusion labels such as "no relevant outcome data" are ambiguous and should be avoided. Every exclusion reason must be traceable to a specific, named criterion.
"Specifying the criteria used to decide what evidence was eligible or ineligible in sufficient detail should enable readers to understand the scope of the review and verify inclusion decisions."
Page et al. (2021), PRISMA 2020 Explanation and Elaboration, BMJPati and Lorusso (2018) describe the function of criteria precisely: they filter the evidence base to ensure only studies relevant to your research question are included, and they do so in a way that is documented, consistent, and reproducible. The criteria are not a post-hoc rationalisation of the papers you happened to find, they are a pre-specified decision system that you commit to before you see your search results.
02 · Frameworks
Choosing your framework: PICO, PCC, SPIDER, and CIMO
Your eligibility criteria must be derived from a structured research question framework. The framework you choose determines the architecture of your criteria, and choosing the wrong one for your research type is one of the most common errors in management SLRs. Most guidance defaults to PICO, which was designed for clinical intervention studies. For much of management research, it is a poor fit.
The four frameworks relevant to business and management researchers are presented below, with guidance on when each is appropriate.
Use when: Your review examines the effect of a specific managerial intervention, policy, or practice on a measurable outcome, and you are comparing conditions (with vs. without the intervention, or between two approaches).
Management examples: Effect of board independence on firm performance; impact of training programmes on employee turnover; comparative effectiveness of supply chain strategies.
Use when: Your review maps or explores a concept across contexts rather than testing an intervention effect. No comparator or specific outcome is required. Suitable for scoping reviews and conceptual SLRs.
Management examples: How sustainability is conceptualised in SME research; corporate governance practices in emerging economies; entrepreneurial intention across cultural contexts.
Use when: Your review includes qualitative or mixed-method studies. SPIDER explicitly captures research design and type, making it appropriate for reviews that synthesise qualitative evidence or combine qualitative and quantitative findings.
Management examples: Employee experiences of organisational change; managerial sensemaking in crisis; qualitative studies of leadership in family firms.
Use when: Your review is interested in how and why an intervention works, not just whether it does. CIMO captures the contextual conditions and causal mechanisms that PICO ignores, making it the most appropriate framework for much of management and organisational research.
Management examples: How organisational culture mediates the effect of leadership style on innovation; mechanisms by which CSR influences employee engagement; contextual conditions for dynamic capability development.
The PICO trap in management research
PICO was designed for randomised controlled trials in clinical medicine, where populations, interventions, and outcomes are measurable and comparable across studies. Many management SLRs default to PICO without considering whether it fits their research question, and then struggle to derive meaningful criteria from a framework that was never designed for their context. If your review is asking how or why something works, or exploring a concept across contexts, CIMO or PCC will serve you significantly better. Mannarath (2026) and Mohamed Shaffril et al. (2021) both emphasise the importance of selecting a framework appropriate to the research context rather than defaulting to the most familiar option.
03 · Writing Criteria
How to write criteria that are specific, observable, and consistently applicable
The most common failure in criteria design is not vagueness, it is untestability. A criterion is only useful if two independent reviewers can apply it to the same abstract and reach the same decision. Xiao and Watson (2019) identify this as the practicality test: criteria must be capable of classifying research, reliably interpreted, and result in a manageable volume of literature. The key word is reliably interpreted.
-
01Anchor every criterion to a framework element
Each criterion should map explicitly to an element of your chosen framework. If you are using CIMO, each criterion should relate to Context, Intervention, Mechanism, or Outcome. Criteria that cannot be linked to a framework element are usually scope decisions that belong in your rationale section, not your criteria list.
-
02Distinguish conceptual criteria from methodological criteria
Conceptual criteria address topic relevance, the subject, population, and scope of your review. Methodological criteria address the characteristics of eligible studies, study design, unit of analysis, publication type, date range, and language. Both are essential. Most criteria errors occur when researchers mix these two types without separating them, creating criteria that are hard to apply consistently.
-
03Write criteria in testable, observable language
Replace evaluative language with observable conditions. "Studies of sufficient quality" is not a criterion, it is a judgment. "Peer-reviewed articles published in indexed journals" is a criterion. "Studies addressing corporate governance" is vague. "Studies examining board composition, director independence, or governance mechanisms at the firm level" is testable. Williams et al. (2021) are explicit: inclusion criteria must result from rational decision-making, and the SRRT should justify the criteria they applied.
-
04Pilot your criteria before full screening
Apply your criteria to a sample of 50 papers before beginning full screening. This is not a methodological formality, it is a quality control mechanism. Xiao and Watson (2019) are direct: criteria should be piloted before adoption. A pilot reveals ambiguous criteria, generates agreed-upon interpretations for borderline cases, and prevents the single most costly error in SLR execution: discovering mid-screening that your criteria are inconsistently applicable.
-
05Document criteria changes and justify restrictions
If your criteria evolve during the screening process, document the changes and re-screen papers that were assessed before the change. Williams et al. (2021) note that as the team works through the inclusion decision process, criteria may evolve, and those modifications must be documented and reported. PRISMA 2020 also advises providing rationales for notable restrictions, such as limiting to studies published after a particular year because that marks the emergence of the phenomenon you are studying.
Worked example: a full criteria set for a management SLR
The following illustrates a complete criteria set for a hypothetical management SLR examining how board diversity influences innovation in listed firms. It uses the CIMO framework and separates conceptual from methodological criteria.
Review question: In what contexts, through what mechanisms, and with what outcomes does board diversity influence firm-level innovation in publicly listed companies?
| Framework Element | Criterion | Type |
|---|---|---|
| Context | Studies examining publicly listed firms (stock exchange listed companies in any country) | Inclusion |
| Context | Studies of private firms, SMEs, non-profit organisations, or public sector entities | Exclusion |
| Intervention | Studies examining at least one dimension of board diversity (gender, age, nationality, educational background, or functional expertise of board directors) | Inclusion |
| Intervention | Studies examining management team diversity (CEO, TMT) without a board-level dimension | Exclusion |
| Mechanism | Studies that theorise or empirically examine a mediating or moderating mechanism between board diversity and innovation outcomes | Inclusion |
| Outcome | Studies measuring at least one innovation outcome: R&D expenditure, patent counts, new product introductions, innovation intensity, or innovation performance scales | Inclusion |
| Outcome | Studies measuring firm performance outcomes only (ROA, Tobin's Q, market returns) without an innovation-specific outcome variable | Exclusion |
| Methodological | Peer-reviewed journal articles published in English | Inclusion |
| Methodological | Published between January 2000 and December 2025 (rationale: corporate governance codes mandating diversity reporting emerged from 2000 onward) | Inclusion |
| Methodological | Conference papers, book chapters, dissertations, working papers, editorials, and grey literature | Exclusion |
| Methodological | Conceptual papers, theoretical frameworks, and literature reviews not reporting primary empirical data | Exclusion |
A note on journal quality as a criterion
Williams et al. (2021) explicitly advise against using journal quality ratings (ABS, ABDC, JQL) as an inclusion criterion. Restricting to ABS 3★ and above, for example, introduces selection bias and may exclude relevant empirical work published in field-specific or regional journals. Assess papers on the criteria relevant to your research question, not on editorial decisions made by journal ranking bodies.
04 · Common Errors
Six common criteria design errors in management SLRs
These errors recur across published management SLRs and are consistently identified in methodological quality assessments of the field. Each is avoidable with careful protocol design.
Criteria so broad that they admit papers only tangentially related to the research question. This inflates the screening pool, extends review time significantly, and risks including studies whose findings are not meaningfully comparable.
Ground each criterion in your framework. If a paper can satisfy your criteria without actually addressing your research question, the criteria are too broad. The pilot phase will expose this.
A criterion requiring full-text reading to apply cannot drive abstract screening. If you cannot decide yes or no from the title and abstract, the criterion belongs in your full-text review stage, not your abstract screening criteria set.
Test each criterion against 10 abstracts before screening begins. If you cannot apply it consistently from the abstract, either rewrite it in terms observable from the abstract or move it to full-text eligibility assessment.
Many management SLRs specify conceptual criteria carefully but omit methodological criteria entirely, leaving unstated whether they include qualitative studies, conference papers, book chapters, or pre-prints. Smela et al. (2023) observe that SLR practice varies considerably in this dimension, and omissions create replicability problems.
Always specify: publication type, date range (with justification), language restrictions, and whether grey literature is included. Each decision should be documented with a brief rationale.
Restricting to studies published after a certain year is legitimate, but only when you can justify why that date is meaningful. "Published after 2010" with no rationale is a decision that peer reviewers will question. PRISMA 2020 specifically advises providing rationales for notable eligibility restrictions.
Anchor your date restriction to a substantive reason: the year a key regulation was introduced, the year a construct first appeared in the management literature, or the year a dataset became available. State this in your methods section.
Restricting to English-language publications is a practical necessity for most researchers, but it is also a source of selection bias that must be acknowledged. Mohamed Shaffril et al. (2021) note that this is a transparency requirement: all inclusion terms must be defined and justified, and all exclusions must be reasoned.
State the language restriction explicitly in your criteria table. Acknowledge in your limitations section that restricting to English may exclude relevant studies published in other languages, particularly from non-Anglophone research communities.
Discovering mid-screening that a criterion is ambiguous and quietly refining it is one of the most serious methodological integrity failures in SLR execution. Papers screened before the modification were assessed against different criteria than those screened after. The resulting sample is internally inconsistent.
Document any criteria change, the reason for it, the date it was made, and re-screen all papers assessed before the change. If the re-screening burden is prohibitive, consider whether the modification was truly necessary or whether the original criterion can be clarified through an operational definition instead.
05 · PRISMA 2020 Reporting
Documenting and reporting your criteria for PRISMA 2020 compliance
PRISMA 2020 (Page et al., 2021) Item 5 sets out what you must report about your eligibility criteria. Rethlefsen and Page (2022) clarify how this reporting connects to the PRISMA flow diagram. The following checklist covers what is required for a methodology section that satisfies peer reviewers at ABS-ranked management journals.
Essential reporting elements (PRISMA 2020 Item 5)
- All study characteristics used to decide eligibility, organised by your chosen framework (PICO, PCC, SPIDER, or CIMO)
- Eligibility criteria for report characteristics: year of dissemination, language, and publication status (peer-reviewed only, or including grey literature)
- Clear distinction between studies ineligible because the outcome of interest was not measured versus studies ineligible because relevant outcome data were not reported, these are different exclusion reasons and must be recorded separately
- Any groups used in the synthesis and how they link to the comparisons in your objectives
Connecting criteria to your PRISMA flow diagram
Rethlefsen and Page (2022) clarify that the PRISMA 2020 flow diagram requires you to track the number of records excluded at each stage and, critically, the reasons for exclusion at full-text stage. Each exclusion reason at full-text stage should map to a specific, named criterion from your criteria table. Vague flow diagram entries such as "not relevant" or "out of scope" do not satisfy PRISMA 2020 reporting requirements.
The standard approach is to number your criteria in your criteria table (IC1, IC2, EC1, EC2, etc.) and use those codes when recording exclusion reasons in your screening log. This produces a fully auditable, PRISMA-compliant record that links every exclusion decision to a specific criterion.
What management journal peer reviewers look for
Williams et al. (2021) note that management SLRs should provide a complete and detailed description of steps applied, criteria employed in those steps, and reasoning for steps and criteria utilised. In practice, peer reviewers at ABS 3★ and 4★ journals expect to see: a clearly formatted criteria table (not a paragraph description); explicit framework linkage; a rationale for notable restrictions; and a PRISMA flow diagram where exclusion numbers add up correctly and exclusion reasons are specific.
The most common PRISMA compliance failure in management SLRs
The single most frequently cited criteria-related reviewer comment in management SLRs is the failure to link flow diagram exclusion numbers to specific criteria. If your PRISMA flow diagram shows 412 records excluded at abstract screening with the reason "not relevant to research question," a reviewer cannot verify whether your criteria were applied consistently. Replace vague aggregate labels with criterion-coded exclusion reasons.
06 · AI-Assisted Screening
How AI-assisted tools interact with your criteria
AI-assisted screening tools use your eligibility criteria as their operating instructions. The precision and testability of your criteria directly determines the quality of AI-assisted screening, this is a relationship that most discussions of AI in SLRs do not address directly.
A well-written criterion, specific, observable, anchored to a framework element, and testable from the abstract, gives an AI screening tool clear decision boundaries. An ambiguous criterion gives the tool an ambiguous signal, producing screening decisions that are inconsistent, unreliable, and require extensive human review to validate.
Abogunrin et al. (2026), reviewing AI adoption in SLRs for health technology assessment bodies, found that human oversight remained essential across all 112 identified AI-assisted SLRs, and that no fully autonomous AI screening process was reported in the literature. The implication for criteria design is direct: even with AI assistance, the quality of your criteria determines the quality of your screening. AI does not compensate for vague criteria; it amplifies the consequences of them.
The most effective AI-assisted screening workflow keeps the researcher in control of every final decision while using AI to rank papers by relevance and flag borderline cases for closer review. This model, AI-ranked, human-decided, fully documented, maps directly to PRISMA 2020's requirements for transparent, verifiable eligibility decisions.
Reporting AI use in your methodology section
If you use AI assistance in your screening, PRISMA 2020 requires you to report the tool used, how it was applied, and how human oversight was maintained. State which criteria were provided to the AI tool as its decision framework, how the tool ranked or classified papers against those criteria, and that all final inclusion/exclusion decisions were made by the human researcher with reference to documented criteria. This is what peer reviewers at management journals now expect.
References
- Abogunrin, S., Liu, Y., & Zerbini, C.H. (2026). A systematic literature review (SLR) on the adoption of artificial intelligence-assisted SLRs: implications for health technology assessments. International Journal of Technology Assessment in Health Care, 42(1), e29.
- Mannarath, A.K. (2026). Systematic literature review for an effective research: a structured framework for social science researches. Quality & Quantity, 60, 4773–4799.
- Mohamed Shaffril, H.A., Samsuddin, S.F., & Abu Samah, A. (2021). The ABC of systematic literature review: the basic methodological guidance for beginners. Quality & Quantity, 55, 1319–1346.
- Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D., … & Moher, D. (2021). Updating guidance for reporting systematic reviews: development of the PRISMA 2020 statement. Journal of Clinical Epidemiology, 134, 103–112.
- Page, M.J., Moher, D., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D., … & McKenzie, J.E. (2021). PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ, 372, n160.
- Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D., … & Moher, D. (2021). The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ, 372, n71.
- Pati, D., & Lorusso, L.N. (2018). How to write a systematic review of the literature. Health Environments Research & Design Journal, 11(1), 15–30.
- Rethlefsen, M.L., & Page, M.J. (2022). PRISMA 2020 and PRISMA-S: common questions on tracking records and the flow diagram. Journal of the Medical Library Association, 110(2), 253–257.
- Smela, B., Toumi, M., Świerk, K., Gawlik, K., Clay, E., & Boyer, L. (2023). Systematic literature reviews over the years. Journal of Market Access & Health Policy, 11(1), 2244305.
- Snyder, H. (2019). Literature review as a research methodology: an overview and guidelines. Journal of Business Research, 104, 333–339.
- Williams, R.I., Clark, L.A., Clark, W.R., & Raffo, D.M. (2021). Re-examining systematic literature review in management research: additional benefits and execution protocols. European Management Journal, 39, 521–533.
- Xiao, Y., & Watson, M. (2019). Guidance on conducting a systematic literature review. Journal of Planning Education and Research, 39(1), 93–112.