Defining Research Visibility

The contemporary research environment is characterised by unprecedented growth in the production, digitisation and circulation of scholarly knowledge. Researchers now publish through journals, repositories, preprint servers, institutional platforms, research databases, conference systems, datasets, software repositories and other digital infrastructures. At the same time, researchers and institutions are increasingly evaluated through combinations of bibliometric, altmetric, institutional and qualitative indicators.

This environment creates an important distinction between producing research and making research visible.

A research study may be methodologically rigorous, theoretically important and potentially useful while remaining difficult for other researchers to discover. Conversely, a highly visible publication may receive substantial attention without necessarily representing the highest quality research. This distinction is particularly important because research evaluation systems increasingly employ quantitative indicators, while established responsible metrics frameworks caution against reducing scholarly assessment to single numerical indicators. The Leiden Manifesto, for example, argues that quantitative evaluation should support rather than replace qualitative expert assessment and that indicators should be interpreted in relation to disciplinary and institutional contexts (Hicks et al., 2015).

A parallel and closely related movement is the 2013 San Francisco Declaration on Research Assessment (DORA), which explicitly denounced the use of journal-level indicators, such as the Journal Impact Factor, as proxies for the quality of individual articles or researchers (DORA, 2013). DORA and the Leiden Manifesto together form the backbone of what is now widely termed responsible research metrics, and both inform the position taken in this Bulletin: that visibility indicators are diagnostic aids, not verdicts.

The concept of research visibility therefore requires systematic conceptual clarification.

Google Scholar's own inclusion guidance illustrates one part of the problem. Scholarly works must be technically discoverable by automated crawlers, have appropriate bibliographic information and provide accessible scholarly content or abstracts. Google Scholar recommends individual URLs for scholarly works and supports structured bibliographic metadata such as citation_title, citation_author and citation_publication_date (Google Scholar, n.d.).

However, technical discoverability represents only one component of the broader scholarly visibility process.

A research article may be technically indexed but rarely read. It may be read but not cited. It may be cited but not translated into practice. It may influence policy without generating substantial citation counts. It may be highly visible internationally while remaining inaccessible to communities that could benefit from it because of language, paywalls, infrastructure or disciplinary barriers.

Visibility should be treated as a multidimensional scholarly construct rather than a synonym for citation count or search engine ranking.

This Research Bulletin establishes the conceptual foundation for the ResearchTrics Research Bulletin Series. It expands the original conceptual note with a fuller literature grounding, worked illustrations, a structural case study of the African research system, and a preliminary discussion of how the proposed framework might eventually be operationalised and validated.

Several constructs are routinely used when discussing the performance and influence of research: research quality, research productivity, research visibility, research impact, research influence, scholarly engagement, citation influence, academic reputation, research discoverability, and knowledge translation.

These constructs overlap but are not interchangeable. For example, research productivity concerns the volume or rate of research outputs. Research quality concerns characteristics such as methodological rigour, theoretical contribution, validity and scholarly significance. Research impact concerns changes or effects attributable, at least in part, to research. Research visibility, by contrast, concerns the extent to which research becomes observable, discoverable, accessible and connected within relevant knowledge environments.

2.1 Comparing the constructs

Table 1 summarises the core distinctions among five constructs that are frequently, and often incorrectly, treated as synonyms in institutional research assessment.

Table 1. Distinguishing research productivity, quality, visibility, impact and reputation.

The distinction is consequential. A highly productive researcher may have low visibility if their outputs are poorly indexed or difficult to discover. A high quality article may have limited visibility because it is inaccessible or poorly represented in scholarly databases. A visible article may eventually produce impact, but visibility itself does not establish that impact has occurred.

The conceptual proposition of this Bulletin is therefore:

Research visibility is an enabling condition for scholarly engagement and potential influence, but it is neither equivalent to research quality nor sufficient evidence of research impact.

For purposes of the ResearchTrics framework, research visibility is defined as:

The degree to which a research output, researcher, research group or institution can be discovered, accessed, recognised, connected, engaged with and represented within relevant scholarly and knowledge ecosystems.

This definition contains several important elements.

3.1 Degree

Visibility is not necessarily binary. A research output can be minimally visible, locally visible, discipline specific, nationally visible, internationally visible, or broadly visible across scholarly and non scholarly knowledge ecosystems. Research visibility should therefore be conceptualised as a continuum rather than a simple visible or not visible classification.

3.2 Research object

Visibility can apply to multiple units of analysis:

• Individual research outputs
• Researchers
• Research groups
• Institutions
• Datasets
• Research instruments
• Software
• Research projects
• Research programmes

This broader unit of analysis approach is consistent with the development of open science, in which publications are increasingly connected to data, software, methodologies and other research outputs. UNESCO's Recommendation on Open Science explicitly recognises publications, data, metadata, software, source code, protocols and workflows as components of open scientific knowledge (UNESCO, 2021).

3.3 Scholarly ecosystem

Visibility does not occur in a vacuum. A research output exists within interconnected infrastructures involving researchers, institutions, journals, databases, repositories, citation networks, research communities, policy and practice communities, and society at large. Research visibility is therefore partly relational: it is produced as much by the surrounding infrastructure as by the properties of the output itself.

A central principle of the ResearchTrics framework is the separation of visibility from impact.

4.1 Research visibility asks

Can the relevant community discover, access, recognise and engage with this research?

4.2 Research impact asks

What difference has this research made?

Impact may include changes in scientific knowledge, theory, methodology, clinical practice, educational practice, policy, organisational behaviour, technology, professional practice, public understanding, or societal outcomes.

Visibility may facilitate impact, but the relationship is not deterministic. A useful conceptual sequence is:

Research production -> visibility -> engagement -> use -> influence -> impact

However, this should not be interpreted as a universal linear causal chain. Research can generate impact through pathways that are not captured by conventional visibility indicators, while highly visible research may have limited substantive influence. This distinction is consistent with responsible research assessment principles, which caution against false precision and emphasise the importance of context, qualitative judgement and multiple indicators (Wilsdon et al., 2015).

ResearchTrics proposes eight foundational dimensions. Each is defined below together with representative indicators; Table 2 provides a consolidated summary for quick reference.

5.1 Discoverability

Discoverability refers to the extent to which research can be located through relevant scholarly discovery systems. Indicators may include indexing, search engine retrieval, database presence, metadata completeness, persistent identifiers, keyword alignment, title clarity, abstract discoverability, and linkage to researcher profiles.

Google Scholar's technical guidance demonstrates the importance of crawlability, individual article URLs, searchable full text or abstracts, and appropriate bibliographic metadata for scholarly indexing (Google Scholar, n.d.).

5.2 Accessibility

Accessibility concerns whether researchers and other legitimate knowledge users can actually obtain and use the research once they discover it. Potential indicators include open access availability, repository availability, accessible abstracts, accessible datasets, accessible supplementary materials, language accessibility, machine readability, and technological accessibility.

UNESCO's Open Science Recommendation places substantial emphasis on making scientific knowledge openly available, accessible and reusable, including publications, research data, metadata, software and methodologies (UNESCO, 2021).

5.3 Scholarly Recognition

Scholarly recognition refers to the extent to which a research output or researcher is identifiable and attributable within scholarly systems. Relevant signals may include ORCID identity, institutional affiliation, author disambiguation, publication attribution, researcher profiles, journal records, database records, and persistent identifiers.

This dimension is particularly important in environments where researchers have multiple name variants, institutional affiliations or fragmented publication records. Recent ORCID community reporting on Africa illustrates the practical stakes: persistent identifiers are being promoted specifically to help elevate African research internationally by resolving identity fragmentation across databases (ORCID, 2026).

5.4 Scholarly Engagement

Visibility becomes more meaningful when other people interact with the research. Engagement may include article views, downloads, saves, discussions, social dissemination, academic presentations, requests for materials, educational use, research community discussion, and subsequent scholarly use.

The altmetrics movement was an explicit response to the limitations of traditional, slow filters such as peer review and citation counts, calling for new, faster and broader indicators of engagement drawn from online scholarly tools (Priem, Taraborelli, Groth and Neylon, 2010). Engagement should nevertheless be interpreted cautiously because attention is not synonymous with scientific value.

5.5 Connectivity

Research exists within networks. Connectivity concerns the degree to which a research output, researcher or institution is connected to other components of the knowledge ecosystem. Examples include co-authorship networks, institutional collaboration, citation relationships, researcher networks, interdisciplinary connections, project networks, dataset to publication relationships, and researcher to instrument relationships. This dimension provides a foundation for research graph and scholarly network analysis.

5.6 Citation Visibility

Citation visibility concerns the extent to which a research work becomes incorporated into subsequent scholarly literature. Citation indicators may include citation count, citation growth, citing document diversity, disciplinary reach, geographic reach, citation network centrality, and citation trajectory.

Citation counts should not be interpreted as direct measures of research quality. Responsible research evaluation frameworks explicitly caution against relying on single metrics and emphasise disciplinary context and qualitative assessment (Hicks et al., 2015; Wilsdon et al., 2015).

5.7 Openness

Openness concerns the extent to which research outputs and associated research resources can be accessed, reused, verified and connected. This includes open access publications, open data, open methods, open software, open protocols, transparent reporting, reusable metadata, and persistent research infrastructure.

UNESCO identifies open scientific knowledge, open science infrastructures, open engagement of societal actors, and open dialogue with other knowledge systems as core pillars of open science (UNESCO, 2021).

5.8 Knowledge Reach

Knowledge reach concerns the extent to which research travels beyond its immediate publication environment. Potential domains include other academic disciplines, other institutions, other countries, professional communities, policymakers, practitioners, industry, educators, and wider society. Knowledge reach is therefore broader than citation influence.

Table 2. Summary of the eight ResearchTrics dimensions of research visibility.

The dimensions proposed above can be represented as an integrated model, moving from a research output or researcher through progressively wider circles of discovery, access, recognition, engagement, connection and reach, toward potential scholarly and societal influence.

This representation should not be interpreted as a strictly causal model. Rather, it represents an interconnected system in which dimensions can reinforce one another. For example, better metadata can lead to greater discoverability, which increases the probability of access, which creates more opportunity for engagement, which in turn creates more opportunity for citation and reuse. Similarly, a strong researcher identity can improve attribution, which strengthens network connectivity, which improves the discovery of related outputs.

Future ResearchTrics studies should empirically examine these relationships rather than assuming them.

The framework implies that research visibility should not be represented by a single raw indicator. For example, citation count is not equal to research visibility; downloads are not equal to research visibility; a Google Scholar ranking is not equal to research visibility; and social media attention is not equal to research visibility. Each represents only a partial signal.

A more defensible conceptual representation is:

Research Visibility = f (Discoverability, Accessibility, Recognition, Engagement, Connectivity, Citation Visibility, Openness, Knowledge Reach)

The exact mathematical operationalisation of this construct should be developed empirically rather than predetermined. This is important because composite indicators can create false precision when dimensions are arbitrarily weighted. The Leiden Manifesto specifically emphasises the need to account for differences between fields, avoid misplaced concreteness, and regularly scrutinise indicators (Hicks et al., 2015).

Figure 2 illustrates why a single number is insufficient. It shows two hypothetical researchers with a similar citation count but very different underlying visibility profiles: one is well indexed, open and well networked, while the other is comparatively closed and disconnected despite similar citation performance. The figures are illustrative only and are not drawn from an empirical study; they are included to demonstrate the diagnostic value of a multidimensional profile over a single score.

Consequently, future ResearchTrics work should investigate dimensionality, weighting, reliability, construct validity, convergent validity, discriminant validity, measurement invariance, disciplinary normalisation, geographical bias, career stage effects, institutional effects, and susceptibility to gaming.

The concept has particular relevance for African research systems. Research generated in African universities and research institutions may face structural visibility barriers involving fragmented scholarly infrastructure, inconsistent researcher identification, limited institutional repositories, metadata deficiencies, restricted access to publications, language barriers, limited international collaboration, inadequate digital research infrastructure, and underrepresentation in global knowledge databases.

Empirical bibliometric work supports this characterisation. Using Web of Science publication data, Tijssen (2007) found that African-authored articles accounted for less than 1.5 per cent of all indexed publications between 1980 and 2004, and that only a small fraction of the journals listed in the Science and Social Sciences editions of the Journal Citation Reports originated from African publishers. More recent analyses confirm that sub-Saharan Africa remains the most underrepresented world region in both Scopus and Web of Science (Asubiaro, Onaolapo and Mills, 2024), and that even inclusive, open bibliographic sources such as OpenAlex, while an improvement, still show gaps in the coverage and metadata quality of African research outputs (Chavarro and Alperin, as reported in the OpenAlex comparative analysis, 2024).

A large share of African scholarship is also published in journals that are not indexed internationally, which supports the view that a substantial proportion of the continent's research is, in bibliometric terms, hidden rather than absent (see Figure 3). Editorial capacity is itself part of the visibility problem: a qualitative study of journals in Kenya, Ethiopia, Nigeria and Mozambique found that editorial practices, ranging from peer review workflows to digital platform integration, are underexplored despite being central to whether a journal and its articles can be found, accessed and cited (Gates Open Research, 2025).

A visibility framework must therefore avoid treating international citation as the sole representation of scholarly value. The Leiden Manifesto explicitly calls for protection of locally relevant research and recognition of disciplinary differences in publication and citation practices (Hicks et al., 2015). Similarly, UNESCO's Open Science Recommendation emphasises equitable access to scientific knowledge and recognises that open science should operate across countries, disciplines, languages and career stages (UNESCO, 2021).

Research visibility should consequently be capable of recognising both global reach and locally relevant scholarly contribution. Regional infrastructure such as African Journals Online (AJOL) and the growing adoption of persistent identifiers through initiatives such as ORCID's African engagement programme represent practical steps toward narrowing this gap (ORCID, 2026).

The conceptual distinction established in this Bulletin has implications for universities, research institutions, publishers, funding agencies and individual researchers. Research evaluation systems should distinguish at least four questions, summarised in Table 3.

Table 3. Four questions a mature research evaluation system should distinguish.

A mature research evaluation system should therefore use a portfolio of evidence rather than treating any single metric as definitive. This position aligns with established responsible metrics guidance. The Leiden Manifesto recommends that quantitative indicators support qualitative assessment (Hicks et al., 2015), while the Metric Tide review commissioned by the UK Higher Education Funding Council for England similarly concludes that metrics should support, not supplant, expert judgement, and calls for transparency and the use of multiple indicators rather than a single metric (Wilsdon et al., 2015).

If research visibility is multidimensional, research infrastructure should also be multidimensional. A future oriented research visibility infrastructure should connect researcher identity, publications, journals, institutions, research projects, datasets, research instruments, methods, citation networks, collaboration networks, and research opportunities into a single interoperable structure.

Such a connected architecture moves beyond the conventional researcher profile. It creates a research knowledge graph in which scholarly entities and relationships can be represented and analysed. This provides an important conceptual basis for the development of ResearchTrics as a research intelligence environment.

Based on the conceptual analysis presented above, ResearchTrics proposes the following preliminary, seven level framework.

Table 4. The seven levels of the proposed ResearchTrics Research Visibility Framework (RRVF).

The framework should be treated as conceptual and provisional until empirically validated.

Translating the RRVF from a conceptual scheme into a validated measurement instrument raises several methodological questions that subsequent ResearchTrics bulletins should address in turn.

12.1 Dimensionality and weighting

Before any composite score is proposed, it is necessary to establish whether the eight dimensions are empirically distinguishable, for example through exploratory and confirmatory factor analysis of candidate indicators, and whether a simple additive model, a weighted model, or a latent variable model best represents the underlying construct.

12.2 Normalisation

Citation and engagement behaviour vary enormously across disciplines, publication types and career stages. Any visibility metric must be field normalised and career stage aware to avoid systematically disadvantaging early career researchers, humanities and social science scholars, or researchers working in under indexed regions such as much of sub-Saharan Africa, as discussed in Section 8.

12.3 Resistance to gaming

Any indicator that becomes a target risks distortion of the behaviour it measures. A defensible visibility metric should be resistant to manipulation, for example through coordinated self-citation, artificial engagement or metadata stuffing, and should be regularly audited, consistent with the Leiden Manifesto's principle of continual scrutiny of indicators (Hicks et al., 2015).

12.4 Governance and transparency

Consistent with DORA (2013) and the Metric Tide (Wilsdon et al., 2015), the underlying data sources, computation methods and known limitations of any ResearchTrics visibility metric should be published openly, and the metric should always be presented alongside, rather than instead of, qualitative expert review.

This first ResearchTrics Research Bulletin establishes several questions for subsequent investigation. Future bulletins in this series should examine the following:

1. What are the empirically observable dimensions of research visibility?
2. Are the proposed eight dimensions statistically distinguishable?
3. Can research visibility be modelled as a latent variable?
4. What indicators provide the strongest evidence for each dimension?
5. How should research visibility be normalised across disciplines?
6. Does research visibility differ systematically by career stage?
7. Are there geographic biases in existing visibility systems?
8. How does open access influence discoverability?
9. What is the relationship between visibility and citation growth?
10. Can researcher identity fragmentation reduce visibility?
11. How can research visibility metrics avoid rewarding gaming?
12. How should locally relevant research be represented?
13. Can research visibility be measured longitudinally?
14. What forms of visibility predict subsequent scholarly influence?
15. Can a validated Research Visibility Metric be developed?

These questions provide the foundation for the next stages of the ResearchTrics Research Bulletin programme.

This Bulletin is conceptual rather than empirical, and several limitations should be acknowledged. First, the eight proposed dimensions and the seven level RRVF are provisional; they have been derived from a synthesis of existing scholarly communication, open science and responsible metrics literature rather than from primary data collection or psychometric testing. Second, the illustrative figures presented in Section 7 (Figure 2) use hypothetical data and are intended only to demonstrate the diagnostic logic of a multidimensional profile; they should not be read as empirical findings. Third, the figures presented on the African research system in Section 8 draw on real, cited bibliometric studies, but these studies differ in period, database coverage and methodology, and should be read as indicative of a well documented structural pattern rather than as a single precise estimate. Finally, the framework does not yet specify measurement instruments, data sources or validation procedures in sufficient detail for operational deployment; this is deliberate, and is the explicit subject of the research agenda in Section 13.

Research visibility represents an important but insufficiently differentiated dimension of contemporary scholarly communication.

This Bulletin defines research visibility as the degree to which research outputs, researchers, research groups or institutions can be discovered, accessed, recognised, connected, engaged with and represented within relevant scholarly and knowledge ecosystems.

The proposed framework distinguishes visibility from research quality, productivity and impact. It conceptualises visibility as a multidimensional construct encompassing discoverability, accessibility, scholarly recognition, engagement, connectivity, citation visibility, openness and knowledge reach.

The framework also establishes an important methodological principle: visibility should not be reduced to a single numerical indicator.

A credible research visibility system must account for disciplinary differences, geographic context, career stage, institutional context and the diverse ways in which knowledge travels. Quantitative indicators should complement rather than replace expert judgement, consistent with established responsible research assessment principles.

The ResearchTrics Research Bulletin Series will build upon this conceptual foundation through subsequent conceptual, methodological, bibliometric, psychometric and empirical studies.

Ultimately, the objective is not simply to make research more visible. It is to understand how scholarly knowledge becomes discoverable, connected, usable and influential within an increasingly digital global research ecosystem.

Oladunmoye, E. O. (2026). Defining research visibility: A conceptual framework for understanding the discoverability, accessibility, engagement and scholarly reach of research. ResearchTrics Research Bulletin, 1, 1-34.

Proposed citation key

Oladunmoye (2026)

Recommended internal citation

Research visibility should be understood as a multidimensional construct encompassing discoverability, accessibility, recognition, engagement, connectivity, citation visibility, openness and knowledge reach (Oladunmoye, 2026).

This Bulletin is the foundational conceptual paper for the ResearchTrics Research Bulletin Series. Subsequent bulletins should cite this work where they use, extend, test or critique the ResearchTrics definition and framework of research visibility.

The framework is explicitly presented as provisional and subject to empirical validation. It should not be represented as an established psychometric or bibliometric metric until appropriate validation studies have been conducted.