Testing the knowledge bridging function of non-mainstream journals in scholarly communication

Testing the knowledge bridging function of non-mainstream journals in scholarly communication Victoria Di Césare*, Rodrigo Costas** and Nicolas Robinson-Garcia*** *[email protected] https://orcid.org/0000-0001-7098-1914 EC3 Research Group, Unit for Computational Humanities and Social Sciences (U-CHASS), University of Granada, Spain **r[email protected].nl https://orcid.org/0000-0002-7465-6462 Centre for Science and Technology Studies (CWTS), Leiden University, The Netherlands DSI-NRF Centre of Excellence in Scientometrics and Science, Technology and Innovation Policy (SciSTIP), Stellenbosch University, Stellenbosch, South Africa ***[email protected] https://orcid.org/0000-0002-0585-7359 EC3 Research Group, Unit for Computational Humanities and Social Sciences (U-CHASS), University of Granada, Spain This study explores the knowledge bridging communication function of non-mainstream journals, originally proposed by Chavarro et al. (2017). Based on their definition, and incorporating a multidimensional conception of locality (Di Césare & Robinson-Garcia, 2024), we identify journals that perform this function within a large 2023 OpenAlex dataset. Our findings reveal a predominance of such journals in Asian and South American countries. Large scientific producers like Brazil and Russia stand out as significant contributors to knowledge bridging journals. The Social Sciences domain leads article production, followed closely by the Physical Sciences. The study highlights the value of non-mainstream journals as knowledge bridging sources beyond traditional metrics and underscores the need for greater recognition of their role in scholarly communication. 1. Introduction Much has been debated about the dichotomy between mainstream and non-mainstream journals. This includes differing perceptions of their quality and excellence, which impact recognition during research assessments (Bianco et al., 2016; Vessuri et al., 2014), as well as their uneven visibility and access opportunities, often shaped by economic and language barriers (Tijssen et al., 2006). Such factors contribute to an unequal appraisal of the research published in non-mainstream venues that jeopardizes certain forms of knowledge while promoting other agendas (López Piñeiro & Hicks, 2015). Their distinction cannot be explained based on universalistic criteria, like editorial standards and scientific impact, because particularistic variables such as country, discipline, and language strongly determine the inclusion of journals in databases like Web of Science (Chavarro et al., 2018). These databases, widely considered mainstream, are known for being strongly biased toward Western, Anglophone content (Larivière et al., 2015; van Leeuwen et al., 2001). Consequently, the accepted assumption that non-mainstream, usually implying non-indexed, journals lack value in disseminating research findings is called into question. Chavarro et al. (2017) analyzed the role played by non-mainstream journals and the reasons leading researchers to publish in them. They conducted a series of interviews with Colombian researchers from which they identified three main communication functions. The first function is training, primarily for young researchers and PhD students. In these journals, they 9 find a space for initiation into publishing, incorporation of feedback from editors and peer reviewers, identification of relevant literature in their native languages, and overall capacity building that will eventually help them reach mainstream venues. The second is knowledge bridging, a function that connects articles from mainstream journals to communities with limited or no access to them. This is mainly done with dissemination in non-English languages and open accessibility purposes, in close relation to teaching and introduction of concepts and methodologies to local communities. Finally, the third function is knowledge gap-filling, used to prevent neglect or underrepresentation of certain topics. Communication functions two and three may overlap and very well apply to researchers who can publish in the mainstream but instead choose non-mainstream journals to fulfil these roles. In this study we revisit the knowledge bridging function as described by Chavarro et al. (2017), situating it within a multidimensional framework of local research we recently developed (Di Césare & Robinson-Garcia, 2024). This framework conceptualizes locality through multiple dimensions, allowing for a more nuanced understanding of how knowledge circulates across geographical and linguistic boundaries. Building on this, we define knowledge bridging journals as those that serve local audiences –operationalized by the geographic concentration of citations received– while drawing from globally distributed knowledge sources –as indicated by the geographic origin of their cited references. We also limit our analysis to journals publishing in non-English languages. Accordingly, we identify knowledge bridging journals by a combination of global referencing, local citing, and non-English publishing. With this research, we intend to deepen our understanding of the role different journals play in a scientific communication still deeply intersected by the mainstream/non-mainstream divide, while granting them recognition as relevant sources beyond traditional impact indicators. 2. Data and methods We worked with a large OpenAlex dataset comprising 59,230 journals that published articles during 2023, enriched with Web of Science Master Journal List (MJL), Scopus, and DOAJ language data. We use the OpenAlex version provided by the InSySPo program through Google BigQuery (public database last updated March 2025) (Mazoni & Costas, 2024). We established a threshold to retain only those journals with a proportion of articles with references equal to or greater than 50% to avoid skewed reference processing. All computations were run in R (version 4.4.0). Both the R and SQL codes can be found in Di Césare (2025). Figure 1 provides an overview of how we operationalize the knowledge bridging function. We combine the use of references, citations, and language to identify knowledge bridging journals. The intervening variables, as well as their definitions, measurement methods, and thresholds are based on Di Césare & Robinson-Garcia (2024). The referenced proportion variable was defined as the proportion of appearance of the maximum referenced country per journal according to the geographical origin of references. Similarly, the citing proportion variable captures the maximum citing country based on the origin of citations. In addition, the non-English publishing variable considers whether the journal publishes in languages other than English according to MJL, Scopus, DOAJ, and OpenAlex data. In Appendix 1 we list examples of sources that meet the conditions to be identified as knowledge bridging journals. 2 Figure 1: Operationalization of the knowledge bridging communication function across local research dimensions. To build the journal subsets, we applied the third quartile of each continuous variable’s distribution as a cut-off point. For referenced proportion, we selected journals in which the most frequently referenced country accounted for less than 42% of all cited references –indicating a more global referencing pattern. For citing proportion, we included journals where over 86% of citations came from a single country, reflecting a locally concentrated readership. In cases where the same country is both the most frequently referenced and the top citing country for a given journal, but its proportions remain within the established thresholds, the journal is still considered to exhibit a global referencing pattern. In other words, a share of references from the top-referenced country below 42% is indicative of geographic dispersion, even if that same country also dominates the journal’s citing pattern (Appendix 1). Language was treated as a dichotomous variable, with all non-English outputs considered local. We identified language combining OpenAlex information with that of other databases, prioritizing the information obtained from Scopus, Web of Science, and DOAJ, when available. This methodological decision was made in light of the accuracy issues identified by Céspedes et al. (2025). As a result, we obtained three subsets of 44,270 journals with global references, 13,424 journals with local citations, and 7,984 journals with non-English language publications. 3. Results In Figure 2, we identify and characterize the knowledge bridging journals within the context of this OpenAlex dataset (N = 59,230). According to the conditions and thresholds outlined before, 1,271 journals (S1) simultaneously exhibit global references, local citations, and non-English language articles (A). Other possible intermediate subsets include those combining global references and non-English languages (S2, n = 4883), journals with local citations and non-English articles (S3, n = 3090), and those containing global references but local citations (S4, n = 6192). When comparing the knowledge bridging journals to the other subsets in terms of their average production (B) and average impact (C), we find that S2 published the most articles in 2023 (x p = 27) and received the most citations (x i = 28). In 3 contrast, the average production and impact of knowledge bridging journals (x p = 20, x i = 8), S3 (x p = 18, x i = 9), and S4 (x p = 19, x i = 7) were approximately one-third lower than the former mean values. Figure 2: Intersections for the (A) identification of knowledge bridging journals (S1), their (B) average scientific production and (C) average impact compared to subsets S2, S3 and S4. Figure 3 shows the total and percentage distribution of knowledge bridging journals’ production in the top 20 countries. In terms of total publications within journals that perform this function (A), Brazil clearly dominates the list with 5,260 articles, followed by Russia (n = 3,594) and Indonesia (n = 3,573). From Turkey (n = 1,363) and Spain (n = 1,298) onwards, values drop by half, with varied countries like Japan, Mexico, France, Argentina, China, and the United States each accumulating between 800 and 600 publications each. More than half of these top 20 countries belong to North America and Europe, with Europe contributing the most cases. However, their combined production (n = 9,285) does not surpass the total output of the South American and Asian countries in the list (n = 13,406). (A) also shows that, when considering these countries’ total 2023 output, their share of publications within knowledge bridging journals is very low, ranging from 0.08% in the case of China to 3.7% for Brazil. As for the share of publications of each country total output that is published in knowledge bridging journals (B), we observe that the distribution of countries changes significantly. The African Republic of Guinea-Bissau reports the highest proportion of articles in knowledge bridging journals (24%) relative to its 2023 total production, followed by Nicaragua (9%). Beyond these two, the remaining countries display quite homogenous shares below 5% and decreasing. Most of them are from South and Central America and the Caribbean, two regions that introduce countries at the highest places of this top 20 which are not present in (A) because of their more limited overall production of scientific papers. Such are the cases of Honduras, Paraguay, Bolivia, and Guatemala, whose 2023 production ranges between 20 and 50 articles. The middle section of this skewed distribution includes only a few Asian and European countries (e.g. Timor-Leste and Russia), while Africa contributes one more towards the end (Mozambique). Brazil and Russia stand out in both (A) and (B) because they are large 4 countries in terms of their overall scientific output that also contribute significantly to knowledge bridging journals. Figure 3: Top 20 countries within knowledge bridging journals arranged according to their (A) total production and (B) publications share. Figure 4 displays the distribution of knowledge bridging journals according to OpenAlex’s categorization into fields and domains. It is important to note that OpenAlex assigns these 5 categories to individual articles. As a result, when we aggregate them at the journal level, a single journal may be associated with multiple unique fields and domains simultaneously. The Social Sciences field of the homonymous domain presents the highest proportion of knowledge bridging journals (88%, n = 1119 journals), followed by Health Professions (54%, n = 685) from the Health Sciences domain, and Computer Science (53%, n = 676) from the Physical Sciences. The lowest proportions of knowledge bridging journals are found in Chemical Engineering (2%, n = 25) from the Physical Sciences, Veterinary (3%, n = 42), and Dentistry (5%, n = 68), both fields belonging to Health Sciences. The Social Sciences domain leads the distribution with 94% (n = 1195) of knowledge bridging journals containing articles assigned to one or more of its fields. Not far from those values, we find that 83% (n = 1049) of the knowledge bridging journals belong to the Physical Sciences, with fields like Environmental Science (52%, n = 666) and Engineering (35%, n = 442) being also quite predominant in them. The Health and Life Sciences domains are represented more moderately in these journals (70%, n = 888, and 59%, n = 744 respectively), with Medicine (51%, n = 651) closely following the Health Professions field within the first domain, and Agricultural and Biological Sciences (46%, n = 591) clearly dominating the second one. Figure 4: Distribution of the share of knowledge bridging journals by OpenAlex fields and domains categories. 4. Discussion In this work in progress, we have operationalized the knowledge bridging communication function to test Chavarro et al.'s (2017) definition. The results obtained so far indicate that knowledge bridging journals have a predominance of Asian and South American countries as article producers. We observe that these Global South regions have a stronger tendency to publish in their own languages in this type of journals, which are also largely grounded in knowledge from other parts of the world but primarily cited locally. The cases of Brazil and Russia are quite singular, as they are large scientific producers which also contribute 6 significantly to knowledge bridging journals. The Social Sciences stand out as the main domain into which these journals are categorized. However, the Physical Sciences also play a prominent role, meaning that concepts and methodologies from its fields could be communicated, translated, or adapted to local communities through the knowledge bridging journals. While our findings align with the model proposed in Chavarro et al. (2017), further research is needed to assess whether contributors to knowledge bridging journals are indeed primarily early-career researchers, as well as to address the persistent challenge of identifying knowledge bridging journals within anglophone communities. Moving forward, we also expect to delve deeper into the precise origin of references and citations, with the goal of establishing a bridge not only between mainstream and non-mainstream journals but also between scientific publications from the Global North and the Global South. Ultimately, we intend to continue working on the operationalization of the training and knowledge gap-filling communication functions, where the dimensions of the local intersect with epistemic approaches and academic trajectories. Open science practices We used open data from the OpenAlex database to conduct this research. Its origin and characteristics are fully described in the Data and methods section to allow for replicability. Furthermore, we shared the complete R and SQL code through an open repository on GitHub, which is also cited in this paper. Author contributions VDC: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing - original draft, Writing - review & editing. NRG: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing - review & editing. RC: Supervision, Writing - review & editing. Competing interests The authors declare no competing interests. Funding information This work is part of the COMPARE project (Ref: PID2020-117007RA-I00) funded by the Spanish Ministry of Science (Ref: MCIN/AEI/10.13039/501100011033 FSE invierte en tu futuro). Victoria Di Césare is currently supported by a FPI grant from the Spanish Ministry of Science (Ref: PRE2021-097022). Nicolas Robinson-Garcia is currently supported by a Ramón y Cajal grant from the Spanish Ministry of Science (Ref: RYC2019-027886-I). References Bianco, M., Gras, N., & Sutz, J. (2016). Academic Evaluation: Universal Instrument? Tool for Development? Minerva, 54(4), 399–421. https://doi.org/10.1007/s11024-016-9306-9 Céspedes, L., Kozlowski, D., Pradier, C., Sainte‐Marie, M. H., Shokida, N. S., Benz, P., Poitras, C., Ninkov, A. B., Ebrahimy, S., Ayeni, P., Filali, S., Li, B., & Larivière, V. (2025). Evaluating the linguistic coverage of OpenAlex: An assessment of metadata 7 accuracy and completeness. Journal of the Association for Information Science and Technology, asi.24979. https://doi.org/10.1002/asi.24979 Chavarro, D., Ràfols, I., & Tang, P. (2018). To what extent is inclusion in the Web of Science an indicator of journal ‘quality’? Research Evaluation, 27(2), 106–118. https://doi.org/10.1093/reseval/rvy001 Chavarro, D., Tang, P., & Ràfols, I. (2017). Why researchers publish in non-mainstream journals: Training, knowledge bridging, and gap filling. Research Policy, 46(9), 1666–1680. https://doi.org/10.1016/j.respol.2017.08.002 Di Césare, V. (2025). Local journals code (Version 1.0.0) [R, SQL]. https://github.com/vdicesare/Local.Journals (Original work published 2025) Di Césare, V., & Robinson-Garcia, N. (2024). What is local research? Towards a multidimensional framework linking theory and methods. Zenodo. https://doi.org/10.5281/zenodo.14033472 Larivière, V., Haustein, S., & Mongeon, P. (2015). The Oligopoly of Academic Publishers in the Digital Era. PLOS ONE, 10(6), e0127502. https://doi.org/10.1371/journal.pone.0127502 López Piñeiro, C., & Hicks, D. (2015). Reception of Spanish sociology by domestic and foreign audiences differs and has consequences for evaluation. Research Evaluation, 24(1), 78–89. https://doi.org/10.1093/reseval/rvu030 Mazoni, A., & Costas, R. (2024, June 6). Towards the democratisation of open research information for scientometrics and science policy: The Campinas experience. Leiden Madtrics. https://www.leidenmadtrics.nl/articles/towards-the-democratisation-of-open-research-i nformation-for-scientometrics-and-science-policy-the-campinas-experience 8 Tijssen, R. J. W., Mouton, J., van Leeuwen, T. N., & Boshoff, N. (2006). How relevant are local scholarly journals in global science? A case study of South Africa. Research Evaluation, 15(3), 163–174. https://doi.org/10.3152/147154406781775904 van Leeuwen, T. N., Moed, H. F., Tijssen, R. J. W., Visser, M. S., & Van Raan, A. F. J. (2001). Language biases in the coverage of the Science Citation Index and its consequences for international comparisons of national research performance. Scientometrics, 51(1), 335–346. Vessuri, H., Guédon, J.-C., & Cetto, A. M. (2014). Excellence or quality? Impact of the current competition regime on science and scientific publishing in Latin America and its implications for development. Current Sociology, 62(5), 647–665. https://doi.org/10.1177/0011392113512839 9