Citation Cliques in Low Impact Journals
Panagiotis-Alexios Spanakis, Grigorios Alexandrou, Diomidis Spinellis · 2026
Read Citation Cliques in Low Impact Journals by Panagiotis-Alexios Spanakis, Grigorios Alexandrou, Diomidis Spinellis (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-11930.txt.
Introduction
Panagiotis-Alexios Spanakis, Grigorios Alexandrou, and Diomidis Spinellis examine citation cliques in low-impact journals. Using Crossref records matched by subject and h5, they find authors in low-impact venues cite co-authors 6.7 times more often and cite back 4.7 times more than authors in high-influence venues. A hybrid screening pipeline flagged 277 outliers with 93.5 percent case purity, showing closed, inward-looking citation clubs that distort citation-count scores.
Catalog record
| CREATOR | Panagiotis-Alexios Spanakis, Grigorios Alexandrou, Diomidis Spinellis |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56349 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.11930
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Citation Cliques in Low Impact Journals Panagiotis-Alexios Spanakis, Grigorios Alexandrou, Diomidis Spinellis Abstract This exploratory study examines how low-impact journals, defined through subject-normalized Eigenfactor percentiles, are associated with denser and more reciprocating patterns of author-to-author citations. Using Crossref records, we assign journals to broad subject areas, compute subject-specific Eigenfactor scores, propagate venue quality to works and authors, match authors in low– (Case) versus high-influence (Control) venues by subject and h 5 h_{5} , and analyze citation edges for cohesion and anomalies. Across a 10% sample of 9 431 9\,431 matched pairs, authors in low-impact venues exhibit significantly higher cohesion: 6.7 × \times higher co-author citation rates and 4.7 × \times higher reciprocity in the aggregate Case–Control comparison. A subject-aware hybrid detection pipeline flags 277 outliers with 93.5% Case purity; these outliers display an 11 × \times clique-strength lift relative to non-outliers, revealing a stark “Two Worlds” segregation ( r = 0.71 r=0.71 ) where low-impact venues operate as closed citation economies. The largest detected component ( n = 23 n=23 ) displays a hub-and-spoke topology in which peripheral “Sycophants” funnel citations to central “Beneficiaries” through coordinated bursts, confirming a directed flow imbalance rather than reciprocal exchange among equals. Overall, cohesion, rather than broad asymmetry, accounts for the main Case–Control differences, suggesting that low-impact venues foster segregated, inward-looking citation economies that distort bibliometric indicators. Keywords: scientometrics; low-impact journals; citations; cliques; social network analysis; eigenfactor 1 Introduction Quantitative indicators such as citation counts, the h -index, and journal metrics have become central to the assessment of scholarly impact ( Garfield 2006a ; Hirsch 2005a ; Hicks et al. 2015a ) . These measures offer convenient proxies for influence, but they also create incentives for strategic behavior and manipulation ( Goodhart 1984a ; 2 ; Baccini et al. 2019a ) . When deployed uncritically, citation-based indicators risk rewarding volume or visibility over substance, with unintended consequences for research integrity and evaluation systems. Beyond well-known practices such as excessive self-citation ( Van Noorden & Chawla 2019a ; Wilhite & Fong 2012a ) , more coordinated behaviors have emerged. Authors may form tightly connected citation cliques — groups that disproportionately reference one another, often within low-impact journals as defined by metrics such as the Journal Impact Factor ( Garfield 1972a ) or subject-normalized Eigenfactor percentiles ( Bergstrom et al. 2008a ) . Such practices distort indicators, undermine trust in the scholarly record, and complicate fair comparison across fields. These dynamics exemplify what has recently been described as citation orchestration ( Evdaimon & Ioannidis 2024a ) , where citations serve strategic rather than intellectual functions. Low-impact journals provide fertile ground for these dynamics. Their emphasis on publication volume, rapid turnaround, and relaxed editorial oversight ( Beall 2012a ; Abalkina et al. 2025a ) creates environments where abnormal citation patterns can thrive. Systemic distortions such as national home bias ( Qiu et al. 2025a ) further interact with venue incentives, amplifying the risk that rankings and impact assessments reflect coordination rather than genuine uptake of ideas. This exploratory study investigates whether low-impact journals are significantly associated with cliques and abnormal citation patterns. Using large-scale bibliographic data, we implement a subject-aware pipeline that links journal influence to authors, matches researchers publishing in Bottom– and Top-tier impact journals of comparable productivity (by h_{5} index), and analyzes their citation networks for cohesion and anomalies. We find that extreme anomalies cluster predominantly among authors publishing mostly in low-impact venues: they concentrate in unusually dense, inward-looking structures and form compact outlier groups that set them apart from their Control counterparts. These results provide empirical evidence that venue incentives can shape citation behavior in ways that distort bibliometric indicators. Throughout this paper, authors publishing primarily in low-impact venues are referred to as Cases , and their matched counterparts in high-impact venues as Controls . Terms such as “syndicate,” “sycophant,” and “beneficiary” are used as structural labels describing observed network roles and do not imply intent or misconduct. Similarly we use the terms Control and Case to classify authors publishing mainly in high– or low-impact journals, again without implying any judgement or view regarding the authors’ and their works’ scientific worth. 2 Related Work Prior research has highlighted the fragility of citation-based indicators when used as proxies for scholarly quality. Performance measures not only shape incentives ( Goodhart 1984a ) , but also invite gaming, from excessive self-citation to coordinated cartels ( Biagioli et al. 2019a ; 2 ) . The wider literature on the “metrics culture” documents how quantitative indicators reshape scholarly communication, creating feedback loops between evaluation systems and researcher behavior ( Hicks et al. 2015a ; 2 ) . These mechanisms underpin the emergence of citation cartels ( Fister Jr. et al. 2016a ; Catanzaro 2024a ) , inflated impact factors ( Ioannidis et al. 2024a ; Catanzaro 2024a ) , paper mills ( Christopher 2021a ; Abalkina et al. 2025a ) , AI-generated papers ( Spinellis 2025a ) , and other forms of metric manipulation that erode the interpretive value of bibliometric indicators. Venue-level frameworks were developed to mitigate some of these distortions. The Eigenfactor metric ( Bergstrom et al. 2008a ) introduced a network-based approach that discounts journal self-citations and weights each citation by the prestige of the citing venue. This recursive weighting sought to capture influence rather than popularity and to counteract local citation loops. However, later analyses have shown that such corrections remain sensitive to database coverage, field size, and reference density ( Baccini et al. 2019a ) , indicating that structural biases persist even in sophisticated venue-level measures. At the author level, organized citation exchanges have been formalized as citation orchestration ( Evdaimon & Ioannidis 2024a ) . This framework quantifies concentrated citation flows through indicators such as the ratio of total citations to the squared author’s h -index ( C/h^{2} ), the number of distinct authors accounting for half of one’s citations ( A50\%C ), and the count of hyper-collaborators sharing over fifty joint papers ( A50 ). Together, these metrics reveal unusually cohesive collaboration networks that may signal collective metric inflation. Earlier studies of citation cartels and excessive self-referencing ( Fister Jr. et al. 2016a ; Baccini et al. 2019a ; Kojaku et al. 2021a ; Ioannidis 2025a ; Heneberg 2016a ) align with this view, emphasizing how reciprocal ties among small groups of authors can yield disproportionate bibliometric visibility. Such coordinated behaviors blur the line between collaboration and manipulation, challenging the assumption that citation networks represent organic intellectual influence. Systemic and infrastructural distortions further complicate these dynamics. National home bias inflates domestic citation counts, particularly in large scientific systems such as China’s ( Qiu et al. 2025a ) , while technical manipulation has emerged through practices such as “sneaked references”, which are inserted into metadata but absent from article text ( Besançon & Cabanac 2024a ) . Industrial-scale paper mills produce fabricated papers that populate citation networks with artificial linkages ( Abalkina et al. 2025a ) . Recent investigations reveal even more elaborate schemes, including the creation of fictitious author identities and peer reviewers to infiltrate legitimate journals. As reported by Naddaf 2025a , these fake personas — sometimes maintained across multiple journals — enable mills to generate both papers and favorable reviews,