Interlocking Editorship Networks

• Interlocking editorship is a phenomenon where scholars serve on multiple journal boards, creating quantifiable links through shared editorial memberships.
• Quantitative methods such as centrality metrics and the Jaccard coefficient are used to map policy diffusion and structural clusters within scholarly networks.
• This network structure highlights both benefits, like expedited best-practice dissemination, and risks such as editorial concentration and conflicts of interest.

Interlocking editorship refers to the network phenomenon wherein individual scholars serve concurrently on the editorial boards of multiple scholarly journals. This overlapping board membership generates explicit connections among journals, which serve as quantitative proxies for the similarity of editorial policies, disciplinary orientations, and institutional influences. The formal paper of interlocking editorship leverages techniques from social network analysis to understand both the structural properties of journal clusters and the dynamics of gatekeeping, policy diffusion, and concentration of editorial power within scientific fields.

1. Formal Structure and Measurement

Interlocking editorship is defined as a relationship between journals whereby two or more share one or more editorial board members. The network is constructed using a dual-mode affiliation structure (editors ↔ journals) which is subsequently projected into a single-mode network (journals only). In such networks, an undirected edge exists between journals $i$ and $j$ if $\exists$ an editor $e$ such that $e \in$ Board($i$) $\cap$ Board($j$). The edge weight $w_{ij}$ is given by $w_{ij} = |$Board($i$) $\cap$ Board($j$)$|$, representing the number of overlapping editors.

Quantitative analysis involves calculating standard centrality metrics:

Measure	Equation	Interpretation
Degree centrality	$d_i = \sum_j a_{ij}$	Direct connections
Closeness centrality	$C_i = (n - 1) / \sum_j d(i, j)$	Average proximity
Betweenness centrality	$$B_i = \sum_{s \neq i \neq t} \frac{\sigma_{st}(i)}{\sigma_{st}}$$	Brokerage/gatekeeping

where $a_{ij} = 1$ if journals $i$ and $j$ share at least one editor, $d(i, j)$ is the shortest path, and $\sigma_{st}$ denotes the number of shortest paths from $s$ to $t$ with $\sigma_{st}(i)$ those passing through $i$ (Barabesi, 2011, Baccini et al., 2011, Baccini et al., 2019, Baccini et al., 2023).

Similarity between journal boards is frequently measured using the Jaccard coefficient:

$J_{ij} = \frac{|A_i \cap A_j|}{|A_i \cup A_j|}$

where $A_i$ and $A_j$ are sets of editors (Baccini et al., 2019, Baccini et al., 2023, Baccini et al., 2020).

Dense subnetworks are identified using $m$-slice thresholds ($m$ = minimum overlapping editors), revealing tightly-knit communities and isolating peripheral journals.

2. Editorial Policy Diffusion and Homophily

The overlap of editorial board memberships is empirically validated as a proxy for editorial policy similarity: Journals sharing several editors are more likely to align in terms of research priorities and methodologies. Notably, highly interconnected journal pairs, such as Journal of Informetrics and Scientometrics (16 shared editors), manifest substantial policy commonality with a focus on quantitative methods (Barabesi, 2011).

Social and intellectual homophily is evident both within boards and across journals in the same network cluster. Community analysis demonstrates that clusters based on shared editors correspond with intellectual proximity (e.g. co-citation patterns) and author mobility. In fields such as information and library science, these communities are especially well-defined, reflecting strong associations between intellectual and social structuring (Baccini et al., 2019, Baccini et al., 2020, Baccini et al., 2023).

3. Cluster Structure and Subfield Segmentation

Detailed network studies reveal that interlocking editorship does not produce a uniformly connected field; rather, it gives rise to multiple cohesive subgroups and isolated clusters. For instance, Library and Information Science (LIS) and Management Information Science (MIS) form two primary weakly-linked subfields identifiable in the LIS network, with smaller professional or subject-matter journals remaining relatively isolated (Barabesi, 2011). Similarly, economics journal networks are characterized by a sparse global density (∼2–9%) but a large interconnected core ("giant component"), alongside specialized clusters for macroeconomics, theory, and applied domains (Baccini et al., 2011, Baccini et al., 2023).

Community detection algorithms (e.g. Louvain, Leiden) consistently reveal that clusters within the interlocking editorship network are structurally similar irrespective of whether the analysis considers all editors, only editorial leaders, or excludes leaders. This redundancy of network links is interpreted as social and intellectual homophily within and between boards (Baccini et al., 2023).

4. Implications for Editorial Power, Gatekeeping, and Diversity

Network centrality metrics identify journals and individual editors who serve as gatekeepers or opinion leaders in their fields. Highly central journals (by degree, closeness, betweenness) exert pivotal influence over the evaluation criteria and the shaping of disciplinary discourse. The concentrated selection of prominent editors as board members, especially in high-prestige journals, amplifies the power of certain individuals and elite institutions (e.g. U.S. and U.K. universities account for the majority of economics editorial seats) (Baccini et al., 2023).

Significant risks are associated with editorial concentration:

• Homogenization of editorial policies, reducing diversity in research perspectives.
• Barriers for innovative research and emerging scholars.
• Reinforcement of institutional and geographic dominance (e.g., ∼33–36% of economics board seats held by U.S.-based scholars).

Conversely, interlocking editorship can expedite the dissemination of best practices and methodological convergence, improving consistency across journals. In select cases (i.e., journals focusing on gender studies), there is evidence for strategic improvement of gender diversity on boards; however, most networks remain male-dominated (∼24–25% women editors) (Baccini et al., 2023).

5. Conflict of Interest, Favoritism, and Regulation

Recent empirical work demonstrates that interlocking editorship may also create adverse effects in editorial decision-making. Editors frequently handle submissions from recent collaborators, constituting a conflict of interest (COI) that persists despite explicit recusal policies. Analysis of 500,000 papers handled by 60,000 editors shows that nearly 3% of journals have a COI rate of ≥10%, and that papers with COI are accepted significantly faster—suggesting potential favoritism. Regression discontinuity analyses indicate that policy interventions have at most modest effects in curbing COI behavior (Liu et al., 2023).

Table: Conflict of Interest Prevalence

COI Rate (%)	Journals	Example Titles	Notes
≥10	∼3%	PLOS Medicine (17%)	High impact journals
≥2	~50%	Journal of Fungi (14%)	Widespread

Recommended regulatory responses include more robust enforcement, mandatory disclosure, audits, and automated COI flagging.

6. Comparative and Multilayer Network Models

Analysis of combined network layers—based on co-citations (CC), interlocking authorship (IA), and interlocking editorship (IE)—demonstrates that the fused journal similarity network captures the strongest structure from the editorship layer ($\sqrt{R_d} \approx 0.8$ for IE, vs. $0.45–0.55$ for CC/IA) (Baccini et al., 2020). Partial distance correlation analysis confirms that clusters in the fused network align most closely with editor overlaps, reinforcing the view that editorial gatekeeping defines the boundaries of scholarly communities more persistently than co-authorship or citation patterns.

Multilayer network fusion is formalized as:

$J(A,B)=\frac{|A \cap B|}{|A \cup B|}$

with similarity matrices fused via iterative cross-diffusion updates, borrowing ties from k-NN neighborhoods and integrating structures across layers (Baccini et al., 2020).

7. Evolution and Future Directions

As disciplinary fields mature, interlocking editorship networks are predicted to achieve higher density and centralization—reflecting growing consensus on research aims and methodologies, as well as the consolidation of editorial power within central journals. Longitudinal analyses may reveal whether such consolidation fosters scholarly innovation or entrenches paradigmatic orthodoxy (Barabesi, 2011, Baccini et al., 2011). Comparative cross-field studies enable the evaluation of the uniqueness of network patterns, providing insight into domain-specific mechanisms of editorial influence.

An outstanding area for future research is the differentiation of editor roles (e.g. editor-in-chief vs. advisory board) and their respective impact, along with causal evidence connecting editorial overlap to actual publication outcomes. Enhancement of policy and assignment algorithms, transparency practices, and structural reforms are central to preserving the integrity of editorial decision-making and the broader scholarly enterprise.

• (Barabesi, 2011): Network analysis of editorial boards in Information Science & Library Science.
• (Baccini et al., 2011): Interlocking editorship in economics journals.
• (Baccini et al., 2019): Intellectual and social similarity among scholarly journals.
• (Baccini et al., 2020): Similarity network fusion for scholarly journals.
• (Baccini et al., 2023): Geographic diversity, gender composition, and interlocking editorship in economics journal boards.
• (Liu et al., 2023): Empirical paper of editors handling collaborator submissions and COI policy effectiveness.