ArticlePDF Available

Academic influence and invisible colleges through editorial board interlocking in communication sciences: a social network analysis of leading journals

Authors:

Abstract and Figures

Editorial boards (EBs) play a crucial role in setting journals’ scientific output, generally determining what legitimate science is and thus establishing a benchmark on what should be published. EB members are generally prominent scholars in a given discipline and, as such, are habitually connected to several journals’ EB, a phenomenon known as editorial board interlocking. This exploratory study analyses the EB interlocking in a sample of communication journals. Specifically, we apply social network analysis to investigate how 41 JCR (Journal Citation Report) communication journals are connected through EB interlocking. Using graph theory and social network analysis, we identified the scholarly journal network, the high influential journals and scholars, and the cohesive subgroups (i.e. invisible colleges) within the communication field. Our findings shed some important light on the network structure of EBs in communication sciences, arguing that the exploration of editorial interlocks is a complemental approach to understand academic journals’ influence within academic fields.
Content may be subject to copyright.
Vol.:(0123456789)
Scientometrics
https://doi.org/10.1007/s11192-020-03401-z
1 3
Academic inuence andinvisible colleges througheditorial
board interlocking incommunication sciences: asocial
network analysis ofleading journals
ManuelGoyanes1,2· Luisde‑Marcos3
Received: 17 July 2019
© Akadémiai Kiadó, Budapest, Hungary 2020
Abstract
Editorial boards (EBs) play a crucial role in setting journals’ scientific output, generally
determining what legitimate science is and thus establishing a benchmark on what should be
published. EB members are generally prominent scholars in a given discipline and, as such,
are habitually connected to several journals’ EB, a phenomenon known as editorial board
interlocking. This exploratory study analyses the EB interlocking in a sample of commu-
nication journals. Specifically, we apply social network analysis to investigate how 41 JCR
(Journal Citation Report) communication journals are connected through EB interlocking.
Using graph theory and social network analysis, we identified the scholarly journal network,
the high influential journals and scholars, and the cohesive subgroups (i.e. invisible colleges)
within the communication field. Our findings shed some important light on the network struc-
ture of EBs in communication sciences, arguing that the exploration of editorial interlocks is
a complemental approach to understand academic journals’ influence within academic fields.
Keywords Editorial boards interlocking· Communication sciences· Journals· Social
network analysis· JCR· Diversity
Editorial boards are crucial agents in the governance and development of academic disci-
plines (Willett 2013). As gatekeepers of knowledge (Metz etal. 2016), play a key role in
shaping what is published and thus what informs theory development, research, and prac-
tice (Mauleón etal. 2013; Burgess and Shaw 2010). Extant research on bibliometric and
scientometric studies (Feldman 2008; Mauleón et al. 2013; Teixeira and Oliveira 2018)
suggest that potential candidates to scientifically join the editorial board of a top-tier jour-
nal should meet certain criteria, such as a strong record of publications, a great expertise
and prestige in the field, and considerable amount of peer citations. Editorial policies are
typically deployed by scholars who are members of the board (Burgess and Shaw 2010).
* Manuel Goyanes
mgoyanes@hum.uc3m.es
1 Department ofCommunication, Carlos III University, Getafe, Madrid, Spain
2 University ofSalamanca, Political Science, Salamanca, Spain
3 Department ofComputer Science, University ofAlcalá, Alcalá, Spain
Scientometrics
1 3
The crossed presence of academics in different boards is a phenomenon called interlock-
ing editorship and is considered a proxy of the similarity of editorial policies (Teixeira
and Oliveira 2018). This article explores how communication journals and editorial board
members are connected through editorial board interlocking.
Through a social network analysis of 41 JCR (Journal Citation Report) journals indexed
between quartile one (Q1) and quartile two (Q2) in the category of “communication,” we
try to elucidate how editorial board interlocking establishes and creates a social structure
within communication journals. Therefore, our aim is to examine how the memberships of
editorial boards of major journals are linked and identify influential scholars, nationalities,
and academic affiliations (i.e., universities). In short, the aim of this paper is to: 1) map
the editorial board interlocking of the communication field, 2) identify the most influential
scholars, journals, academic affiliations and nationalities that shape communication edito-
rial boards and 3) identify the cohesive subgroups of scholars’ and journals’ networks (i.e.
invisible colleges). This study contributes to a growing body of scholarship on the rela-
tional networks of editorial boards and its implications for the development of the form and
contents of top-tier communication journals.
Editorial boards interlocking
One of the most important factors for academic success represented in tenure and hiring
decisions is based on publication in prestigious, indexed, peer-reviewed journals (Zdeněk
2018). The quality of such research is often validated under the form of journals impact
rankings (Dahdouh-Guebas et al. 2003) in such a way that publication in higher impact
journals is equated with quality of scholarships, reputation, and status (Kurmis 2003). In
today’s’ academy, there is little doubt that scientific journals provide the principal means of
science dissemination (Goyanes 2017; Demeter 2018), and publishing in top-ranked jour-
nals has now become an essential academic activity (Burgess and Shaw 2010; Mauleón
etal. 2013). However, publication in leading journals is not a simple task, and therefore,
the production process requires a chain of agents to ensure certain quality standards. One
of such agents is the journal editorial or advisory board.
According to extant research editorial boards (EBs) assist the journal’s editor(s) in three
major ways (Willett 2013): (1) promote the journal, (2) serve as referees when articles are
submitted for review and (3) provide strategic and tactical advice to the journal future devel-
opment. Their role in shaping journals’ output and strategy is reflected by the “gatekeepers
of knowledge”, a metaphor that recognizes their key function in the peer-review process and
thus in the establishment of the norms and values that shape legitimate science (Rosenstreich
and Wooliscroft 2006; Metz etal. 2016). As EB members, they contribute to ensuring the
quality of scientific journals (Willett 2013), are usually selected according to their experience
and prestige in their field (Feldman 2008), and occupy a strategic position in the scholarly
debate (Barzilai-Nahon 2009). Being invited to join the board of a top-tier journal is a highly
prestigious appointment and an important form of academic recognition (Mauleón et al.
2013) favorably regarded in academic promotion processes (Bedeian etal. 2009).
As an institutional body in the governance of academic communities (Dhanani and
Jones 2017), EBs play a decisive role by delineating who and what deserves to be pub-
lished and thus legitimizing certain research topics and scholars. As a result, EBs have a
significant influence on academic careers and progression, signalling the issues, priorities,
and approaches that are worth to be studied, problematized, or even discussed further. In
Scientometrics
1 3
doing so, they provide status and visibility to certain scholars, by conferring authority and
recommending their work to be published (Willett 2013), providing also the technical and
tactical strategies to improve or set the future development of journals (Burgess and Shaw
2010; Petersen etal. 2017).
EB members are appointed by invitation (Dhanani and Jones 2017), generally by those
at the top of the EB hierarchy, i.e., editors-in-chief or journal editors. Both instances are the
most influential agents in the recruitment process, holding extensive discretion over their
selection (Metz etal. 2016). According to Mauleón etal. (2013), the existence of a previ-
ous female editor-in-chief predicts the presence of women in EBs, explained by the gender
homogeneity in academic networks (Burgess and Shaw 2010). As appointment to member-
ship is usually by invitation (Lee 1997), many EB members are recommended by former
editors and academic friends (Feldman 2008). In short, the selection of EB members is
usually based on the trajectory and excellence of scholars, which generally means pub-
lishing in top journals (Burgess and Shaw 2010). However, recent scholarship cast doubt
on this rational view. For instance, Bedeian etal. (2009) show that there is no significant
correlation between board memberships and publications, suggesting that the meritocratic
argument is not supported by empirical data and thus, academic patronage might be the
best explanation (Burgess and Shaw 2010).
The names and positions of scholars in the EB hierarchy are usually displayed in jour-
nals’ print versions and on the official home page of journals’ websites (Burgess and Shaw
2010). Therefore, EB members are often visible, although the recruitment process by
which individuals become members is not so transparent (Ozbilgin 2004). Although EB
selection has become more formalized over time (Cascio 2008), it is generally assumed
that high academic performance (mainly in terms of one’s publication record) is a criterion
in the selection of EBs (Dhanani and Jones 2017). Therefore, the scientific reputation of
a journal is the product of the reputations of its EB (Willett 2013), signaling the “status
of the journal to potential subscribers, authors and researchers” (Burgess and Shaw 2010:
630).
As previously outlined, EB members are usually prominent scholars with a strong
record of publications (Teixeira and Oliveira 2018) and highly appreciated by scientific
peers (Andrikopoulos and Economou 2015). As in the business realm, an EB member may
occupy a seat in more than one directory board, a phenomenon known as editorial board
interlocking (Baccini and Barabesi 2010). Several disciplines have analysed journals’ gov-
ernance more than others, but there is a general consensus that EB interlocking creates
a network structure of elite scholars, influencing the employment of particular research
methods, topics, and theories (Teixeira and Oliveira 2018).
The potential consequences of EB interlocking also relate to the creation of a subgroup
of scholars linked to some core journals, thus directly exerting influence on the vision and
main paradigms of such journals. Since EBs’ opinions, suggestions, visions, expectations,
and reviews are reflected in many different journals, they can establish clear patterns of
knowledge production/dissemination. As a result, editorial interlocks could create ‘invis-
ible colleges’ which may develop their own norms, values and standards, monitoring also
the progress of the field (Zuccala 2006). Previous studies have examined interlocking edi-
torship within Economics (Baccini and Barabesi 2010), Information and Library Sciences
(Baccini and Barabesi 2011), Finance (Andrikopoulos and Economou 2015), or Knowl-
edge Management (Teixeira and Oliveira 2018). However, none work had considered EB
interlocking in Communication Sciences, despite the crucial implications for knowledge
production that this phenomenon might entail. This study seeks to palliate this gap by
Scientometrics
1 3
identifying the most influential journals and scholars and mapping how they are structur-
ally connected in a social network.
Method
Data collection
Data for this study came from the public web pages of the selected journals. We decide to
take the Journal Citation Report (JCR) to examine our research questions because it is the
most influential ranking in sciences, usually taken as reference in academic promotions,
scholars’ evaluations, and research founding (Alvesson etal. 2017). Similarly, we select
2016 for the analysis because we started our data collection in 2017, and therefore 2017
was the most recent data available. Journals indexed by the JCR ranking are divided into
four different quartiles, according to their impact factor: Q1, Q2, Q3 and Q4. Those jour-
nals with a higher impact factor are thought to be more influential than those with lower
ones, and hence Q1 and Q2 journals are arguably the most influential in a given field. In
communication during 2016, there were 79 JCR journals in total, of which 41 were indexed
in Q1 and Q2. “Appendix 1” presents the journals included in this study. For this study, we
analyse the gender composition, the organizational affiliation, and the country of affiliation
of each EB.
The country of affiliation was coded according to the geography in which EB members
have their current academic affiliation (1 = USA, 2 = Asia-Pacific, 3 = Africa and Middle
East, 4 = Europe, 5 = Israel, 6 = UK, 7 = Canada, 8 = Latin America). We follow previous
studies for the coding (Goyanes 2019). We combine continent level, country level, and geo-
graphical level to present the data in a coherent and concise manner. For example, we con-
sider the U.S., Israel, UK, and Canada as independent values at country level because they
are some of the most important countries in terms of total EB members. By contrast, we
consider Africa and Middle East and Latin America as a conglomerate of nationalities at
a continent-regional level because their contribution to the total number of EB members is
limited.
Social network analysis
Connections between scholars and journals in EBs can be analysed by Social Network
Analysis (SNA) methods (Andrikopoulos and Economou 2015; Baccini and Barabesi
2010, 2011; Liwei and Chunlin 2015). SNA studies the connections in social structures
through the use of graph theory. Actors of the network are called nodes. The relationships
or connections between actors are represented as edges between nodes. The social network
of journals and members of EBs can be represented as a bipartite graph. Bipartite networks
have two sets of nodes that usually receive different names: actors and events. The edges of
the bipartite network only connect nodes from opposite sets. In the social network of EBs,
actors represent members of EBs (scholars), events represent journals, and edges represent
a membership relation between a scholar and the EB of a journal. Bipartite networks are
also called affiliation networks or two-mode networks.
A bipartite (two-mode) network can be projected into a unipartite (one-mode) network
that contains only nodes of one set. Nodes in the one-mode projected network are con-
nected if they have an edge to a common node in the original network. For instance, in a
Scientometrics
1 3
one-mode projected network of members of EBs, two scholars are connected if they serve
in the EB of at least one common journal. In a one-mode projected network of journals,
two journals are connected if one or more scholars are part of the EB both journals. In a
one-mode weighted projected graph, each edge of the resulting network contains an inte-
ger value that represents the number of common neighbors (weight) between the nodes
in the original network. For instance, if two journals share ten members of their EBs,
the weighted projected graph of journals has an edge between them with a value of ten.
Although one-mode projections reduce the dimensionality of bipartite graphs by compress-
ing information, it is a common analytic approach in SNA because it facilitates the visuali-
zation of large networks. Besides, several analytical methods can only be used in one-mode
networks.
In this study, SNA was used to analyze the bipartite network formed by journals and
scholars in which connections represent a relation of membership to the EBs. Python’s
NetworkX package1 was used to create the graphs. NetworkX was also used to compute the
metrics of nodes, and create weighted projections of the bipartite network. Graph figures
were produced with the Gephi network visualization software ( Bastian etal. 2009) using
the ForceAtlas2 layout algorithm. The visualization of social networks conveys the results
of the analysis showing patterns of connections and of properties for the actors.
Metrics
SNA provides a variety of metrics to quantify different properties of the network, nodes,
and edges. Following previous works (Andrikopoulos and Economou 2015; Baccini and
Barabesi 2010, 2011; Liwei and Chunlin 2015; Teixeira and Oliveira 2018), the social net-
work of EBs membership formed by scholars and journals was analysed using SNA met-
rics. Density quantifies the number of connections of the network in relation to the maxi-
mum possible number, and it is useful to compare different networks. Centrality measures
quantify the importance or influence of a particular node in the network. In this study, we
analyze the following centrality metrics of journals and members of their EBs: degree cen-
trality, closeness centrality, and betweenness centrality.
Degree centrality measures the number of connections of a node. In this study, degree cen-
trality of a given node is computed as the fraction of nodes connected to it. It is normalized by
dividing by the maximum possible number of connections, which is different for each set of
a bipartite graph. Closeness centrality represents the distance to the centre of the network and
can be regarded as a measure of how long it will take to spread information from a given node.
It measures the average shortest distance to all other nodes of the network. In this study, close-
ness centrality is normalized by the minimum possible distance (Borgatti and Halgin 2014).
Higher values represent more centrality. The maximum possible closeness centrality is one.
A value of one means that the node is connected to all other nodes by the minimum possible
distance, which is one for all nodes in the other set (e.g., a scholar is member of the EB of all
journals) and two for the nodes in the same set (e.g., a journal shares at least one common
member of its EB with all other journals). Betweenness centrality measures the number of
times that a given node is part of the shortest path between two other nodes. Betweenness
centrality reflects the amount of influence a node has over the flow of information or resources
in the network. It can be used to find nodes that act as a bridge between different parts of the
1 NetworkX package documentation https ://netwo rkx.githu b.io/.
Scientometrics
1 3
network. In this study, betweenness centrality of a given node is computed as the sum of the
fraction of all pairs of shortest paths that pass through it. The value is normalized by the maxi-
mum possible value, i.e., the maximum number of shortest paths of the complete graph, which
for a bipartite graph is limited by the relative size of the two sets of nodes (Borgatti and Halgin
2014).
Cohesive groups
In one-mode social networks, cohesive groups are subsets of actors among whom there are
strong or intense ties (Wasserman and Faust 1994). In this work, we analyzed the cohesive
subgroups of the two different one-mode weighted projections: the projection of members of
EBs (actors) and the projection of journals (events). Cohesive groups of actors are subsets
of scholars that interlock in multiple EBs. Cohesive groups of journals are subsets of events
that interlock multiple members. In one-mode weighted networks, cohesive subgroups will be
determined by subsets of actors or events with numerous connections or with connections that
show high weight values.
Cohesive groups were assessed by using two different strategies in the one-mode weighted
projections: k-cores and the weight of connections (m-slices). A k-core is a subgraph in which
each node is adjacent to at least k others. A k-core of actors will include only members of
EBs that are connected with at least k other members by serving in common EBs. A k-core
of events will include the journals that are connected with at least k other journals by having
common members in their EB. We can consider different threshold values of k to form cohe-
sive subgroups. Subsets of actors or events that have more connections will be part of cohesive
groups at higher threshold levels, whereas actors or events that have fewer connections will
only appear in cohesive groups at less strict threshold levels. An m-slice is a subgraph that has
the edges with weight equal to or larger than m, as well as the nodes that are adjacent to these
nodes. An m-slice of actors will include the actors among whose connections the weight is
equal to or larger than m, i.e., the members that serve in at least m common EBs. Similarly, an
m-slice of events will include the journals that share at least m members. We can also consider
different threshold values of m to form the cohesive subgroups. Subsets of actors or events
among whose ties have large values can be part of cohesive groups at higher threshold lev-
els, whereas actors or events among whose connections have small values would only appear
in cohesive groups at less strict threshold levels. Although analyzing the derived networks
at increasing values of k and m results in a hierarchical series of cohesive subgroups, previ-
ous studies suggest that both thresholds can be chosen ad-hoc by the researcher (Teixeira and
Oliveira 2018).
The two strategies show different aspects of subgrouping. While the k-core approach shows
if the network has a highly connected core, the weight of connections (m-slices) may show the
presence of cohesive subgroups in which connections are more intense, but such groups may
not necessarily be connected among them.
Results
Social network
The network of journals and members of EBs has 2097 nodes (2056 scholars and 41 jour-
nals) and 2719 edges. Figure1 shows the central connected component of the social network
Scientometrics
1 3
of members and journals represented as a graph. At a visual interpretation, we observe a
fracture between the upper left part of the network and the central part, which included most
journals and scholars. Journals that are in the central part form close ties by sharing mem-
bers of their EBs. We also observe a smaller fracture, which included the three journals that
are close together in the right: J. of Advertising, Int. J. of Advertising and J. of Advertis-
ing Research. These journals share many members of their EBs between them but only a
small number with the central part. There are also several journals in the periphery which
share only one or two members with other journals of the central part of the network. These
journals are Research on Language and Social Interaction (bottommost journal in Fig.1),
Telecommunications Policy, and Comunicar (both in the bottom left in Fig.1). The network
of journals and members of EBs is classified as a disconnected network since three journals,
Written Communication, IEEE Transaction on Professional Communication, and Technical
Communication, do not share any member of their EB members with any other journal.
Each of them forms a separate component (not included in Fig.1).
Fig. 1 Social network of members of the editorial boards of communication journals. White nodes repre-
sent journals. Size is proportional to the number of members. Grey nodes represent members. (Color figure
online)
Scientometrics
1 3
The density of the bipartite network is low (0.03). The density of the projected one-
mode network of scholars is 0.05, which is also low. The density of the projected one-mode
network of journals is 0.32 which is significant and higher than the values reported in other
fields: 0.22 in Knowledge Management and Intellectual Capital (Teixeira and Oliveira
2018), 0.09 in Library Science (Baccini and Barabesi 2011), 0.02 in Economics (Baccini
and Barabesi 2010) and 0 in Finances (Andrikopoulos and Economou 2015). The density
of the network represents the occurrence of interlocking between EBs. Since interlocking is
interpreted as alignments between journals, a higher-density may suggest that journals are
aligned in their policies and preferences; thus existing a state of agreement or cooperation
among EBs towards common causes and viewpoints. This may result in a lack of balance
in the preferred topics, theories, and methods suggesting the existence of an elite group of
researchers and journals in the field of communication. However, previous works analysed
larger sets of journals so we have to be careful when comparing density since it may also
be that the use of high-ranked journals in our study is the main source of higher density.
Central actors
The average degree of editors in the two-mode network is 1.31 (normalized degree average
is 0.032, SD = 0.020). The number of memberships ranges from 1 to 8 (normalized degree
centrality range is 0.024–0.195). The distribution presented in Table1 shows that a sub-
stantial number of scholars (1638) are only connected with one journal. Table2 presents
scholars with the highest degree centrality. Journals have an average of 65.43 members in
their EBs (normalized degree centrality of journals M = 0.032, SD = 0.013).
Average closeness centrality of scholars (M = 0.392, SD = 0.050) and jour nals
(M = 0.245, SD = 0.032) shows that the distance among members of EBs and journals is
small. Closeness centrality is normalized, and the maximum value is 1, which represents
the minimum possible distance. Scholars with the highest closeness centrality are pre-
sented in Table3. Values are particularly high, suggesting that these scholars are close to
all other nodes of the network.
Betweenness centrality of scholars (M = 0.001, SD = 0.004) suggests that, on average,
around 0.1% of all the shortest paths go through a given scholar. Top 10 highest ranked
scholars are presented in Table4. The highest value is 0.058. Since the value is normal-
ized to the maximum number of shortest paths and the number of nodes is high (there are
also many possible paths), even low values represent a significant amount of betweenness
Table 1 Distribution of degree
for scholars in the bipartite
network of journals and members
of EBs
Degree Frequency
1 1638
2 277
3 87
4 27
5 11
6 10
7 5
8 1
Scientometrics
1 3
centrality. Journals average betweenness centrality is 0.055 (SD = 0.036), meaning that, on
average, 6.1% of the shortest paths go through a journal. The top-ranked journal in terms
of betweenness centrality is the Journal of Communication, with a value of 0.166. Since
journals are connected with many scholars (higher degree centrality), more shortest paths
also go through them.
Table 2 Scholars with highest degree centrality
Pos. Name Affiliation G. Geo-
graphic
area
Degree Normal-
ized
degree
1 Michael Delli Carpini Univ. of Pennsylvania M 1 8 0.195
2 S. Shyam Sundar The Pennsylvania State Univ M 1 7 0.171
2 Zizi Papacharissi Univ. of Illinois at Chicago F 1 7 0.171
2 Claes de Vreese Univ. of Amsterdam M 4 7 0.171
2 Patricia Moy University of Washington F 1 7 0.171
2 David Tewksbury Univ. of Illinois, Urbana-Champaign M 1 7 0.171
3 Homero Gil de Zúñiga Univ. of Vienna M 4 6 0.146
3 Natalie Jomini Stroud Univ. of Texas at Austin F 1 6 0.146
3 Ronald Rice Univ. of California, Santa Barbara M 1 6 0.146
3 Michael Slater Ohio Northern Univ M 1 6 0.146
3 W. Lance Bennett Univ. of Washington M 1 6 0.146
3 Daniel Hallin Univ. of California San Diego M 1 6 0.146
3 Sharon Dunwoody Univ. of Wisconsin-Madison F 1 6 0.146
3 Jesper Stromback Univ. of Gothenburg M 4 6 0.146
3 Jörg Matthes Univ. of Vienna M 4 6 0.146
3 R. Lance Holbert Temple Univ M 1 6 0.146
Table 3 Scholars with highest closeness centrality
Pos. Name Affiliation G. Geo-
graphic
area
Closeness cent.
1 Zizi Papacharissi Univ. of Illinois at Chicago F 1 0.569
2 S. Shyam Sundar The Pennsylvania State Univ M 1 0.557
3 Homero Gil de Zúñiga Univ. of Vienna M 4 0.548
4 Carolyn Lin Univ. of Connecticut F 1 0.545
5 David Tewksbury Univ. of Illinois, Urbana-Champaign M 1 0.544
6 Natalie Jomini Stroud Univ. of Texas at Austin F 1 0.544
7 Michael Delli Carpini Univ. of Pennsylvania M 1 0.536
8 Silvia Knobloch-Westerwick The Ohio State Univ F 1 0.533
9 Travis Dixon Univ. of Illinois, Urbana-Champaign M 1 0.526
10 Ronald Rice Univ. of California Santa Barbara M 1 0.525
10 Louis Leung Hong Kong Shue Yan University M 2 0.525
Scientometrics
1 3
Gender andgeographic representation
In the dataset of members of EBs, 766 are females (37.26%), and 1247 (60.65%) are males.
There are also 43 members (2.09%) whose gender could not be determined. Figure2 pre-
sents a graphical representation of the gender distribution of the central connected com-
ponent of the network. Although we can observe journals with more female members and
journals with more male members, there is no clear pattern. Scholars with highest degree
centrality (Table2) are two females and four males. If we look at scholars with a degree of
three or more (top 141), we find 45 females (31.91%) and 96 males (68.09%). So, regard-
ing the number of memberships, females are underrepresented in the top positions. Top 10
scholars with the highest closeness centrality are four females and six males (Table3) sug-
gesting a distribution closer to the distribution of the complete dataset. However, it is not
mirrored in the top 100, which includes 30 females and 70 males. So male members also
occupy more central positions in the network. Top 10 scholars with the highest between-
ness centrality are five females and five males (Table4), while the top 100 are 42 females
and 58 males. Although females are also underrepresented, the distribution of top scholars
regarding betweenness centrality is similar to the distribution of the dataset.
Network metrics for members of EBs do not follow a normal distribution. Non-para-
metric methods that compare the medians of groups were then used to analyze the inter-
relationships among scholars. Medians of the three network metrics return the same value
when grouped by gender. For instance, if we consider degree centrality, most scholars are
members of only one EB. For this reason, we decided to perform the analysis with a sub-
set of the dataset that included only scholars that are part of two or more EBs (N = 418).
Although the median of degree centrality is still the same for male and female members,
small differences can be found regarding closeness centrality and betweenness central-
ity. However, they are not statistically significant (H = 1.83, p = 0.176, and H = 0.16,
p = 0.689 respectively in Kruskal–Wallis tests). We also repeated the analysis with the
subset of scholars that are members of three or more EBs (N = 141) finding the same
results (H = 0.00, p = 0.981, and H = 0.32, p = 0.570 respectively in Kruskal–Wallis tests).
This suggests that although female members are less in overall numbers, they occupy
Table 4 Scholars with highest betweenness centrality
Pos. Name Affiliation Gender Geo-
graphic
area
Between. cent.
1 Shintaro Okazaki King’s College London M 6 0.058
2 Kathleen Tyner Univ. of Texas at Austin F 1 0.045
3 Sharon Dunwoody Univ. of Wisconsin-Madison F 1 0.041
4 S. Shyam Sundar Pennsylvania State Univ M 1 0.041
5 Zizi Papacharissi Univ. of Illinois at Chicago F 1 0.038
6 Rich Ling Nanyang Technological Univ M 2 0.036
7 Karen Tracy Univ. of Colorado F 1 0.034
8 Jörg Matthes University of Vienna M 4 0.033
9 Nick Yee Standford Univ M 1 0.029
10 Carolyn Lin Quantic Foundry F 1 0.027
Scientometrics
1 3
similar positions in terms of the number of connections, positioning, and distance to other
members.
Descriptive statistics of network metrics are similar for the different geographic areas
(Table 5). Figure3 presents the geographic distribution of the affiliation of members of
EBs (as reported by journals). One individual affiliated to the World Economics Journal
was not included. We observe that members affiliated to USA institutions dominate the
central part of the network. The left part shows the most diversity in terms of geographical
representation, including an important number of members from European and UK institu-
tions. Table6 presents the geographic distribution of top-ranked scholars for each of the
network metrics considered in this study. Members affiliated to institutions of the USA
are overrepresented in the sample (61.86%) and top positions. Assuming that top-ranked
individuals are the central actors of the network, the representation of institutions from the
USA is even higher in top positions regarding the number of connections (70.2%), distance
to other actors (82%), and communications paths that go through them (73%). Members of
Fig. 2 Gender distribution of the social network of members of the editorial boards of communication jour-
nals. Light grey nodes represent male members. Dark grey nodes represent female members. White nodes
represent journals, and members whose gender could not be determined
Scientometrics
1 3
European institutions are particularly underrepresented in the top positions of all metrics:
10.6% for degree centrality, 7% for closeness centrality, and 8% for betweenness centrality,
while they represent 15.03% of all members of the dataset.
Medians of degree centrality and betweenness centrality return the same value when
members are grouped by geographic area. Medians of closeness centrality are significantly
different when grouped by geographic area (H = 123.09, p < 0.001 in a Kruskal–Wal-
lis test). Scholars of institutions from the USA (Mdn = 0.402), Israel (Mdn = 0.409), and
Middle East-Africa (Mdn = 0.393) occupy central positions in the network. Scholars
from Asian (Mdn = 0.378), European (Mdn = 0.367), UK (Mdn = 0.373) and Canadian
(Mdn = 0.373) institutions follow. Scholars from Latin American institutions tend to be in
peripheral positions (Mdn = 0.348).
Cohesive subgroups
Cohesive groups were used to evaluate the existence of segmentation in communication
journals and scholars. The 5-core of journals is the network in which each journal is con-
nected (shares a common member of its EB) with at least the other 5 journals through EB
interlocking. We selected a k value of 5 to compare with previous studies. The 5-core of
journals (Fig.4) includes 32 journals and has a density of 0.50, which is higher than the
density of the network of journals (0.32). The discipline of Communication has more jour-
nals (32 of 41) in its core when compared with other disciplines like Knowledge Manage-
ment and Intellectual Capital, which reported a core of 13 out of 27 journals (Teixeira and
Oliveira Teixeira and Oliveira 2018). However, the density is lower (0.50 vs. 0.66), sug-
gesting the presence of less interlocking in the core.
Cohesive groups of journals were also identified by the number of members which
interlock two journals. Following previous studies (e.g., Teixeira and Oliveira 2018), a
threshold of five or more editors (5-slice) was defined as parameter for network sectioning,
which resulted in four cohesive groups (Fig.5). The first subgroup is formed by two jour-
nals, Journal of Public Relations Research and Public Relations Review. The second sub-
group is formed by three journals, Journal of Advertising, Journal of Advertising Research,
and International Journal of Advertising. The third subgroup is formed by two journals,
Table 5 Statistics of network metrics by geographic area
* < 0.001
Geographic
area
NMean degree SD degree Mean
closeness
cent.
SD closeness
cent.
Mean
between.
cent
SD between.
cent
USA 1272 0.035 0.022 0.401 0.048 0.001 0.004
Asia–Pacific 184 0.032 0.016 0.383 0.049 0.001 0.004
Middle east-
Africa
23 0.031 0.014 0.384 0.049 * *
Europe 309 0.032 0.020 0.373 0.053 * *
Israel 22 0.036 0.019 0.404 0.042 * *
UK 173 0.031 0.016 0.375 0.040 0.001 0.001
Canada 47 0.031 0.015 0.384 0.042 0.001 0.003
Latin America 26 0.029 0.012 0.350 0.060 * *
Scientometrics
1 3
Fig. 3 Geographic distribution of the social network of members of the editorial boards of communica-
tion journals. White nodes represent journals. Colored nodes represent the affiliation of members: USA –
Turquoise, Asia–Pacific – Red, Middle East-Africa – Grey, Europe – Green, Israel – Black, UK – Yellow,
Canada – Blue, Latin America – Orange. (Color figure online)
Table 6 Geographic distribution of top-ranked scholars for each network metric
Geographic area Degree cent.
degree > = 3 (top
149)
% Degree cent.
degree > = 3
Closeness
cent. top
100
Betweenness
cent. top 100
Total % Total
USA 99 70.2 82 73 1240 61.6
Asia-Pacific 11 7.8 7 9 180 8.9
Middle east-Africa 1 0.7 0 0 22 1.1
Europe 15 10.6 7 8 300 14.9
Israel 3 2.1 1 0 33 1.6
UK 8 5.7 2 7 169 8.4
Canada 3 2.1 0 3 44 2.2
Latin America 1 0.7 1 0 24 1.2
Scientometrics
1 3
Science Communication and Public Understanding of Science. The final subgroup is the
main group, and it is formed by 23 journals that cover different subject areas of the field,
including journalism, communication, and political communication. Eleven journals did
not form part of any group. Subgroups inform about the main areas and topics in the field
of communication. Three independent themes were found: public relations, advertising,
and science communication. The size and density of the main subgroup suggest that most
Fig. 4 The 5-core of journals. Size of nodes is proportional to degree
Fig. 5 Cohesive groups of journals which share five or more members (5-slice) of their EBs. The thickness
of the edges is proportional to the weight of the connection
Scientometrics
1 3
journals have connections based on their EBs, and information may flow at a faster pace
within the subgroup. This group of journals well connected together suggests the existence
of an elite of journals which includes most of the Q1/Q2 ranked journals, although it also
has to be pointed out that several journals occupy more prominent positions than others.
Journal of Communication and Communication Research are in the central position (high-
est degree centrality) because they share a substantial amount of members of their EBs
with many other journals. However, the strongest connection is between the Journal of
Advertising and the International Journal of Advertising with 61 interlockings.
k-cores were used to find the centre of the network of scholars. Since journals have a
substantial amount of seats in their EB, the degree of the one-mode network of scholars is
particularly high (M = 96.32, SD = 61.15); and k has to be set to high values to cut off the
network significantly. The 100-core of scholars has 459 nodes and a density of 0.27. The
130-core has 132 scholars who sit in 16 different journals. The density of the 130-core is 1.
Therefore, the centre of the network of scholars is formed by a substantial group of scien-
tists completely connected through EB interlocking.
Cohesive groups of scholars were identified using the number of common journals in
which they serve as editors. A threshold of 4 common EBs (4-slice) was defined as param-
eter for network sectioning, which resulted in three cohesive groups (Fig.6). The threshold
was adjusted ad-hoc. The first cohesive group (left in the figure) includes 9 scholars, who
mostly focus on the topic of health communication. The second cohesive group includes 20
scholars who are connected by several common themes, including communication, polit-
ical communication, and journalism. The third cohesive group (right part in the figure)
includes 2 scholars who focus on journalism studies. The size and density of the larger
cohesive subgroup suggest that these scholars form an ‘invisible college,’ a connected elite
that may develop its own norms, values, and standards, monitoring the progress in the field
(Zuccala 2006).
Interrelationships amonginstitutions
The relationships between scholars, journals, and institutions can be represented as a mul-
tipartite graph in which journals are connected with scholars, and scholars are connected
to institutions. This multipartite graph can then be projected into a bipartite graph that
Fig. 6 Cohesive groups of scholars who are part of four or more EBs (4-slice)
Scientometrics
1 3
connects journals and institutions. A journal is connected to an institution if at least one
member of the EB is affiliated with the institution. The bipartite graph of journals and
institutions has 718 nodes (687 institutions and 41 journals) and 2013 edges. Degree cen-
trality (M = 0.075, SD = 0.093) shows that each institution is connected to 2.93 journals
on average. Average closeness centrality is high (M = 0.537, SD = 0.047), suggesting that
institutions are close together when connected through members of EBs. Average between-
ness centrality is low (M = 0.001, SD = 0.05), suggesting that the flow of information is
evenly distributed since only a small proportion of all communication paths go through
each institution. Network metrics of top institutions are presented in Table7. The number
of scholars of each institution in the dataset is also included. Institutions of the USA are
overrepresented in top positions. Top institutions from other geographical areas as ranked
by degree centrality are the University of Amsterdam (ranked 9), the Chinese University of
Hong Kong (ranked 22), Cardiff University (ranked 29), the University of Vienna (ranked
30), and the University of Haifa (ranked 33). The number of common journals, in which
Table 7 Network metrics of top institutions
Institution Degree Normal-
ized
degree
Closeness cent. Betweenness cent. # Scholars
Univ. of Texas at Austin 25 0.609 0.806 0.066 37
Univ. of Illinois at Urbana-Cham-
paign
24 0.585 0.753 0.025 29
Michigan State Univ 23 0.560 0.775 0.042 34
Univ. of Wisconsin-Madison 23 0.560 0.726 0.026 32
Univ. of Southern California 20 0.487 0.742 0.047 27
Pennsylvania State Univ 20 0.487 0.738 0.026 33
Fig. 7 Weighted projection of institutions that share members of EBs in more than 12 journals (13-slice).
The thickness of the edges is proportional to the journals with shared members. The size of each node is
proportional to the number of scholars of the institution in the dataset
Scientometrics
1 3
scholars serve as editors, was used to try to identify cohesive groups of institutions. A
threshold of more than 12 journals in which institutions share at least one member of the
EB was defined as parameter for network sectioning. Only one cohesive group was found
(Fig.7) with 17 institutions and 54 edges. The reduced number of institutions that are part
of this cohesive group and the density (0.40) suggests that there is also an elite of institu-
tions that shape the field.
The number of scholars that each institution has in the complete dataset is highly cor-
related with degree centrality (r = 0.942, p < 0.001), closeness centrality (r = 0.891,
p < 0.001) and betweenness centrality (r = 0.873, p < 0.001). Simple linear regression mod-
els show that the number of scholars also predicts all network metrics and that the lin-
ear models fit the data reasonably well: degree centrality (F = 5343, p < 0.001, R2 = 0.89),
closeness centrality (F = 2601, p < 0.001, R2 = 0.79) and betweenness centrality (F = 2177,
p < 0.001, R2 = 0.76). This suggests that institutions with more members in different EBs
have more connections to other institutions, are closer to them occupying central positions
in the social network, and are part of more communication paths among members of EBs.
This result may seem expected because the connections between journals and institutions
of the projected graph are made through the scholars that sit together in EBs. However, the
number of scholars is not necessarily an indicator of the connections between institutions.
For instance, a given institution can have many members that are only part of a small num-
ber of journals, thus limiting the possible connections that the institution can make with
other institutions through EBs. Also, EBs can have little diversity accumulating numerous
members affiliated only to a few institutions.
Conclusions
Editorial boards, as gatekeepers of knowledge, are key agents in the development of scien-
tific fields (Metz etal. 2016). Their crucial role in ensuring the quality of scientific jour-
nals (Willett 2013) and in the promotion and review process of manuscripts, make them
a fundamental body in shaping the form and content of scientific outlets (Rosenstreich
and Wooliscroft 2006). This exploratory paper contributes to the limited, but fundamental
research in this area by examining the EB interlocking in communication sciences. Using
graph theory and SNA, we analyse editorial board interlocking and identify the central
journals, institutions, and scholars describing their gender and geographic representation.
Our results offer relevant descriptive insights to understand why and how these influent
agents are structured, linked, and associated in what can be conceptualized as an “elite”
academic network.
In relation to the specific composition of the social network of EBs and journals, our
results show that the network is tightly connected with a centre that includes most jour-
nals and individuals. Average closeness centrality is particularly high, suggesting that the
distance between nodes is low. i.e., scholars could reach all other scholars through a small
number of connections. All this suggests that the field of communication is a compact dis-
cipline and that the scholarly practices related to the social network of EBs are collabora-
tively performed with a certain degree of internal cohesion. Furthermore, scholars have the
possibility to exert their influence in academic decisions effectively since they can reach
other actors promptly.
Results also indicate central scholars (individuals) in the network, and therefore the most
important ones in shaping the form and reach of the social network structure of EBs: M.
Scientometrics
1 3
Delli Carpini (rank#1 in degree centrality: member of more EBs than any other scholar),
Z. Papacharissi (rank#1 in closeness centrality: closest average distance to all other schol-
ars) and S. Okazaki (rank#1 in betweenness centrality: number of shortest communication
paths that go through him). This points to leaders in the discipline but also shows how
they can influence scholarly activities: through direct participation in the EBs of journals
(degree) (Willett 2013), by having closer network connections with other scholars (close-
ness), or by acting as gatekeepers of new and crucial information (betweenness). Cohe-
sive group analysis furthers points in this direction, suggesting the existence of an invis-
ible college of individuals that may shape the research agenda of the field. However, such
influence is more evenly distributed among journals, with an “elite” of publication ven-
ues, which includes most journals ranked in the Q1/Q2 of the JCR. Thematic groups found
include a core with journals and scholars focused on journalism, communication, and polit-
ical communication. Results also suggest the existence of additional thematic groups on
health communication, science communication, public relations, and advertising.
With regard to the gender representation of EBs, female members are underrepresented
in the dataset (40%). They are even more underrepresented in top positions when it comes
to the number of memberships (degree) and distance to other members of the network
(closeness centrality). However, the statistical differences for all network metrics when
analysing the complete dataset were not found significant, suggesting that the number and
quality of connections are similar for female and male members. Similarly, the geographic
representation of EBs indicates that institutions of the USA are overrepresented in the data-
set (61.82%). They also occupy most of the top positions in the network metrics related to
the number of connections (70.2%), distance to other members of the network (82%), and
the number of communication paths that go through them (73%). This suggests that the
epistemic culture and norms and values of scientific production from North-America are
paradigmatic in this field of research.
Despite our results providing empirical support for the existence of a geographical
imbalance in the governance of EBs, such findings should be framed according to the real
domination of US scholars and institutions in science production/dissemination in com-
munication research. Previous studies suggest that US communication scholars occupy a
leading position in communication research production in top-tier journals, and therefore,
it is reasonable that they also occupy a leading position as EB members.
In relation to the associations among academic institutions, results indicate that The
University of Texas at Austin is ranked in the first position of all metrics. It has more mem-
bers in EBs than any other institution, their scholars can reach all other actors rapidly, and
they are also part of more communication paths. Assuming that these metrics are relevant
to determine the promotion and review of research, the University of Texas at Austin is the
central institution in shaping the form and content of communication journals. The number
of scholars of each institution predicts the number of connections with other institutions
and also its capacity to access and spread information, thus reflecting its centrality in the
network. The number of scholars that an institution has in the different EBs of Q1/Q2 com-
munication journals is, therefore, a good indicator of the relevance of that institution in the
social network as a gatekeeper of knowledge. From a methodological perspective, these
findings imply that the overall number of scholars in EBs can be used to measure the cen-
trality of that institution, facilitating future estimations. From a practical perspective, these
findings suggest that institutions seeking to improve their relevance in the research field of
communication should try to get more members into the EBs of Q1/Q2 journals.
Finally, at a methodological level, this study uses SNA to investigate the connections
between scholars, journals, and institutions in the field of communication research by
Scientometrics
1 3
conducting an analysis of members of EBs of JCR communication journals. To the best of
our knowledge, this is the first study that explains the structure of EBs using SNA in com-
munication research. This approach can complement others such as scientometrics, which
stresses interlocking between journals, providing valuable insights about how scholars of
EBs interrelate. The correspondence of our findings with previous studies on gender and
geographic diversity in research argues for the methodological validity as a tool for study-
ing the relationships between scholars, journals, and institutions through the composition
of EBs.
Limitations andfuture research
Six main limitations, which need to be addressed in future research, are noteworthy. First,
the ranking and quartiles selected to form the sample are biased towards English-speaking
countries. JCR journals generally employ English as lingua franca, which can explain the
geographical domination of the United States, United Kingdom, or Canada. Therefore,
future studies may also consider the full sample of JCR journals or take other more inclu-
sive rankings (such as Scopus SJR). Second, to gather data related to editorial members’
nationality, we take the country of their academic affiliation as reference. However, in the
communication field, there are many scholars with a very international background, mean-
ing that their country of origin might not be aligned with their country of residence. Third,
many EBs’ affiliations were incorrectly reported or outdated in journals’ websites. When
we detected these mistakes, we directly corrected them in our dataset. However, during
data gathering, we did not double-check whether or not the affiliations publically avail-
able were correctly or incorrectly reported by journals. Fourth, other geographical catego-
rizations of EBs nationality might be considered. Fifth, our results map some imbalances
in the governance of EBs, showing the potential invisible colleges that might exist in the
field. However, our findings do not empirically demonstrate if those inequalities cause real
effects (or what effects were really produced). In short, results demonstrated an unbalance
between institutions/geography, but the extent this unbalance materializes in real prejudice
cannot be shown. Future studies may examine the existence of biases in research published
when EBs and authors of papers share similar characteristics like gender, nationality, affili-
ation, or co-authoring. A more challenging endeavour would be investigating how papers
published by the invisible college impact the communication field. Since the invisible col-
lege may result in a convergence of research streams, future research could investigate
whether and how topics, theories and methods are clustering or dispersing in published
research. Further work could also examine the similarities and differences in the edito-
rial policies of communication journals to further determine the influence of the invisible
college.
Sixth, this study analyses the structure of the social network of connections of EBs in
the field of communication research without considering its possible connections with
other fields. From this perspective, results show that the field presents a compact and cohe-
sive discipline in its internal structure of the connections, which reflect scholarly practices
of EBs. However, this could also mean that there is an excessive level of homogeneity
in those practices that result from the lack of interdisciplinary work that can bring new
theories and interpretations. As future work, we suggest studying other fields of research
to analyse and compare their form, and also to analyse the relations between communica-
tion research and other fields to determine interdisciplinary patterns in the composition of
Scientometrics
1 3
EBs. Further, this study analyses the connections between journals and scholars but not
the content of these connections. Future research could investigate how communication
journals and members are connected by papers published using co-citation networks and
co-author networks to determine their structure and to what extent the imbalances or elites
are also present there. Please note that our study shows the connections between journals
and scholars indirectly through EBs. Despite these limitations, this article presents original
and robust findings to calibrate the power “behind the scenes” in the communication field,
addressing the main actors and institutions that shape the form and governance of elite
journals.
Appendix1: Journals included inthestudy
New Media & Society, Journal of Computer-Mediated Communication, Journal of Com-
munication, Media Psychology, Communication Research, Journal of Advertising, Com-
munication Theory, Information, Communication & Society, Public Understanding of Sci-
ence, Political Communication, International Journal of Advertising, Comunicar, Journal
of Advertising Research, Journalism Studies, Research on Language and Social Interac-
tion, Science Communication, Communication Monographs, Journal of Public Relations
Research, Journal of Health Communication, Human Communication Research, Telecom-
munications Policy, International Journal of Press-Politics, International Journal of Com-
munication, Health Communication, Journalism, Journal of Social and Personal Rela-
tionships, Environmental Communication, European Journal of Communication, Public
Opinion Quarterly, Television & New Media, Journal of Broadcasting & Electronic Media,
Public Relations Review, Mass Communication and Society, Journalism & Mass Commu-
nication Quarterly, Journal of Language and Social Psychology, International Journal of
Public Opinion Research, Written Communication, Media Culture & Society, Games and
Culture, IEEE Transactions on Professional Communication, Technical Communication.
References
Alvesson, M., Gabriel, Y., & Paulsen, R. (2017). Return to meaning: A social science with something to say.
Oxford: Oxford University Press.
Andrikopoulos, A., & Economou, L. (2015). Editorial board interlocks in financial economics. International
Review of Financial Analysis, 37, 51–62. https ://doi.org/10.1016/j.irfa.2014.11.015.
Baccini, A., & Barabesi, L. (2010). Interlocking editorship. A network analysis of the links between eco-
nomic journals. Scientometrics, 82(2), 365–389. https ://doi.org/10.1007/s1119 2-009-0053-7
Baccini, A., & Barabesi, L. (2011). Seats at the table: The network of the editorial boards in information
and library science. Journal of Informetrics, 5(3), 382–391. https ://doi.org/10.1016/j.joi.2011.01.012.
Barzilai-Nahon, K. (2009). Gatekeeping: A critical review. Annual Review of Information Science and Tech-
nology, 43, 1–79. https ://doi.org/10.1002/aris.2009.14404 30117 .
Bastian, M., Heymann, S., & Jacomy, M. (2009). Gephi: An open source software for exploring and manip-
ulating networks. Paper presented at the Third International AAAI Conference on Web and Social
Media, San José, CA, USA.
Bedeian, A. G., Van Fleet, D. D., & Hyman, H. H., III. (2009). Scientific achievement and editorial board
membership. Organizational Research Methods, 12, 211–238. https ://doi.org/10.1177/10944 28107
30931 2.
Borgatti, S., & Halgin, D. (2014). Analyzing affiliation networks. In J. Scott & P. J. Carrington (Eds.), The
SAGE handbook of social network analysis (pp. 417–433). London: SAGE Publications Ltd.
Scientometrics
1 3
Burgess, T. F., & Shaw, N. E. (2010). Editorial board membership of management and business journals: A
social network analysis study of the Financial Times 40. British Journal of Management, 21, 627–648.
https ://doi.org/10.1111/j.1467-8551.2010.00701 .x.
Cascio, W. F. (2008). How editors are selected. In Y. Baruch, A. Konrad, H. Aguinis, & W. Starbuck (Eds.),
Opening the black box of editorship (pp. 231–238). London: Palgrave Macmillan.
Dahdouh-Guebas, F., Ahimbisibwe, J., Van Moll, R., & Koedam, N. (2003). Neo-colonial science by the
most industrialised upon the least developed countries in peer-reviewed publishing. Scientometrics, 56,
329–343.
Demeter, M. (2018). Changing center and stagnant periphery in communication and media studies: National
diversity of major international journals in the field of communication from 2013 to 2017. Interna-
tional Journal of Communication, 12, 29.
Dhanani, A., & Jones, M. J. (2017). Editorial boards of accounting journals: gender diversity and interna-
tionalisation. Accounting, Auditing & Accountability Journal, 30, 1008–1040. https ://doi.org/10.1108/
AAAJ-08-2014-1785.
Feldman, D. C. (2008). Building and maintaining a strong editorial board and cadre of ad hoc reviewers.
In Y. Baruch, A. Konrad, H. Aguinis, & W. Starbuck (Eds.), Opening the black box of editorship (pp.
68–74). London: Palgrave Macmillan.
Goyanes, M. (2017). Desafío a la investigación estándar en comunicación: Crítica y Alternativas. Barce-
lona: Editorial UOC.
Goyanes, M. (2019). Editorial boards in communication sciences journals: Plurality or standardization?
International Communication Gazette. https ://doi.org/10.1177/17480 48518 82532 2.
Kurmis, A. P. (2003). Understanding the limitations of the journal impact factor. JBJS, 85, 2449–2454. https
://doi.org/10.2106/00004 623-20031 2000-00028 .
Lee, T. (1997). The editorial gatekeepers of the accounting academy. Accounting, Auditing & Accountability
Journal, 10, 11–30. https ://doi.org/10.1108/09513 57971 01586 94.
Liwei, Z., & Chunlin, J. (2015). Social network analysis and academic performance of the editorial board
members for Journals of Library and Information Science. Collnet Journal of Scientometrics and
Information Management, 9(2), 131–143. https ://doi.org/10.1080/09737 766.2015.10699 47.
Mauleón, E., Hillán, L., Moreno, L., Gómez, I., & Bordons, M. (2013). Assessing gender balance among
journal authors and editorial board members. Scientometrics, 95, 87–114. https ://doi.org/10.1007/
s1119 2-012-0824-4.
Metz, I., Harzing, A. W., & Zyphur, M. J. (2016). Of journal editors and editorial boards: Who are the trail-
blazers in increasing editorial board gender equality? British Journal of Management, 27, 712–726.
https ://doi.org/10.1111/1467-8551.12133 .
Ozbilgin, M. (2004). “International” human resource management: Academic parochialism in editorial
boards of the “top” 22 journals on international human resource management. Personnel Review, 33,
205–221. https ://doi.org/10.1108/00434 80410 51805 59.
Petersen, J., Hattke, F., & Vogel, R. (2017). Editorial governance and journal impact: A study of man-
agement and business journals. Scientometrics, 112(3), 1593–1614. https ://doi.org/10.1007/s1119
2-017-2434-7.
Rosenstreich, D., & Wooliscroft, B. (2006). How international are the top academic journals? The case of
marketing. European Business Review, 18(6), 422–436.
Teixeira, E. K., & Oliveira, M. (2018). Editorial board interlocking in knowledge management and intel-
lectual capital research field. Scientometrics, 117(3), 1853–1869. https ://doi.org/10.1007/s1119
2-018-2937-x.
Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applications. Cambridge: Cam-
bridge University Press.
Willett, P. (2013). The characteristics of journal editorial boards in library and information science. Inter-
national Journal of Knowledge Content Development & Technology, 3, 5–17. https ://doi.org/10.5865/
IJKCT .2013.3.1.005.
Zdeněk, R. (2018). Editorial Board Self-Publishing Rates in Czech Economic Journals. Science and engi-
neering ethics, 24, 669–682. https ://doi.org/10.1007/s1194 8-017-9922-2.
Zuccala, A. (2006). Modeling the invisible college. Journal of the American Society for Information Science
and Technology, 57(2), 152–168. https ://doi.org/10.1002/asi.20256 .
... Two journals are connected when they share editorial board members. This phenomena, referred to as "interlocking editorship", has been used to discover the functioning and organization of academic communities in the communication sciences (Goyanes and Marcos, 2020), economics (Baccini and Barabesi, 2010), finance (Andrikopoulos and Economou, 2015), knowledge management (Teixeira and Oliveira, 2018), bioinformatics (Malin and Carley, 2007), management (Burgess and Shaw, 2010), and the information and documentation sciences (Baccini and Barabesi, 2011). The results of these studies reveal distinct disciplinary structures: a high level of cohesion in the communication sciences (Goyanes and Marcos, 2020), a high level of dispersion in finance (Andrikopoulos and Economou, 2015), regional subgroups in economics (Baccini and Barabesi, 2010), and a centrality of US universities in management (Burgess and Shaw, 2010). ...
... This phenomena, referred to as "interlocking editorship", has been used to discover the functioning and organization of academic communities in the communication sciences (Goyanes and Marcos, 2020), economics (Baccini and Barabesi, 2010), finance (Andrikopoulos and Economou, 2015), knowledge management (Teixeira and Oliveira, 2018), bioinformatics (Malin and Carley, 2007), management (Burgess and Shaw, 2010), and the information and documentation sciences (Baccini and Barabesi, 2011). The results of these studies reveal distinct disciplinary structures: a high level of cohesion in the communication sciences (Goyanes and Marcos, 2020), a high level of dispersion in finance (Andrikopoulos and Economou, 2015), regional subgroups in economics (Baccini and Barabesi, 2010), and a centrality of US universities in management (Burgess and Shaw, 2010). ...
... To date, studies on interlocking editorship have considered one discipline and, by examining its most influential journals, have analyzed the degree of cohesion of the journal networks, identified communities or subgroups, and signaled central journals and editors (Andrikopoulos and Economou, 2015;Barabesi, 2010, 2011;Goyanes and Marcos, 2020;Malin and Carley, 2007;Teixeira and Oliveira, 2018). Although the cohesion levels of the journal network vary depending on the discipline examined, all of the cited works found a core and a periphery in the journal network. ...
Article
Full-text available
One means of examining the structure of a scientific discipline consists of analyzing the editorial boards of its most influential journals. This work analyzes the networks among the top 100 sociology journals, formed by multiple memberships of editors to distinct boards (interlocking editorship). The results suggest that a small number of editors having similar profiles simultaneously serve on a variety of journal boards; top journals form a large and united community, with Anglo-Saxon universities dominating the same; universities from countries located in the Southern Hemisphere hold a marginal position; and general sociology journals are more connected than specialized sociology journals. These findings contribute to debates on disciplinary segmentation, degrees of internationalization, openness to non-academic sectors and trajectories of the scientific “elite”.
... We will interpret the historic development of the field by building on the conceptual framework of "invisible colleges" (cf., de Solla Price, 1965;Goyanes & De-Marcos, 2020;Vogel, 2012), which is applied to make sense of the academic communication between researchers to elucidate the past of the LMX field. Exploring this allows us to present the changing perspective of emerging and shifting colleges of literature that LMX studies have cited in a specific time period. ...
... Given the importance of formal publications (the "how") for the dissemination of knowledge, allocation of resources and professional recognition seems to be important for tracking and exploring scholarly communication. Bibliometric methods are a useful approach for exploring such issues (Goyanes & De-Marcos, 2020;Zupic & Čater, 2015). ...
Article
Full-text available
This study examines the evolution of leader–member exchange (LMX) research. We apply bibliometrics to analyze the extant literature published from the beginning of the construct development, complementing existing qualitative and meta-analytic reviews of the LMX field. We use a combination of three bibliometric techniques—document co-citation analysis, co-word analysis, and bibliographic coupling. Our approach enables us to identify the most influential topics, determine the underlying structure and development of the field by interpreting bibliometric results against the backdrop of the invisible colleges framework, as well as detect emerging topics. We conclude by providing important theoretical and methodological implications, suggesting the emergence of new sub-fields and future opportunities for new connections among specific existing sub-streams of LMX, leadership, and management research.
... Likewise, exploring the gender and geographic representation of the most productive scholars constitutes a vital research agenda to further capture research bias in scientific production. Altogether, this study contributes to current conversations at the intersection of Scientometrics studies, gender, and geographical bias in global knowledge production (e.g., Aguinis et al., 2018;Baruch, 2001;Begeny et al., 2018;Brown et al., 2020;Goyanes & De-Marcos, 2020;Leahey, 2006;Nisonger, 2002;Teele & Thelen, 2017;Van Arensbergen et al., 2012), providing insightful theoretical implications and empirical findings on academic representation and research trajectories. ...
Article
Full-text available
Examining research patterns across scientific fields constitutes a growing research enterprise to understand how global knowledge production unfolds. However, scattered empirical evidence has casted light on how the publication diversity of the most productive scholars differ across disciplines, considering their gender and geographical representation. This study focuses on the most prolific scholars across three fields (Communication, Political Science, and Psychology), and examine all journals where they have published. Results revealed the most common journals in which prolific scholars have appeared and showed that Communication scholars are more prone to publish in Political Science and Psychology journals than vice-versa, while psychologists' largely neglect them both. Our findings also demonstrate that males and US scholars are over-represented across fields, and that neither the field, gender, geographic location, or the interaction between gender and geographic location has a significant influence over publication diversity. The study suggests that prolific scholars are not only productive, but also highly diverse in the selection of the journals they publish, which directly speaks to both the heterogeneity of their research contributions and target readers.
... In the field of financial economics, Andrikopoulos and Economou [17] detected a core−periphery structure of editors of 20 journals by employing a network analysis. Goyanes and de-Marcos [18] conducted a social network analysis to reveal the interlocking structure of 41 communication journals' editorial boards. By mapping interlocking editorial board social networks in knowledge management and intellectual capital fields, Teixeira and Oliveira [19] explored which journals were the most influential. ...
Article
Full-text available
The co-editor networks of academic journals are generally examined at the journal level. This paper investigates the geographies of the global co-editor network in oncology through the lens of cities. After using different network methods to analyze the global co-editor network, we found that the network can be characterized by a core-periphery structure. The dense core is occupied by many highly interconnected cities, whereas the periphery contains many cities maintaining loose connections with the core cities. The core shows an asymmetric dual sub-core structure. The greater sub-core is constituted by Northern American cities with New York, Washington DC, Boston, Houston, and Los Angeles in the center, whereas the smaller sub-core is formed by Asian cities and centered on Tokyo, Seoul, Osaka, Beijing, and Shanghai. The European core cities do not form a well-outlined sub-core but produce a ringlike shape around the Northern American core. This structure of the co-editor network is a consequence of the prestige effect still characterizing global science. Many European and Chinese journals tend to employ Northern American editors (US-based editors in the first place) to help increase the reputation of the journal. However, US-based journals are more interested in recruiting American editors from the top-ranked national cancer centers and universities rather than outside of the country.
... The relations among editors consists in sitting together in the board of a journal. When tools for individuating clusters are applied to the network of editors, the resulting groups are interpreted as a kind of invisible college [27,28]. This structure can be explored also in an interlocking editorship perspective: in this kind of network nodes are journals and the similarity between a pair of journals is computed by using the number of their common editors. ...
Article
Full-text available
This paper explores intellectual and social proximity among scholarly journals by using network fusion techniques. Similarities among journals are initially represented by means of a three-layer network based on co-citations, common authors and common editors. The information contained in the three layers is then combined by building a fused similarity network. The fusion consists in an unsupervised process that exploits the structural properties of the layers. Subsequently, partial distance correlations are adopted for measuring the contribution of each layer to the structure of the fused network. Finally, the community morphology of the fused network is explored by using modularity. In the three fields considered (i.e. economics, information and library sciences and statistics) the major contribution to the structure of the fused network arises from editors. This result suggests that the role of editors as gatekeepers of journals is the most relevant in defining the boundaries of scholarly communities. In information and library sciences and statistics, the clusters of journals reflect sub-field specializations. In economics, clusters of journals appear to be better interpreted in terms of alternative methodological approaches.
... A social network is a web of relationships connecting different actors together (e.g., individuals, organisations, nations). The purpose of analyzing networks in scientific research is to evaluate the performance of certain research actors through the structure and patterns of their relationships, as well as to guide research funding and development of science [76]. Following previous works [52,77], SNA can be conducted through a variety of metrics such as ego-density at the network level; degree, betweenness and closeness centrality, efficiency and constraint at the actor level; and tie strength at the tie level [78,79]. ...
Article
Full-text available
This paper explores the evolution of research collaboration networks in the ‘stakeholder theory and management’ (STM) discipline and identifies the longitudinal effect of co-authorship networks on research performance, i.e., research productivity and citation counts. Research articles totaling 6,127 records from 1989 to 2020 were harvested from the Web of Science Database and transformed into bibliometric data using Bibexcel, followed by applying social network analysis to compare and analyze scientific collaboration networks at the author, institution and country levels. This work maps the structure of these networks across three consecutive sub-periods (t1: 1989–1999; t2: 2000–2010; t3: 2011–2020) and explores the association between authors’ social network properties and their research performance. The results show that authors collaboration network was fragmented all through the periods, however, with an increase in the number and size of cliques. Similar results were observed in the institutional collaboration network but with less fragmentation between institutions reflected by the increase in network density as time passed. The international collaboration had evolved from an uncondensed, fragmented and highly centralized network, to a highly dense and less fragmented network in t3. Moreover, a positive association was reported between authors’ research performance and centrality and structural hole measures in t3 as opposed to ego-density, constraint and tie strength in t1. The findings can be used by policy makers to improve collaboration and develop research programs that can enhance several scientific fields. Central authors identified in the networks are better positioned to receive government funding, maximize research outputs and improve research community reputation. Viewed from a network’s perspective, scientists can understand how collaborative relationships influence research performance and consider where to invest their decision and choices.
Article
This study employs content analysis and social network analysis to investigate the degree of de-Westernization in the journalism research of the BRICS (acronym of Brazil, Russia, India, China, and South Africa) nations. A total of 10,929 citations between 2010 and 2019 from five top academic journals of journalism research were analyzed. It is found that Anglosphere scholarship dominates the journalism knowledge production across the five countries. The BRICS journalism researchers primarily relied on domestic and Western scholarships to produce knowledge while they were largely isolated from each other and other researchers in the Global South. Three types of citation practice have been discussed, including closed, dependent, and diversified citations. The concept of linguistic segregation is discussed to explain the citation patterns of the BRICS journalism researchers, especially the central role of Anglosphere scholarship. Furthermore, inter-regional referencing and international cooperation are suggested as potential directions to promote the process of de-Westernization and to deepen globalization in the field of journalism research. Finally, limitations and suggestions are also presented.
Article
This article analyses the connections between members of the editorial board of a scientific journal which are formed based on their reviews of scientific articles. It is shown that the connections can be represented as a graph. The research uses the data for six years of article reviewing in the scientific and technical Journal of "Almaz — Antey" Air and Space Defence Corporation. The methods of analysis are combinatorics and graph theory, as well as the relevant graph characteristics: adjacency matrix, incidence matrix, reachability matrix, graph fullness and connectivity, nearest neighbours graph and graph spanning tree. It is shown that cooperation of the reviewers of the Journal helps plot a connected graph with links between any two vertices, i.e. between any reviewers. The graph is analysed and the methods of its application to calculate the Journal’s scientometric indicators are demonstrated. As the research reveals, a journal that publishes articles in numerous disciplines forms connections between all reviewers and this parameter can be indicative of interchangeability within the scientific fields or, conversely, of a joint work in this scientific field when reviewing manuscripts. Based on the research results, it is possible to search for new reviewers in the areas where competencies are underdeveloped. And by the areas where competencies are strong, we can determine the core competence of the reviewers of a scientific journal, reflecting the main focus of the evaluated scientific research. The work resulted in the formulated scientometric indicators of the journal, which can be used to search for and involve new reviewers or to represent data on a strong team of reviewers on a specific topic, as well as on a new scientific field just emerging for research.
Article
It seems have become a common practice for the editorial board members to publish the publications in their “own” journals. In view of the important position of the board members, this article devotes to exploiting the difference of the publications by board and non-board members from the perspectives of publication delay and influence. The Journal of Informetrics and Journal of the Association for Information Science and Technology are taken as the resulting representative of emerging and verteran journals. After extensive analysis, we find that (1) the scientific publications by board members have shorter publication delays than those by non-board members; (2) the first or corresponding authorship dominates among the publications by board members; and (3) compared to those by non-board members, the academic articles by board members have more influence on average in term of citation counts. This may be a Matthew effect for editorial board membership.
Article
Full-text available
This research-based essay examines the national diversity of editorial boards from a selection of journals in communication sciences. Specifically, it reviews the board composition of 39 Journal Citation Report journals indexed in quartile one (Q1) and quartile two (Q2) in the category of 'communication', proposing a typology of dominant nationalities. The most distinguished countries are the United States, United Kingdom, Canada, Australia and Germany, monopolizing 79.4% of total members. The exaggerated domination of certain geographies is surprising given the increasing acknowledgement of plurality as a constitutive value of scientific progress. The article then problematizes why plurality is limited and, therefore, identifies a body of social and cultural bonds that underpin the domination of certain epistemic cultures. The study finally proposes an agenda that moves beyond the current status quo, and considers how these actions are likely to promote a more pluralistic and diverse intellectual terrain.
Article
Full-text available
In this research, we explored what has changed in the field of communication and media studies since Edmund Lauf’s research in 2005, in which he analyzed publication patterns of leading communication journals from 1998 to 2002. We compared the results of our current analysis of 14,925 articles published in 72 Web of Science-ranked communication journals from 2013 to 2017 with Lauf’s earlier data. We found that most leading journals still publish articles almost exclusively from the developed world, and we found the same bias regarding the composition of journal editorial boards. Analysis shows a decreasing contribution of the U.S., while Asia and Western Europe greatly increased their participation, and developing regions are still underrepresented. Our research shows that the field is still deliberately dominated by Western articles in Western journals edited by Western editorial board members. Given this, we suggest that the international community of communication scholars develop strategies to expand common standards for a more balanced international contribution pattern.
Article
Full-text available
In striving for academic relevance and recognition, editors exert a significant influence on a journal’s mission and content. We examine how characteristics of editors, in particular the diversity of editorial teams, are related to journal impact. Our sample comprises 2244 editors who were affiliated with 645 volumes of 138 business and management journals. Using multi-level modeling, we relate editorial team characteristics to journal impact as reflected in three widely used measures: Five-year Impact Factor, SCImago Journal Rank, and Google Scholar h5 index. Results show that multiple editorships and editors’ affiliation to institutions of high reputation are positively related to journal impact, while the length of editors’ terms is negatively associated with impact scores. Surprisingly, we find that diversity of editorial teams in terms of gender and nationality is largely unrelated to journal impact. Our study extends the scarce knowledge on editorial teams and their relevance to journal impact by integrating different strands of literature and studying several demographic factors simultaneously. Results indicate that the editorial team’s scientific achievement is more decisive than team diversity in terms of journal impact. The study has useful implications for the composition of editorial teams.
Article
Full-text available
This article investigates whether editorial board members of selected economic journals publish their research papers in their 'own' journal. Journals were selected from the Journal Citation Report(®) from the categories Business; Business, Finance; and Economics. Only research articles published between 2012 and 2015 were included in the analysis. We recorded ratios concerning the share of articles authored by editorial board members, the share of editorial board members publishing in their own journals and ratios representing their publication output. The average share of articles authored by editorial board members ranges from 0.6 to 17.5%. The average share of editorial board members publishing in their own journals ranges from 5.6 to 24.4%. Considering only editorial board members publishing in their own journals, the share of their articles in their journals ranges from 8.2 to 71.4%. While the share of board members publishing only in their own journals, to the number of board members publishing in their own journals, the ratio in a quarter of journals is equal to zero, with a maximum reach of 85.7%. All observed ratios are significantly positively correlated with the gap between impact factor and impact factor without Journal Self Cites; and negatively correlated with the Article Influence Score. A cluster of journals in which a high proportion of editorial board members publish and simultaneously these members publish in their own journal at a high rate was identified.
Article
Due to the importance of academic journals, it is critical to understand how they can create influence on research fields. An alternative way to understand how scholarly journals create influence in academic fields is to analyze the editorial board interlockings (EBI) and how they may generate networks within academic fields. EBI is the term used to describe the general circumstance where the same scholar is a member of multiple editorial boards, which creates a social network dimension within and between academic journals. The aim of this paper is to examine EBI phenomenon within knowledge management and intellectual capital fields (KM–IC). Assuming that EBI creates a social structure within scholarly journals, this paper investigates how KM–IC journals are connected through EBI, which journals are the most influential within KM–IC field and identify if KM–IC scholarly journal network breaks into subgroups. Social network analysis was the method applied using data from KM–IC ranking. Results identified the scholarly journal network, the high influence journals, and the cohesive subgroups within KM–IC field.
Purpose Editorial boards of academic journals represent a key institutional mechanism in the governance and functioning of the academic community. Board members play an important role in knowledge production and development of the discipline. The purpose of this paper is to enquire into the diversity characteristics of boards of accounting journals. Design/methodology/approach Drawing on a diversity framework that distinguishes between societal diversity and value of diversity, the paper examines two board characteristics: gender diversity and internationalisation. Moreover, it examines the influence of three journal and two editor characteristics on board diversity and analyses trends over time. Findings On gender, overall board trends are consistent with societal diversity and value of diversity: boards reflect the gender profile of senior academics. Further, female representation on boards is broadly consistent across the different journal nationalities; has improved over time; has experienced a convergence in “gender sensitive” sub-disciplines; and is influenced by female editorship. However, inequities appear to be present at the highest level: women appear to be less well represented than men as editors and women also have a lower representation on boards of higher ranked journals than on those of lower ranked journals. On internationalisation, once again, overall trends broadly reflect societal diversity and value at diversity. However, international scholars are less well represented on 4* boards than on 2* and 3* boards and on US boards than on Australian and UK boards. Further, there are signs of weakening US dominance in non-US journals. Originality/value Drawing on the diversity framework, this is the first study to comprehensively examine gender diversity and internationalisation of accounting boards.
Chapter
Some of the most important decisions you will make as an editor—and some of the decisions that will have the greatest impact on the success of your editorship—have to be made before the first manuscript ever crosses your desk or pops up on your computer screen. Whom you choose as your associate editors, whom you choose as your editorial board members, and whom you invite to be ad hoc reviewers determine the quality of the reviewing “service” your faculty colleagues will experience as “customers.” Moreover, these choices send important signals to the marketplace about your values, priorities, and preferences as an editor.
Chapter
Selection of a new editor for a journal is (or at least it should be) a significant undertaking in time and effort, for it will have long-lasting effects. Editors are gatekeepers of knowledge. As such they affect the development of a field, as reflected in the articles their journals publish, and they also affect the development of individual professional careers. The tone of the letters the editor writes to authors may instill feelings of hope, anger, self-confidence, despair, or a variety of other reactions. In all cases, the editor is the face that represents the journal to the outside world. Hence, the decision to select a new editor should not be taken lightly.