Male Presence in Gender Research Networks in the Communication field in Spain

The purpose of this paper is to learn about the scientific productivity and collaboration of authors who have published papers on gender in Spanish communication journals by paying special attention to the role played by men. The citation network was analysed in order to identify the researchers network and the participation of men therein. The hypothesis put forward is that men have a secondary role in scientific communities that research gender in the communication field. Methodologies from the analysis of social networks were applied. The information extracted was analysed using Ucinet software. It was confirmed that men have a marginal role in the network and that women tend to cite other women in their publications.

he words of Sánchez de Madariaga are revealing when she states that "the most recent research has shown how, in many scientific knowledge fields, from humanities to medicine, via social sciences and engineering, sex and gender variables are still not sufficiently taken into account" (2011, p.1). The same paper containing the aforementioned researcher's prologue stresses that women are still not highly represented.
Women's access to the fields of research and technology encompasses a true obstacle course. For centuries, scientific academies and universities vetoed women in multiple ways. Arguments included a "feminine nature" perceived as incompatible with scientific knowledge. The consideration of Donna Haraway (1991) when she reflects that science is concerned with knowledge and power is important because it is in the link between science and power (especially if it is understood as domination) that the quid of the question could lie. Scientific literature has shown the under-representation of women in practically all academic fields and it is confirmed that there are fewer women, and those in the field hold peripheral positions in research networks (Ceci et al., 2011;Eccles, 2007;Ecklund et al., 2012;West et al., 2013).
While participating in scientific tradition's demonstration of this reality, we want to ask whether something similar occurs with regard to men in collaboration networks focusing on studies from a gender perspective that have been conducted in recent years. In other words, we want to answer the following questions: Have men been fully included in research whose central topic is women's studies? Conversely, are those researching these topics primarily women? Does a gender bias occur in women's studies?
There is no research that has analysed sex in gender research or that studies the role played by men in this academic community. The main purpose put forward in this paper is to learn about the scientific productivity and collaboration of Spanish authors who have published papers from a gender perspective in communication journals indexed in SCOPUS during the period 2008-2014. The citation network was built to identify the structure of researcher networks and the role played by men therein. This information can serve as a basis to understand the scant presence of men in this kind of research and contribute to the design of policies to expand this topic among male researchers.

Justification and Theoretical-Methodological Framework
Traditionally, women's studies in communication were started and have been developed by women researchers, as in other fields of knowledge where research has also always been linked to the commitment of action in social sciences. Male participation began to rise in the last 10 years, and a study specifying the reason for this participation is important and may explain whether, in addition to scientific interest, the tendency towards researching sensitive topics, including gender, and the ease with which they are published contributes to this. In gender research, this is not an isolated fact, therefore, it is to be expected that the same thing happens in other branches deriving from or complementary to women's studies: in masculinity studies, the majority of the researchers are men (Pini & Pease, 2013) and in queer studies, the researchers are women and men, but militants of this perspective or vital position (Allen, 2010). But it may be that the scarcity of men is not only due to a lack of interest, but also because research networks led by women are hard to access.
Bibliometrics was used to analyse this phenomenon. Bibliometrics applied to social sciences with regard to author productivity helps to prove the productive life of an author, as well as his or her consideration in the scientific community, in the words of Rubio (1999). Likewise, the role of co-authorship, which has risen in the last decade and is linked to state or private support figures and experimental works, according to the same author, is essential in order to mark so-called "invisible colleges", understood as professional groups united along research lines, personal and teaching relationships, etc. These methods have produced important papers in other areas of knowledge (Chi & Young, 2013;Clark et al., 2014;Kraus et al., 2012), which validates them in order to put them into practice in communication and gender studies.
There are multiple papers that have used bibliometrics in the field of communication, but none of them have influenced their practice on gender studies in a comparative way. Work has been done on quantitative or qualitative methods of journal articles in communication (Castillo, Almansa & Álvarez, 2013) on the provenance of authors and their points of reference in Spanish journals (Fernández, Masip & Bergillos, 2013), on co-citations (Piedra, 2010) or on the "Academy" itself through the components of the doctoral thesis committees in communication (Casanueva & Caro, 2013). Gender journals remain at the fringes of this kind of research on researchers, therefore, there is no work quantifying the gender of research in women's studies. There are publications on women researchers in general (Alcalá et al., 2005) in the country or in autonomous communities, and some specific ones on women researchers in social sciences (Maz et al., 2011), but no meta-research on sex in gender research in communication.
The first stage we intended to reach was the small-scale discovery of invisible colleges where the domain of women researchers does or does not leave room for men researchers. De Solla Price (1963) points to the existence of groups of scientists that are essentially made up of a contact and informal communication that creates a stable social structure (highlighting the role of the elite within them). The concept of "invisible college" with a long track record in psychology studies, is reduced to scientific collaboration that for different reasons becomes reiterated.
The concept of "invisible colleges" is not new in literature. It is linked to the origins of modern science with a clear objective: to serve as a communication vehicle between researchers and research. In fact, scientific journals managed to harmonise this link between those who read them and alerted those who worked on a specific topic (Gracia, 2005). Therefore, such concept is linked to the actual development of science in that it recognises that the person conducting the research not only needs to communicate his or her results, but also needs to know what other people think and are studying. If we consider science as an organisation (Carpintero & Peiró, 1981), it is possible to analyse academic production in both its qualitative and quantitative dimensions by taking the material and the social into account, and invisible colleges can be understood as social networks where the analysis unit is the social relationship (Miceli, Bruno & Pulh, 2011). Thus, for the aforementioned authors, a measurable aspect of an academic relationship in terms of content would then be the bibliographical reference, and the frequency and/or volume of the exchanged material would be a measurement factor of the intensity of the social relationship.
Scientific journals come about with the precise purpose of becoming a communication network among those working on the same topic, and "constitute a kind of distance work group" (Gracia, 2005, p.13). The same author makes explicit the differences between the concept of "visible colleges" and the concept of "invisible colleges." The former are identified through joint signatures on scientific papers and citation networks, while invisible colleges are social networks and groups that are identified through paths other than those of citation networks, given that they come before them. Electronic communication encourages the consolidation of these networks.
As maintained by Peñaranda-Ortega and Osca-Lluch (2013), they are scientific connection networks that are established informally, but come to be formalised in research articles published jointly. Thus, such networks facilitate the exchange of information and, on many occasions, culminate with an interdisciplinary collaborative exchange that aids the growth of scientific literature, both in a quantitative and qualitative manner. "Currently, the term "invisible college" is attempting to show the close relationship between productivity and collaboration" (Peñaranda & Osca, 2013, p.129) and entails the union of authorship and perspectives "within time (evolutionary invisible college) and space (geographical invisible college) coordinates" (Peñaranda & Osca, 2013, p.130).
There are different studies on invisible colleges and their impact on science. The work done by Monsalve, Cerdá and de Andrés (2004) is interesting. They studied 520 papers signed by 1,172 authors by evaluating the lead author, number of articles written, number of authors comprising it, number of signatures per paper and index of articles per author. They classified the authors as transient (those who signed once during the period studied), aspiring (those who came to sign on up to four occasions), moderate (those whose signature appeared five to nine times) and productive. In the communication field, and at an international level, invisible colleges and their determination through bibliometric studies have also been explored (Chang & Tai, 2005;Tai, 2009).
One of the fundamental questions of this paper is formulated as follows: Is that collaborative exchange joint when it is linked to specific gender studies? Or, conversely, does one sex dominate the leadership of research projects? If so, two new traditional gender concepts reappear: the velvet ghetto and the glass ceiling. Have women gender researchers created a velvet ghetto for themselves in this field? Have they raised a glass ceiling 68 Caro González & Guarinos -Male presence in gender research for male researchers that leads them to play a secondary role in gender research networks?

Objectives and Hypothesis
The specific purpose of this article is to identify the network of Spanish authors conducting research from a gender perspective in communication. For this, the citations produced among those who have published in Spanish communication journals indexed in SCOPUS during the period 2008-2014 were analysed.
The hypotheses upon which we are basing our work relate to the role played by men in the gender researchers network in communication. A superficial observation of gender research forums and of research groups focusing on the subject show that the presence of men working in this field is very scant and they have less weight on research. Our first hypotheses focus on this statement: (1) Hypothesis 1: There are few men conducting research from a gender perspective in communication.
(2) Hypothesis 2: Men conducting gender research in communication play a secondary role in the researcher network.
Should the two hypotheses be proven, we put forward two more hypotheses aimed at identifying a potential reason for the scant presence of men in the gender researchers network. We posit the fact that there is a tendency among researchers to cite those of the same sex. In this case, we are talking about homophily or assortativity, which is the tendency of people to relate to others that are like them (McPherson et al., 2001).
(3) Hypothesis 3: Women show a preference for other women when citing. (4) Hypothesis 4: Men show a preference for other men when citing.

Methodology
To respond to these hypotheses, we turned to a bibliometric study of scientific publications in the field and the citations produced among their authors.
In view of the limited internationalization of publications from communication academics in Spain (Casanueva & Caro, 2013;Masip, 2011), it was decided to analyze the papers published in the Spanish journals of the area. SCOPUS was chosen against other standardized international databases as SSCI. Only one journal in communication was indexed in 2014 in SSCI compared to six in SCOPUS.
Articles on gender published in Spanish journals in the communication field and indexed in the SCOPUS database were identified. The journals were: Comunicar, Estudios del Mensaje Periodístico, Comunicación y Sociedad, Revista Latina de Comunicación Social, Historia y Comunicación Social and Signa. The study period considered was from when the journals were included in SCOPUS up to 2014. The first one to be included was Comunicar (2008)  For the article search, a key word search strategy was followed. A search for the term GENDER was carried out in the fields TITLE, ABSTRACT and KEYWORD. This search generated 60 references. These results were immediately filtered by reading each one to ensure that the research took into account the gender perspective. Only one of the publications did not meet this requirement, therefore, the total number of studies analysed was 59. They are shown in Table 1. The full bibliographical reference can be found in Annex 1. Table 1 Articles that form the research network from a gender perspective in communication

Authors
Year Journal Callejo, J. 2013 Comunicación y Sociedad Castillo, A., & Carretón, M. C. 2010 Comunicación y Sociedad Crescenzi, L., Araüna, N., & Tortajada, I. 2013 Comunicación y Sociedad  Citations were then extracted from each article and a matrix was built with the authors of the articles who were cited. Centrality and homophily statistical procedures were applied to this matrix using the specific UCINET VI software for the analysis of social networks.

The Citation Network
The analysis of social structures must be based on specific data, not on the characteristics of individuals, but rather on their social links. For this, the analysis of social networks is used (Wellman & Berkowitz, 1988). The purpose of the analysis of social networks is to learn about the joint structure of the links between the players, enabling the existence of any general relationship patterns resulting from the abstraction of the elections of individuals or from the relationships between elements to be identified. A network can be defined simply as a set of nodes or elements related to each other. Therefore, the starting point of the analysis of social networks is the study of these two basic units: the nodes representing the players or elements of the network and the relationships between these nodes (Casanueva & Caro, 2013, p.65).
Before starting the analysis of social networks, we obtained confirmation of the first of our hypotheses. The proportion of women and men who study the gender perspective is very unequal: 72% of the authors are women, while the overall proportion in Spanish journals in all fields is 37% (Mauleón et al., 2013), which initially denotes greater interest on the part of women in this kind of research.
To contrast Hypothesis 2, we used the social networks analysis that enables the discovery of structures and associations between elements based on their relationships. In this case, the network comprising 110 authors from the 59 publications selected was analysed. The relationship studied was the citations between them, in other words, when an author of any of the articles cites another from the list. Crane (1972) suggests that it is possible to understand the social and intellectual structure of a field of knowledge or a certain disciplinary focus by studying the different relationships between the authors, in particular by analysing the citations.
In total, the 110 authors cite 1,814 in their bibliographical references, but our interest was limited to the relationships within the network we 74 Caro González & Guarinos -Male presence in gender research defined. These relationships were not symmetrical. We did not take into account the weight of the relationship, that is, the number of times an author cites another, simply the fact that they were cited. We worked with dichotomous relationships. In this way, a matrix of 110 x 110 elements can be built, in which a 1 in each box indicates that the author in the row cites the corresponding author in the column and 0 in the other case.
The matrix can be represented in a chart, such as the one shown in Figure 1. The structure of the network can be analysed through a series of indicators provided by Ucinet VI software (Borgatti, Everett, & Freeman, 2002). The size was 110 and the density 0.0325, which is rather low (only 3.2% of all the potential relationships). This result showed scant cohesion between the authors who take the gender perspective into account in communication. It was observed, therefore, that there were few social relationships and little collaboration in the citation relationships.
The network had 22 isolated elements (authors who are not cited and do not cite other authors in the network). This may be due to the fact that they were recently included in the topic and, therefore, their papers could not have been familiar to other researchers.
Below we show the most relevant authors in this network identified from two measurements of centrality ( Table 2). One of them is the degree of centrality, that is, the authors who are cited more frequently. It corresponds to the number of arrows received by each author in the chart in Figure 1. The most central authors in the network are Bringué Sala, García Muñoz, Martín Llaguno y Tous Rovirosa. Of the 10 with greater centrality, eight are women (a proportion of 80%, higher than that of the total sample, 72%).
The other measurement is intermediation, which refers to authors who have more connections with different groups in the network. In other words, the number reflects the times that the author appears in the paths that connect to all the pairs of authors in the network. These nodes are called "bridge nodes." It can be observed how, with the exception of two of them, the authors were different from the previous ones: Etayo, Piñeiro, Simelio-Sola y Tous Rovirosa. Some of these intermediation positions can be clearly observed in Figure 1. In this case, the prevalence of women is absolute. The scant number of male researchers (28% of the total) and the little relevance they have in the network confirmed the second hypothesis: men play a secondary role in gender research networks in the field of communication.
To confirm the third and fourth hypotheses, a study of the correlation of the citation matrix was conducted with the independent variable "sex of the authors." The degree of homophily was measured, in other words, the preference that people from a group have for those from the same group, in our case, the preference of women for citing other women and of men for citing other men. We again used Ucinet VI and applied the "hemophilic variable" method. The following table outlines the coefficients obtained. It can be observed how the correlation is significant in the case of women, therefore, hypothesis 3 is confirmed. In other words, women show a preference when citing for people of their same sex. In the case of the "men" group (hypothesis 4), the relationship is not significant, therefore, it is shown that there is no homophily when citing among them.

Conclusions
The analysis of the citations in articles from a gender perspective published in communication journals shows significant questions. Firstly, it should be pointed out that the number of articles published from this perspective is very scant. In the period analysed, the six journals considered published 1,763 articles, of which only 59 include a gender perspective (3.3%).
Secondly, it is important to stress that most of the publications are from recent years, which implies that interest in the field of communication in gender is recent, or that, at least, interest in the topic on the part of the publishers of these journals has come about recently.
With regard to researcher networks, it can be observed that gender studies in communication are a question of sex. It is women who make up the majority of the network. It is also women who have greater centrality in said network and serve as elements of connection between the different sub-groups that could be identified. Other studies on Spanish scientific publications in all fields show that men are the ones who play that role (Mauleon et al, 2013).
But the most significant contribution of this paper is the identification of a significant degree of homophily between women who cite articles published by other women. This data suggests the existence of invisible colleges, in which certain positive discrimination towards the female sex can exist when it comes to citing. Could the existence of any kind of "Matilda effect" be suggested (Rossiter, 1993) in the case of men who conduct research on gender matters? This effect, scientifically studied and documented, shows the discrimination suffered by women in science, reflecting the denial of their contributions and discoveries that were assigned to their research colleagues. In this case, the effect would affect only the preference of women for their colleagues of the same sex and not the androcentric subordination behaviour of the woman defined by Rossiter. These conclusions could suggest the existence of intentional behaviour on the part of women; they simply indicate that this tendency exists and that it would be interesting to delve deeper into the reasons therefor.

Limitations and Future Lines
To better understand this phenomenon, it is necessary to conduct new research based on both qualitative and quantitative data. To that end, in future research, in-depth interviews shall be held with women and men gender researchers from the field of communication. A study is also being conducted on the components of research groups, R&D projects and the chairing of doctoral theses and of scientific meetings and conferences on gender in communication. The qualitative evidence will serve to find out whether there are differences between the motivations of women and men when it comes to conducting research in this field and their perception with regard to the role played by men. The quantitative data will help delve more deeply into the researcher network and their relationships, and to find out whether the presence and role of men in these scientific activities, not directly linked to production, remains similar to the information obtained in this study.
We invite researchers from the field (or others) to contribute evidence in favour of or against this hypothesis. In any case, the researchers signing this article defend the need for institutional support for women's development in science and research, with a reminder that, in the case of unequal conditions in Spain, male researchers are 2.5 times more likely to receive a promotion than women researchers , and for the development of policies encouraging male researchers to join this necessary field of research.