Matilda Effect in the Hispanic American Communication co-authorship network El Efecto Matilda en la red de coautorías Hispanoamericana en Comunicación

Resumen Investigaciones recientes con perspectiva de género han confirmado la pervivencia del “Efecto Matilda” en la ciencia, haciendo que las aportaciones y el papel de las mujeres sigan quedando relegados dentro de sus comunidades científicas. En este contexto, la propuesta que ocupa estas páginas se centra en comprobar si el fenómeno también se produce en la comunidad hispanoamericana de investigación en Comunicación, realizando para ello un análisis sociométrico sobre la red de coautorías. Los resultados señalan que pese a que tres de los cinco puestos más centrales de la red están ocupados por féminas, en términos generales se registran diferencias significativas que desplazan a las investigadoras a posiciones más periféricas. Del mismo modo, se ha detectado que las comunidades de investigación formadas por las distintas comunidades tienden a estar organizadas en torno a varones, confirmando así la presencia del Matilda effect en la red social de la disciplina de Comunicación. Cabe señalar que estas conclusiones se extrapolan a unos parámetros intelectuales (Comunicación) y geográficos (España y Latinoamérica) concretos, por lo que la puerta a estudios en otros ámbitos queda abierta.


Introduction
Given that women's scientific contributions have traditionally received less visibility and recognition than those of their male colleagues, there is a need to implement new approaches and reflections. Significant breakthroughs made by women in the past were usually attributed to men, a phenomenon first described by the activist Matilda Joslyn Gage and known today as the "Matilda effect" (Rossiter, 1993;Stamhuis, 1995;Benschop and Brouns, 2003; Lincoln, Pincus, Koster and Leboy, 2012; Kretschmer, Kundra, Beaver and Kretschmer, 2012; Knobloch-Westerwick, Glynn and Huge, 2013). Indeed, the scientific community's acknowledgement that the "glass ceiling" in science remains as strong as ever (Bain and Cummings, 2000;Guil, 2008; Gallego-Morón and Matus-López, 2020) and that, as such, it must be shattered, has given rise to a profusion of papers on the subject since the 1990s.
Gallego Morón and Matus-López (2020: 105) argue in one of their latest collaborations that, "in contrast to the proliferation of research on the causes of vertical segregation in the university, there has been less empirical research on the positive conditioning factors in women's trajectories". Without detracting from this claim, ever since the historian of science Margaret Rossiter coined the term "Matilda effect" in 1993, the marginalization of women has been studied in disciplines as diverse as technology, engineering, the arts, and medicine. However, in light of the scarce research on this effect in the Social Sciences and Communication, this paper reviews the state of the question in the latter. Accordingly, it is worth highlighting the experiment that Knobloch-Westerwick, Glynn, and Huge conducted in 2013, where they noted a certain tendency to associate publications by male authors with higher scientific quality, while at the same time a greater interest in collaboration with men.
The literature review undertaken in this study shows that the tendency to overlook the achievements of female academics and scientists is decreasing, although the practice to minimize the citation of works and papers signed by female authors persists (Davenport and Snyder, 1995 (Carter, Croft, Lukas, and Sandstrom, 2018) and are gauged as less competent and less worthy of a high starting salary (Moss-Racusin, Dovidio, Brescoll, Graham, and Handelsman, 2012). Indeed, Zhang and Li (2020) find that references to specific contributions increase when the author's name evokes the "neutral" gender (Fryer and Levitt, 2004;Sumner, 2018), for the same reason that in the publishing industry, female writers choose "gender-neutral names" as pseudonyms to increase their book sales (Finn, 2016).
Even more controversial, and far from the trend of incorporating a greater gender perspective into science in addressing the lack of opportunities (Alonso, Diz, and Lois, 2016), a certain sector of academia holds that the differences between male and female researchers could be influenced by behavioral factors innate to gender differences (Udry, 1994), which theoretically explain why female researchers tend to publish less ( In contrast to Summer's premise, recent analyses suggest that gender diversity may benefit teamwork, as groups including female researchers are more creative and produce higher quality outcomes. Further, gender variety reinforces objectivity for processing information and reducing unconscious biases (Vásárhelyi, 2020) that, in some instances, lead to self-imposed standards of excellence, since female authors seem more reluctant to accept open review practices (Segado-Boj, Martín-Quevedo, & Prieto-Gutiérrez, 2018).
This research does not seek to elucidate whether the gender of the researcher shapes scholarly collaboration; instead, it addresses how women's status and reputation are undervalued due to the perpetuation of stereotypes that should be eradicated. Thus, this particular research follows the line defended by that group of authors for whom there is a direct impact of gender bias in the perpetuation of the Matilda effect, that is, the systematic under-recognition and denial of women's contributions to Engineering, Technology, Mathematics and, particularly, Communication Sciences.
Within this theoretical framework, it is also relevant to incorporate the concept contributed in 1996 by UNESCO known as the "leaky pipeline," since it constitutes the perfect metaphor to "illustrate the disproportionate flow of women who leave the professional system in the field of science" (Castaño, 2010: 256), in comparison to the bulk of men who survive each rung on the academic ladder or equivalent (Pell, 1996;Blickenstaff, 2005).
Although the underlying reasons for the existence of the Matilda effect and the leaky pipeline are diverse, complex, and difficult to resolve (Jones et al., 2014), this paper delves into the former phenomenon with a methodology focused on identifying gender discrepancies. This analysis is often based on easy-to-obtain metrics, such as the number of publications, citations, or grants received. However, as Faulkner (2009) indicates, what is fundamental is to discern the level of "visibility" achieved by female researchers in their discipline, explore the processes of resistance and marginalization, and, ultimately, propose strategies to promote greater inclusion and success of female academics and scientists in the Communication field.

The co-authorship network as a reflection of the social structure of a discipline
This paper analyzes the social structure reflected in the Spanish and Latin American co-authorship network to identify the marginalization of female researchers in Communication studies. Network analysis is understood as a way to measure the "visibility" or "reputation" of women within an area. Not surprisingly, works focused on other disciplines and geographical contexts have used similar approaches to demonstrate that women in areas such as Accounting tend to occupy the more peripheral places (Dias et al., 2019), while in the case of Urban studies, they tend to have fewer opportunities to direct doctoral theses (Walker and Boamah, 2019). However, in areas such as Knowledge Organization Systems (Karimi, Mayr and Momeni, 2019) or Industrial and Organizational Psychology (Fell and König, 2016), no significant differences have been found in centrality according to gender. In other words, the lack of conclusive results urges further research.
The study of "co-authorships," the practice of collaborating in research that results in papers coauthored by two or more authors (Díaz-Campo and Segado-Boj, 2017), provides the ideal scenario to test the perpetuation of the Matilda effect and to verify whether collaborations are a reflection of existing relationships within any discipline (Otte and Rousseau, 2002). Furthermore, such synergies can be visualized and analyzed as a framework for investigating the social structure of a subject, and this, in turn, not only determines its cognitive level and influences the production of knowledge (Bordons, Aparicio, González-Albo and Díaz-Faes, 2015), but also makes it possible to identify the most authoritative and relevant authors in each field (Kumar, 2015).
In short, network analysis allows us to identify the most eminent researchers in a field since highly regarded academics typically attract more collaborators and co-authors (Leifeld, Wankmüller, Berger, Ingold, & Steiner, 2017). Thus, within the network, the "star" scholars of a subject monopolize the central places and raise their "status" (Coate and Howson, 2014), while those yielding less influence are relegated to the periphery, where women tend to be more often located. However, centrality is not the only index of academic prestige, and other scales such as productivity or citations could be included.

Objectives
The persistence of the Matilda effect in many fields of research has led us to ascertain the extent to which this phenomenon manifests itself in the co-authorship network of Communication studies by comparing the production of women with that of their male colleagues. The aim is to identify how gender differences can mark differences in prestige among researchers with similar interests or, in other words, to discover whether the Matilda effect impacts the centrality of women within the co-authorship network in the Latin American community. Having verified that the proliferation of research on the Matilda effect and the leaky pipeline in the scientific area has hardly been extended to the Hispanic-American field, the present paper aims to fill the gap in the literature by testing the two hypotheses that guide the development of this work: 1) Women are placed in the more peripheral positions in the general network of Hispanic American coauthorships in Communication, while men are located in the central nodes.

Previous studies have pointed out that collaboration among authors is a growing trend both in Latin
2) The communities created within the general co-authorship network are organized around men.
Hypothesis 1 is proposed from a macro perspective, from a broad point of view that takes in the entire research community in Communication created by scholars affiliated with Spanish and Latin American institutions. Hypothesis 2 is defined from a meso perspective; that is, it analyzes the places these researchers occupy within the various stable communities of collaboration in the period under study. Thus, a researcher may occupy a secondary or peripheral position in the general co-authorship network, yet a central position within the research community in which he or she habitually collaborates.
Because the underlying reasons for the existence of the Matilda effect and the leaky pipeline are diverse and largely difficult to resolve (Jones et al., 2014), the authors addressed them with the precise methodology outlined below.

Methodology
This work applied network analysis to the Spanish-American co-authorship community of Communication studies, with a sample comprising 2,343 papers published in JCR journals between 2015 and 2019. The data were retrieved from the Incites service. The search criteria included documents assigned to the knowledge area "Communication" in Web of Science and written by at least one author affiliated with a Spanish or Latin American institution. In this regard, it should be noted that InCites allows institutions to be identified by specific countries and geographic areas, hence our use of the Latin American categorization available in the bibliographic service. Thus, it was found that the average number of authors per document was 1.59, and that of the 2,343 papers retrieved, 643 had some international collaboration (Annex I shows the frequency of appearance of the authors' national affiliations). The downloaded bibliographic information was converted to network data using VosViewer (Van Eck, Waltman, 2010). In addition, the data were manually preprocessed by one of the authors to normalize variants in the names of the same author under a unique identifier. For example, Rosa Berganza could appear in the database as "Berganza, Rosa", "Conde, MR", "Berganza Conde, María" or "Rosa Berganza-Conde, María".
The growing trend in International co-authorship in both Spain and Latin America (Segado-Boj, Prieto-Gutiérrez, Díaz-Campo, 2021 has led to the inclusion of researchers in the network from those countries, mainly from Europe and the United States, with which the former collaborate. The individual degree centrality of all the network nodes was obtained using Pajek software (Batagelj and Mrvar, 1998), recommended for visualizing large networks with thousands or millions of vertices.
In this regard, it should be emphasized that such centrality degree measures the connections that a given node maintains with other vertices (Freeman, 1978), while the scenario formed by co-authorship networks reveals the frequency with which a researcher collaborates -or has collaborated-with other authors (Collazo, Luna and Vélez, 2010; Ávila-Toscano, Vargas-Delgado and Oquendo-González, 2020).
We first calculated individual centrality degree indicators, after which we confirmed that the relationships in the network are usually concentrated in a more or less limited set of researchers before identifying the gender of each scholar. For the sake of simplicity, the only distinction was made between binary genders (male and female) since the frequently termed "first name" is regulated by "severe social and cultural norms" "in subjection to a heteronormative identity regime that obligatorily links sex with gender" (Vacarezza, 2018: 14). Consequently, the categorization was carried out manually, taking into account the name of the authors of each paper and resolving the cases that involved doubt by searching for more information about the author.
Given that the degree centrality values do not correspond to a normal distribution within the sample (p<0.001 according to the Shapiro-Wilk test), a nonparametric test (Wilcoxon-Mann Whitney) was applied to look for significant differences between the two categories. This statistical test shows significant differences in the presence of a quantitative value -in this case, degree centrality -between two sets within a sample, here, male or female. When the result of the p-value was equal to or below 0.05, the hypothesis that there is a significant difference in the centrality of subjects belonging to both groups was verified.
Besides the statistical tools described above, a simplified graphical representation of the network was also made to visualize the different positions occupied by the researchers. This visualization was developed using the Kamada-Kawei (1989) algorithm, which organizes the nodes according to their centrality.
Within extensive co-authorship networks, there are smaller research communities whose members collaborate with each other more frequently than with other nodes (Newman, 2001). Network analysis makes it possible to detect and identify these communities through various clustering algorithms (Porter, Onnela, and Mucha, 2009). These clusters show groups of authors who have formed stable and regular collaboration networks in the period analyzed, without expressly belonging to the same constituted research group or formally collaborating on a research grant. In this context, our proposal elucidated whether female researchers had the same possibilities as their counterparts to occupy central places in their communities. This analysis thus was conducted both at the macro level (the entire network of researchers) and at the meso level (the stable and frequent communities of collaboration between smaller groups of authors).
VosViewer software(Van Eck, Waltman, 2010) was used to achieve the last objective since it is one of the most intuitive tools for visualizing bibliometric networks and locating clusters, i.e., sets of researchers that frequently collaborate showing the existence of formal or informal research communities. Thus, the VoSViewer algorithm allowed us to identify the academics with greater centrality within each cluster, quantifying the connections between the different nodes.
It should be noted that this study only considered clusters comprising at least five interconnected elements. Following a relevance criterion, we only analyzed those clusters with a minimum number of nodes (researchers) to avoid particular or excessively personal aspects of smaller communities that might distort the results. The authors placed in the central positions were identified in terms of the number of co-authorships (number of researchers with whom the subject collaborated, rather than total frequency of collaboration). Likewise, in cases where nodes had identical centrality, all were placed at the center of the cluster, resulting in 36 out of 113 clusters having more than one central node.
Finally, when dealing with two categorical variables (the researcher's gender and the leadership of a research community), the chi-square test was applied to determine whether the two were related. A p-equal to or less than 0.05 implied a relationship between the variables considered. In other words, this result suggested that at least one of the relationships between two of the categories yielded a lower or higher distribution than would be expected from a theoretical projection of the distribution of frequencies.
All statistical tests were performed using the R programming language.

Results
The study sample comprised 2,601 individual authors, where 1,298 (49.9%) had a first name linked to the female gender and 1,303 (50.1%) to the male gender. The average centrality degree was 0.003 (standard deviation = 1.732), and the median in this sense was 2. The results are shown below in a disaggregated manner, considering both the centrality according to gender and the differences in the research community leader.

Differences in centrality according to gender
As shown in Table 1, when disaggregated by gender, our results show a greater degree centrality in the case of men, both in the average and in the median. Source: Authors Likewise, the Wilcoxon-Mann Whitney test was significant (p=.019) when comparing degree centrality according to the gender of each researcher. Although most male and female authors had similar levels, Figure 1 shows more male academics in the leading positions.  Figure 1 shows that, although most researchers share the same centrality degree, in the smallest cases regardless of gender, the situation changes with those with a centrality higher than 6. Moreover, this trend is repeated in Figure 2, which shows a visual representation of the network that includes only those authors who have published at least three papers in the period analyzed (again, mostly men).

Figure 2. Visualization of the co-authorship network
The red nodes represent male scholars, while the gray ones correspond to female scholars. As indicated in the methodology, the visualization places authors with a higher centrality degree, i.e., those who collaborate most with other authors, at the core of the graph. Thus, the size of the nodes is proportional to the number of papers published by each author, and the connections between nodes imply that two authors have collaborated at least once. Similarly, the thickness of these links is directly proportional to the frequency with which two researchers (represented by the corresponding node) have co-authored a paper. That is, the thicker the nexus, the higher the number of papers in which these authors have collaborated.
The results indicate that, although a relatively varied and large group of researchers is indeed located in the center of the network, the central positions are occupied by men to the detriment of women. Thus, female researchers are again relegated to the more peripheral or even disconnected positions in the network. When they appear in the more central places, they do so less often, in an isolated or disconnected way. In other words, the collaboration structure of their male counterparts is characterized by a greater cohesion and connection that, in the end, favors collaborations or co-authorships.
The displacement of female researchers to the periphery of the network can also be seen in Table 2  Source: Authors With these elucidations, it was then of interest to know whether these differences also extend to a leader of a research field since higher standards are imposed on female researchers directing a project (Bornmann et al., 2007).

Differences in community leadership
The results of the Chi-square test suggested that the categories "gender" and "cluster center" are related to each other (DF=1, p=.028). Table 3 shows that the presence of male researchers as cluster center leaders exceeds the expected theoretical projection that would correspond to them if these two values were unrelated.

Source: Authors
According to the statistical test applied, the resulting data pointed to the high probability that male researchers occupy the central position in their community and, by extension, lead their respective research communities. On the other hand, women continued to have fewer opportunities to head the projects in which they are involved, at least within the field of Communication studies.

Discussion and conclusions
The findings of this work support our claim that there is a Matilda effect in relation to the centrality of female authors in the Spanish and Latin American community of Communication researchers. Not surprisingly, by taking degree centrality as an indication of the prestige achieved by scholars, we can confirm there is an inequality in the status of women compared with that of their colleagues in the discipline. However, despite the preference for male specialists in collaboration and co-authorship, as in other academic fields, there are hopeful signs of greater equality in this specific context. After all, the presence of women in the most central positions in the network (see Table 2) shows that, of the five most prominent specialists in the Latin American field, three are women.
Michelle Dion, Jane Lawrence Sumner, and Sara McLaughlin Mitchel argue in their 2018 collaboration that women in the scientific environment receive more recognition as their disciplines and subfields become "more gender diverse" (Dion et al., 2018: 325). In the short to medium term, the authors assess that in fields such as Communication, the bleak situation shown in the graphs on centrality--apart from the elite positions--could be corrected by empowering feminization in the field in question.
Forecasts aside, the current results show that women are experiencing a tendency to be significantly displaced to the periphery of their research communities and that collaboration is organized chiefly around men. Of course, the small group of women researchers that has managed to break through the glass ceiling are placed in more central positions of the networks, and -by extension -project their reputation within the discipline, but the situation is less promising for the bulk of female scholars, who appear far from prestigious positions.
The snapshot shown by the data interpreted here implies that a female researcher in Communication is more likely to encounter dynamics that displace her to peripheral positions in the network of coauthorships, either because of factors related to the perpetuation of stereotypes; for reasons that would explain the phenomenon of "leaky pipelines" to which scholars give dissimilar importance (Blickenstaff, 2005); due to the greater self-demanding nature of women quantitatively affects their production, or for other reasons that could be explored in future works through in-depth interviews or focus groups.
There is no doubt that the dynamics of inequality that were intuited when the two-verified--hypotheses guiding this proposal were put forward will eventually be broken: 1) Women occupy more peripheral places in the network of Latin American co-authorships in Communication, while men are located in the central points, and 2) The smaller communities created within the general network, in turn, revolve around men. Until the gender gap is ended, however, the need for further research is pressing.
To conclude, the results support the theory of marginalization of women highlighted in disciplines such as Urban Studies (Walker and Boamah,  It should also be noted that this work only examined the co-authorship relationships existing at a level of scientific production in Communication, namely, that of the papers published in JCR journals. Although this index is said to include the highest quality and reference results within the discipline, the fact is that scientific production in the field of interest is not limited to this database. In other words, it is possible to check whether the findings provided have been reproduced at other levels, such as in journals indexed in the Emerging Source Citation Index or Scopus. Similarly, in our paper, neither the authors' order of signature nor the status of corresponding author was used as a variable. Nevertheless, such features influence the attribution of merit and responsibility in the published manuscripts (Díaz-Campo and Segado-Boj, 2017). Therefore, it remains to be seen whether female authors also occupy lower positions in the order of authorship and receive recognition as corresponding authors on fewer occasions than their male counterparts.
Likewise, the study of centrality in communities of authorship has been limited to those clusters composed of at least five nodes, so the option of exploring whether the dynamics indicated in this study have been reproduced in smaller groups or communities also remains open.
Another prospective question would lie in whether the lack of balance in the network responds to cultural, political, or structural reasons of the Latin American research space since it is worth asking whether the phenomenon occurs with identical characteristics in scenarios such as the Anglo-Saxon one. The questions to be addressed in future studies are: What does the greater inclusion and success of female academics and scientists in the field of Communication depend on, and what factors would halt the leaky pipeline that alludes to the loss of female talent?
Answering these questions and overcoming the Matilda effect will require raising awareness of the problem and "a gender-blind environment" for what constitutes the scientific core of the Social Sciences (Dion et al., 2018). An ambitious goal in which, by all accounts, efforts will have to be combined without distinction.

Coauthorship network as reflection of the social structure of the discipline
This paper analyzes the social structure reflected in the Spanish and Latin American co-authorship network to identify the marginalization of female researchers in Communication studies. Network analysis is understood as a way to measure the "visibility" or "reputation" of women within an area. Not surprisingly, works focused on other disciplines and geographical contexts have used similar approaches to demonstrate that women in areas such as Accounting tend to occupy the more peripheral places (Dias et al., 2019), while in the case of Urban studies, they tend to have fewer opportunities to direct doctoral theses (Walker and Boamah, 2019). However, in areas such as Knowledge Organization Systems (Karimi, Mayr and Momeni, 2019) or Industrial and Organizational Psychology (Fell and König, 2016), no significant differences have been found in centrality according to gender. In other words, the lack of conclusive results urges further research.
The study of "co-authorships," the practice of collaborating in research that results in papers coauthored by two or more authors (Díaz-Campo and Segado-Boj, 2017), provides the ideal scenario to test the perpetuation of the Matilda effect and to verify whether collaborations are a reflection of existing relationships within any discipline (Otte and Rousseau, 2002). Furthermore, such synergies can be visualized and analyzed as a framework for investigating the social structure of a subject, and this, in turn, not only determines its cognitive level and influences the production of knowledge (Bordons, Aparicio, González-Albo and Díaz-Faes, 2015), but also makes it possible to identify the most authoritative and relevant authors in each field (Kumar, 2015).
In short, network analysis allows us to identify the most eminent researchers in a field since highly regarded academics typically attract more collaborators and co-authors (Leifeld, Wankmüller, Berger, Ingold, & Steiner, 2017). Thus, within the network, the "star" scholars of a subject monopolize the central places and raise their "status" (Coate and Howson, 2014), while those yielding less influence are relegated to the periphery, where women tend to be more often located. However, centrality is not the only index of academic prestige, and other scales such as productivity or citations could be included.

Discussion and conclusions
The findings of this work support our claim that there is a Matilda effect in relation to the centrality of female authors in the Spanish and Latin American community of Communication researchers. Not surprisingly, by taking degree centrality as an indication of the prestige achieved by scholars, we can confirm there is an inequality in the status of women compared with that of their colleagues in the discipline. However, despite the preference for male specialists in collaboration and co-authorship, as in other academic fields, there are hopeful signs of greater equality in this specific context. After all, the presence of women in the most central positions in the network (see Table 2) shows that, of the five most prominent specialists in the Latin American field, three are women.
Michelle Dion, Jane Lawrence Sumner, and Sara McLaughlin Mitchel argue in their 2018 collaboration that women in the scientific environment receive more recognition as their disciplines and subfields become "more gender diverse" (Dion et al., 2018: 325). In the short to medium term, the authors assess that in fields such as Communication, the bleak situation shown in the graphs on centrality--apart from the elite positions--could be corrected by empowering feminization in the field in question.
Forecasts aside, the current results show that women are experiencing a tendency to be significantly displaced to the periphery of their research communities and that collaboration is organized chiefly around men. Of course, the small group of women researchers that has managed to break through the glass ceiling are placed in more central positions of the networks, and -by extension -project their reputation within the discipline, but the situation is less promising for the bulk of female scholars, who appear far from prestigious positions.
The snapshot shown by the data interpreted here implies that a female researcher in Communication is more likely to encounter dynamics that displace her to peripheral positions in the network of coauthorships, either because of factors related to the perpetuation of stereotypes; for reasons that would explain the phenomenon of "leaky pipelines" to which scholars give dissimilar importance (Blickenstaff, 2005); due to the greater self-demanding nature of women quantitatively affects their production, or for other reasons that could be explored in future works through in-depth interviews or focus groups.
There is no doubt that the dynamics of inequality that were intuited when the two-verified--hypotheses guiding this proposal were put forward will eventually be broken: 1) Women occupy more peripheral places in the network of Latin American co-authorships in Communication, while men are located in the central points, and 2) The smaller communities created within the general network, in turn, revolve around men. Until the gender gap is ended, however, the need for further research is pressing.
To conclude, the results support the theory of marginalization of women highlighted in disciplines such as Urban Studies (Walker and Boamah,  It should also be noted that this work only examined the co-authorship relationships existing at a level of scientific production in Communication, namely, that of the papers published in JCR journals. Although this index is said to include the highest quality and reference results within the discipline, the fact is that scientific production in the field of interest is not limited to this database. In other words, it is possible to check whether the findings provided have been reproduced at other levels, such as in journals indexed in the Emerging Source Citation Index or Scopus. Similarly, in our paper, neither the authors' order of signature nor the status of corresponding author was used as a variable. Nevertheless, such features influence the attribution of merit and responsibility in the published manuscripts (Díaz-Campo and Segado-Boj, 2017). Therefore, it remains to be seen whether female authors also occupy lower positions in the order of authorship and receive recognition as corresponding authors on fewer occasions than their male counterparts.
Likewise, the study of centrality in communities of authorship has been limited to those clusters composed of at least five nodes, so the option of exploring whether the dynamics indicated in this study have been reproduced in smaller groups or communities also remains open.
Another prospective question would lie in whether the lack of balance in the network responds to cultural, political, or structural reasons of the Latin American research space since it is worth asking whether the phenomenon occurs with identical characteristics in scenarios such as the Anglo-Saxon one. The questions to be addressed in future studies are: What does the greater inclusion and success of female academics and scientists in the field of Communication depend on, and what factors would halt the leaky pipeline that alludes to the loss of female talent?
Answering these questions and overcoming the Matilda effect will require raising awareness of the problem and "a gender-blind environment" for what constitutes the scientific core of the Social Sciences (Dion et al., 2018). An ambitious goal in which, by all accounts, efforts will have to be combined without distinction.