The rapid rise in popularity of online social networking has been followed by a slew of services aimed at an academic audience. This project sought to explore network structure in these sites, and to explore trends in network structure by surveying participants about their use of sites and motivations for making connections. Social network analysis revealed that discipline was influential in defining community structure, while academic seniority was linked to the position of nodes within the network. The survey revealed a contradiction between academics use of the sites and their position within the networks the sites foster. Junior academics were found to be more active users of the sites, agreeing to a greater extent with the perceived benefits, yet having fewer connections and occupying a more peripheral position in the network.
Discussion and conclusions
Online social networking has developed rapidly in the past decade (boyd and Ellison, 2007; Rainie and Wellman, 2012), including a number of social networking sites (SNS) aimed specifically at academics (Nentwich and König, 2012). Academic SNS are characteristically venture-capital funded startup businesses (CrunchBase, 2012; Empson, 2012; Hammersley, 2010) independent of higher education institutions.
The ways in which academic SNS may benefit academics has received greater focus in the academic literature than the ways that such services are being used in practice. Facilitating research collaboration and enhancing scholarly communication are viewed as the principal benefits of academic SNS (Greenhow, 2009; Weintraub, 2012; Xu, et al., 2008; Zaugg, et al., 2010), underpinned by development of an online academic identity.
Extant studies of academic SNS have begun to shed light upon these issues in practice. The nature of collaboration has been addressed by two studies, focused upon the nature of group formation in Mendeley (Jeng, et al., 2012; Oh and Jeng, 2011). A positive correlation between group size and the extent of interdisciplinarity was found (Oh and Jeng, 2011), and followed by exploring the factors affecting group formation (Jeng, et al., 2012). Construction of an online identity has been addressed by analysis of profile contents from Academia.edu (Almousa, 2011; Menendez, et al., 2012). Differences in profile completion were discovered based on discipline (Almousa, 2011), academic seniority, university ranking, and country (Menendez, et al., 2012), with the latter concluding that the platform serves to preserve existing hierarchies.
However, drawing upon a single platform may present an impoverished view of academics’ online scholarly practices. In contrast, Bukvova (2012) took a holistic approach, presenting an analysis of the types of online channels academics may use, and the ways in which they are used. In this instance, academic SNS are found to predominantly perform the passive role of an online business card , in contrast to the more active potential benefits of collaboration and communication identified by previous work.
Existing studies have relied upon content analysis to explore academic SNS; no previous work has looked at the network structure of academic SNS, although this is a potentially valuable approach to explore the nature of collaboration supported by the platforms, and to explore online academic identity in terms of connectivity and behaviour within the network structure. As there are several sites, there is a question of whether the structure of academic communities is similar in different academic SNS. To address this, the following research questions guided the study:
- What is the structure of academic social networks? Do factors such as discipline or career level correlate with behaviour in the network?
- To what extent do different academic SNS foster similar networks? Can we find any characteristics of academic social networks by comparing a similar sample across different sites?
- Can we explain patterns in the structure of academic social networks? What do academics perceive to be beneficial by being part of the network?
While academic SNS vary to an extent (Bullinger, et al., 2010), SNS are Web sites which allow users to create a profile and make connections with others (boyd and Ellison, 2007). In this study, academic SNS are defined as sites which fit this definition but are aimed explicitly at the academic community. In the early stages of this study a list of academic SNS was compiled by size and relative popularity. This drew upon lists in previous articles (Bullinger, et al., 2010; Huggett, 2010; Leeder, 2008; Menendez, et al., 2012) and Internet searches, and the data updated at the time of writing. Several of the sites listed in previous articles are now defunct; others which no longer fit the definition may have modified their model in order to be profitable. The size and Web ranking of the sites which fit the academic SNS definition are shown in Figure 1; this shows a cluster of large, popular academic SNS with approximately two million users, after which number of users drops dramatically with decreasing Web rank.
Figure 1: Number of users plotted against Web rank of academic SNS. Data collected 23 November 2012.
Social network analysis (SNA) is well suited to the research questions as it allows new insights to be gained by visualising complex networks (Borgatti, et al., 2009; Edwards, 2010). However, as the research questions are not only concerned with network structure, but also with gaining an understanding of the processes which led to their creation, a mixed methods SNA approach was used (Edwards, 2010). Mixed methods SNA is a research methodology based on traditional SNA, acknowledging the benefits but also limitations in approaching social phenomena by examining network structure without investigating the motivations of individuals or social processes which created the network. As such, this approach combines SNA with other qualitative or quantitative research methods to gain a fuller understanding.
Three of the largest academic SNS were chosen for inclusion in the study; Academia.edu, Mendeley and Zotero (Figure 1). Although the number of users was not available for Zotero, it was included as its Web ranking (28,505) was high enough to suggest it would be part of the popular, highly ranked cluster. ResearchGate.net is also part of the cluster but its terms of service prevented data collection. The site was approached to take part in the study, but declined. At each of the three academic SNS, profiles were selected for inclusion on the basis of being affiliated with the U.K. Open University (OU). This offered the opportunity to explore a similar network across three different sites, and the possibility to conduct follow-up research with participants while keeping the amount of data practical for analysis. The networks sampled from each sit are summarised in Table 1.
Table 1: Summary of data collection from academic SNS. Site Date of data collection Number of OU academics Number of first degree contacts Total nodes Total edges Graph type Academia.edu 8 February 2013–15 February 2013 1,045 10,283 11,328 22,342 Directed Mendeley 15 November 2012 70 289 359 412 Undirected Zotero 14 December 2012 115 14 129 18 Directed
In addition to collecting information about profiles’ connections (to form the basis of network graphs), data was also collected about academic position and discipline. Disciplines were coded using the Higher Education Statistics Agency (HESA) subject area codes (HESA, 2013), while positions of academics were coded into the following categories: Undergraduate students; Graduate students; Alumni; Tutor or consultant (a catch-all term used by the university for a range of freelance-type posts); Academic support (including librarians, learning technologists and administrators, for example); Researcher (research assistants, postdoctoral researchers, and research fellows); Lecturer (lecturers, associate lecturers and senior lecturers); and Professor (professors and readers). Data about nodes (academics) and edges (links between academics) were entered into Gephi, in order to visualise the network graphs and conduct basic SNA.
A survey sought to explore the reasons behind the trends observed in network structure and gain insight into the network participants’ perspective. A sample was constructed within the OU-affiliated academics at Academia.edu to reflect differences in seniority and discipline (focusing upon academics in the graduate student, lecturer, researcher, professor and Biological Sciences, Computer Science, Social Studies, Historical & Philosophical Studies, and Education categories, respectively). On this basis, 162 academics were invited to take part in the survey. The sample was drawn from the Academia.edu population as it is the largest network sampled, and sampling within this alone rather than the combined Mendeley and Academia.edu datasets removed the possibility of duplication. In answering Part 2 of the survey, participants were asked to focus upon their use of Academia.edu in particular, as it is the only site common to all included in the sample. The survey comprised two parts: demographic information, and questions relating to their use of academic SNS.
Part 1 collected brief demographic information, including: department; position; SNS they have a profile on and level of use. This provided sub-groups to draw comparisons between responses in the second section. The second part addressed how participants view their participation in the network and potential benefits. An inventory of Likert scale items were included (Bell, 2005; Likert, 1932), with statements drawing upon the existing literature (Table 2). A five-point scale was used, ranging from ‘strongly disagree’ to ‘strongly agree’, allowing a neutral (‘neither agree nor disagree’) mid-point.
Table 2: Inventory of Likert scale items, and the theme each draws upon. Item Rationale & basis Being part of an academic social networking site is useful Initial warm-up question I see my profile as an online business card How academics view the role of profiles — Bukvova, 2012; Veletsianos and Kimmons, 2013 I use my profile as a research journal My online academic and personal identities are separated I actively interact with others via the site I only follow people who I know personally Exploring trends in network structure identified in network analysis I follow people as a way of staying in touch with people I used to work with I follow people who I would like to work with in the future If someone follows me, I follow them back Being able to ask questions of the online community is important Following up on differences in terms of question posing and answering identified in previous studies — Almousa, 2011; Menendez, et al., 2012 Academic social networking sites allow me to draw upon a wider community of expertise when I need help Academic social networking sites are a useful way to support working in collaboration with other researchers Collaborative aspects of academic social networking — Jeng, et al., 2012; Oh and Jeng, 2011 Having a profile will enhance my future career prospects Academic social networking sites are a good way of promoting my own academic publications Exploring importance of academic publications as part of an online academic identity Academic social networking sites are a good way of finding out about new publications of interest
Given the limited existing literature, a free text box was also included to allow participants to raise any benefits of using academic SNS which were not covered by the Likert scale items.
The survey was conducted online, using Bristol Online Surveys , from 23 April to 10 May 2013. 51 responses were returned, a response rate of 31 percent from the 162 academics invited to take part.
The degree distribution of the OU-affiliated academics in the three network samples is shown in Figure 2; the networks demonstrate heavy tailed degree distributions, a hallmark of social networks in a range of contexts (Barabási, 2002).
Figure 2: Degree distribution of the sampled academic SNS; Academia.edu (triangles), Mendeley (circles), and Zotero (squares).
The Zotero data was challenging to code as it is the least demanding in terms of information required from users, and although users can make connections, very few in the sample did; the sample included 129 nodes and only 18 edges. As a result, the analysis of network structure focused upon the Academia.edu and Mendeley networks.
The Mendeley graph (Figure 3) shows both the OU-affiliated academics and their first-degree contacts. Since the network is relatively small, it was feasible to also map and code the first-degree contacts. This approach had been intended with the Academia.edu network but the number of first-degree contacts in this case was over 10,000. The Academia.edu (Figure 4) graph shows OU-affiliated academics only.
Figure 3: Network of OU-affiliated academics and their first-degree contacts sampled from Mendeley. Nodes colour-coded according to discipline.
Figure 4: Network of OU-affiliated academics sampled from Academia.edu. Nodes colour-coded according to discipline.
Both networks show structures characteristic of social networks more generally. The basis of the “heavy tailed” degree distribution can be seen in Figures 3 and 4; while a small proportion of nodes are highly connected, many have low or even zero degree. In visualising the networks this way, two further properties of the structures become apparent. Like many other social networks, each includes a “giant component”. Communities can also be seen, with discipline appearing to play a role in defining community structure within the networks.
Moving from communities to the behaviour of individual nodes within the networks, differences were found according to both discipline and seniority. Two basic metrics were considered: degree (the number of connections a node has) and centrality (a measure of how central a node is within the network; the centrality measure used here was eigenvector centrality, which is a measure of influence a node has within a network) (Wasserman and Faust, 1994). As degree distributions and centrality measures are not typically normally distributed, nonparametric statistical tests (independent samples Kruskal-Wallis tests) were carried out to determine whether there are significant differences according to categories of discipline or seniority. The results of the tests are summarised in Table 3. All tests were significant at the 95 percent level, indicating that there are significant differences in degree and eigenvector centrality according to discipline and position, on both platforms.
Table 3: Results of nonparametric statistical tests. Site and factor Degree Eigenvector centrality Academia.edu, by discipline χ2(14, N=1045)=61.41, p=.000 χ2(14, N=1045)=95.70, p=.000 Mendeley, by discipline χ2(8, N=70)=19.58, p=.012 χ2(8, N=70)=24.57, p=.002 Academia.edu, by position χ2(8, N=1045)=244.89, p=.000 χ2(8, N=1045)=328.15, p=.000 Mendeley, by position χ2(6, N=70)=13.15, p=.041 χ2(6, N=70)=16.70, p=.010
While degree and eigenvector centrality were found to differ significantly according to discipline at both SNS, no consistent trends were discernible in terms of particular disciplines across the sites. In considering seniority, however, a similar trend emerges in both networks. Senior academics on average have a larger number of connections and occupy a more central position in the network, while junior academics are less well connected and more peripheral in the network.
This raises questions about how the connections are made; are more senior academics more highly connected because they more actively seek out colleagues to connect with, or do they passively attract more connections due to reputation? As Academia.edu is a directed network (that is, the connections are not mutual but can be viewed as those an individual has chosen to follow, and others who have chosen to follow them), this can be explored in greater detail. The average in-degree (number of followers) and out-degree (number of people they have chosen to follow) according to seniority is shown in Figure 5. This shows an increase in both in-degree and out-degree with seniority, but a disparity emerges between in-degree and out-degree at the most senior levels.
Figure 5: Average in-degree (black bars) and out-degree (white bars) according to position (Academia.edu only).
The survey sought to explore the differences observed in network structure according to discipline and seniority in further depth. In most of the Likert scale questions (in relation to views about communication, collaboration and the role of profiles), consistent opinions emerged irrespective of discipline or seniority. Notable differences were found for questions relating to the nature of connections and network structure. The survey results are addressed here according to themes from the existing literature and the questions which relate to them (Table 2).
Communication — posing and answering questions
Two questions were included which related to the facility of posing and answering questions via the Academia.edu site: ‘Being able to ask questions of the online community is important’ and ‘Academic social networking sites allow me to draw upon a wider community of expertise when I need help’. Both questions gave a modal category of ‘neither agree nor disagree’, and neither showed significant differences according to participants’ discipline or position.
The qualitative responses accompanying each question suggest that while participants can see the potential benefits of these functions, they have not actively used them themselves or are being fulfilled by other platforms. For example, in response to ‘Being able to ask questions of the community is important’:
“Not on Academia.edu in my experience. Sometimes useful elsewhere, but I’m more likely to ask individuals.” — Lecturer, Education
“Potentially but have not done so myself” — Lecturer, Social Studies
Responding to ‘Academic social networking sites allow me to draw upon a wider community of expertise’:
“Strongly agree in general, but I don’t use Academia.edu for this.” — Lecturer, Education
“I can see it would, but I haven’t used like that” — Lecturer, Historical & Philosophical Studies
Communication — academic publications
Two questions focused upon the role of academic publications, in terms of both finding out about new publications from others and in promoting participants’ own academic publications. Likert scale responses both showed a skew toward agreement, with a modal category of ‘agree’. A larger proportion of participants agree or strongly agree that Academia.edu is ‘a good way of promoting my own academic publications’ (35 of 48 responses) than being ‘a good way of finding out about new publications of interest’ (26 of 49 responses). Qualitative responses suggest that the lower agreement with the statement about finding out about the publications of others is due to participants seeing the potential but not having used it this way, and the limitations of the platform. For example:
“Unfortunately, the news feed is useless — I only want to know what my contacts have published, not who they are following.” — Researcher, Social Studies
For some, dissemination is a major function of their profile:
“It makes publications open access to researchers who may not be affiliated with a university and makes the process more streamline. Meaning they don’t have to go through ORO [the University’s online repository] to get copy or a final proof copy of something that has a copyright restrictions. All academic publications should be open access.” — Lecturer, Education
“I get vanity feedback I can use to justify my research impact.” — Lecturer, Computer Science
While there is higher agreement with the statement ‘Academic social networking sites are a good way of promoting my own academic publications’, there is significant variation within the responses according to position (independent samples Kruskal-Wallis test, χ2(3, N=48)=8.014, p=.046). Graduate students, researchers and lecturers ‘agree’ with the statement on average, however the average response by professors is to ‘neither agree nor disagree’ (Figure 6).
Figure 6: Distribution of responses to ‘Academic social networking sites are a good way of promoting my own academic publications’ according to participants’ position. 1 = ‘strongly disagree’, 5 = ‘strongly agree’.
Responses to the statement ‘Academic social networking sites are a good way of promoting my own academic publications’ also exhibit significant differences according to frequency of use of the site (independent samples Kruskal-Wallis test, χ2(4, N=48)=9.973, p=.041). All categories of activity level show a consistent median category of ‘agree’, with the exception of those who “created a profile but have not used it since” demonstrate a median category of ‘neither agree nor disagree’. There is a relationship here between position and frequency of use of the site, as cross tabulation of position and frequency of use of the site (Table 4) reveals that four of the five academics in the ‘professor’ category have ‘created a profile but have not used it since’.
Table 4: Cross tabulation of frequency of use of the Academia.edu site according to position. Position Most days Most weeks Monthly Rarely (less than once a month) I created a profile but have not used it since Total Graduate student 0 1 3 1 0 5 Researcher 2 1 5 1 0 9 Lecturer 2 11 11 4 3 31 Professor 0 0 0 1 4 5 Total 4 134 19 7 7 50
Collaboration — present and future
Two questions addressed using Academia.edu to support research collaboration, in terms of current practice and improving prospects for their future career. Likert scale responses to both questions gave a modal category of ‘neither agree nor disagree’, and no significant differences were found in responses to either question in terms of position or discipline. In relation to the statement ‘Academic social networking sites are a useful way to support working in collaboration with other researchers’, qualitative responses suggest that it is useful but a function better served by other platforms:
“Same as before — strongly agree in general, but not for Academia.edu” — Lecturer, Education
“not always the time invested is actually worth it — and impact or collaboration opportunities using other means can be better.” — Researcher, Computer Science
Identity — The role of profiles
The majority of participants indicated that they viewed being part of an academic social networking site to be broadly useful; 36 of the 50 respondents either agree or strongly agree with the statement ‘Being part of an academic social networking site is useful’. There was also agreement that academic and personal identities are separated online (28 of the 49 respondents indicating ‘agree’ or ‘strongly agree’). No significant differences according to discipline or position were found. Facebook was raised in some of the qualitative responses in contrast to the academic audience on Academia.edu:
“I have received friend requests through Facebook from people that I have met at a conference. I have always refused these as I want to keep my professional and personal profile separate.” — Graduate student, Historical and Philosophical Studies
“I’m a heavy Facebook user and I try to use it to generate larger-than-academic dialogue about some of the things I’m interested in. it’s going beyond what I take to be the narrow confines of academia that attracts me to FB.” — Lecturer, Social Studies
Two questions provided contrasting ways of conceptualising the function played by an Academia.edu profile, either as an online business card (Figure 7) or as a research journal (Figure 8). The majority of responses indicated agreement with the business card metaphor, and disagreement with characterisation as a research journal. This is also reflected by the responses to the item ‘I actively interact with others via the site’, which shows a modal category of ‘neither agree nor disagree’ and a skew toward disagreement overall. The business card role may be useful in reinforcing face-to-face interactions, for example:
“They’re great enhancements to conferences, and have more or less entirely replaced swapping business cards. It’s much easier to meet someone interesting at a conference and then keep in touch via social networking afterwards.” — Lecturer, Education.
Figure 7: Distribution of responses to ‘I see my profile as an online business card’.
Figure 8: Distribution of responses to ‘I use my profile as a research journal’.
Identity — Exploring trends in network structure
The questions related to this theme were included in order to explore the differences in degree which emerged from the network analysis in terms of academic position. Four questions were included which focus upon reasons for following and making connections with others in the network:
- ‘I only follow people who I know personally’ (Figure 9)
- ‘I follow people as a way of staying in touch with people I used to work with’ (Figure 10)
- ‘I follow people who I would like to work with in the future’ (Figure 11)
- ‘If someone follows me I follow them back’ (Figure 12)
Figure 9: Distribution of responses to ‘I only follow people who I know personally’.
Figure 10: Distribution of responses to ‘I follow people as a way of staying in touch with people I used to work with’.
Figure 11: Distribution of responses to ‘I follow people who I would like to work with in the future’.
Figure 12: Distribution of responses to ‘If someone follows me I follow them back’.
The distribution of responses to the items provide an insight into the network structures observed. One item, ‘I only follow people who I know personally’, demonstrates statistically significant differences according to discipline (Independent samples Kruskal-Wallis test, χ2(3, N=50)=11.899, p=.018). The distributions of responses according to discipline are shown in Figure 13; Computer Scientists tend to agree with the statement, Biologists are ambivalent, and Social Scientists and Humanities (Social Studies, Education, Historical and Philosophical Studies) disagree.
Figure 13: Distribution of responses to ‘I only follow people who I know personally’ according to participants’ discipline. 1 = ‘strongly disagree’, 5 = ‘strongly agree’.
The item ‘I follow people as a way of staying in touch with people I used to work with’ (Figure 14) has a bimodal distribution, which reflects significant differences according to position (Independent samples Kruskal-Wallis test, χ2(3, N=48)=10.563, p=.014). Graduate students and researchers agree to a greater extent than lecturers or professors. This item also shows significant differences according to academics’ frequency of use of the site (independent samples Kruskal-Wallis test, χ2(4, N=48)=10.659, p=.031), reflecting the link between position and frequency of use (Table 4).
Figure 14: Distribution of responses to ‘I follow people as a way of staying in touch with people I used to work with’ according to participants’ discipline. 1 = ‘strongly disagree’, 5 = ‘strongly agree’.
Looking from the past to those who participants may wish to work with in the future, a similar trend is borne out. The responses overall are more consistent, although there are still significant differences according to position underlying this (Independent samples Kruskal-Wallis test, χ2(3, N=49)=10.688, p=.014). Again, the median average for graduate students and researchers is ‘agree’, and lecturers and professors ‘neither agree nor disagree’ (Figure 15). This item also demonstrates significant differences according to frequency of use, being linked to academic seniority (independent samples Kruskal-Wallis test, χ2(4, N=49)=13.542, p=.009).
Figure 15: Distribution of responses to ‘I follow people who I would like to work with in the future’ according to participants’ discipline. 1 = ‘strongly disagree’, 5 = ‘strongly agree’.
The final question in this section focused upon reciprocity in forming connections; ‘If someone follows me, I follow them back’ (Figure 12). A small proportion agreed with the statement, although the majority either disagreed or were ambivalent. No significant differences were found according to position or discipline.
Discussion and conclusions
This study has provided an insight into the nature of connections and network structure in academic SNS. By examining the OU population across three different academic social networking platforms, trends in network structure were identified. The network graphs indicate relationships between both discipline and academic seniority in network structure; discipline appears to play a role in defining communities within the network structure, while the number of connections and position within the network differs according to academic seniority (more senior academics having more connections and occupying a more central position within the network than more junior ones). In examining network structure in academic SNS, this study provides a novel contribution as network structure has not been previously studied in this context.
Studying the network across three different platforms, which bears out the trends identified across them, adds a degree of confidence in the results. While the main analysis focused upon the Academia.edu and Mendeley platforms, the greater popularity of Zotero with students, given that students on the other platforms are the least connected to the network, may account for the lack of connections on the Zotero platform to an extent. However, the study is limited in that it only focuses upon a single higher education institution; it is possible that the trends may be unique to the OU, so a potentially useful future study would be to examine network structure across a sample of contrasting institutions to see whether the trends persist.
The survey findings provide an insight into the observed network structures to an extent. Significant differences were found in terms of how academics at different career levels view making connections. The professor category was found to behave differently to the other groups, being more likely to only follow people who they know personally and less likely to make connections for other reasons. This is interesting and helps to explain the differences in degree to an extent, although further work would be required to gain a better level of understanding. Members of the professor category also showed the lowest frequency of use of the Academia.edu Web site, which was also found to be a significant factor in several of the items showing differences according to seniority. This is also a potentially interesting finding as it underlines the discrepancy between network structure and active use; professors tend to occupy privileged positions in the network despite rarely actively participating.
Agreement with the statement ‘I only follow people who I know personally’ also showed significant differences according to subject area. The median category for Computer Science is ‘agree’; for Biological Sciences, it is ‘neither agree nor disagree’; while for Education, Historical and Philosophical Studies, and Social Studies, it is ‘disagree’. Considering the giant component of the Academia.edu network graph (Figure 16), it is possible that there may be a relationship between these attitudes and how discrete or dispersed the disciplinary communities are, however further work would be required to confirm this.
Figure 16: Giant component of the OU-affiliated academics’ network sampled from Academia.edu.
The Likert scale items that were concerned with how academics view the role of Academia.edu and what it is useful for demonstrated broad agreement across the sample and no significant differences were found according to discipline or position. Agreement that a profile serves as an online business card and disagreement in using it as a research journal confirm findings by Bukvova (2012). No disciplinary differences were found in the items relating to questions, in contrast to findings from Almousa (2011) and Menendez, et al. (2012); however it should be noted that the participants in the sample here had not actively used the facility, but rather could see its potential benefits. Similarly, participants could see the potential for the sites’ use in supporting collaboration, but tended not to have actively used it in this way. This may mirror the finding from Oh and Jeng (2011) that only a small proportion of Mendeley users collaborate via group membership.
A further limitation of the study is that academic SNS each represent only a single channel through which academics communicate, collaborate and construct their identity. The network of connections an academic has on Academia.edu for example is but one of their academic networks; very different networks may result if looking at the connections in the variety of other social platforms or face-to-face interactions. This is a limitation of both the network analysis and the survey, which was sampled from the OU-affiliated academics present in the Academia.edu network. As Academia.edu presents a single facet, participants may be highly connected and active online more generally despite being low-level users of Academia.edu, although the survey would have obscured this as it focused upon their use of Academia.edu alone.
Uncovering differences in network structure according to discipline and position points to a relationship with academic career trajectory and identity. This finding contradicts the perception that the online environment acts as a democratising space, suggesting instead a preservation of ‘off-line’ hierarchy. This may be related to the design of the Academia.edu platform, which privileges institutional affiliation and seniority in the way which profiles are presented and navigated. Further follow-up work is planned in order to investigate academics’ networks of connections more holistically, not restricted to a particular platform but rather across the full spectrum of online channels they use.
About the author
Katy Jordan is a Ph.D. student in the Institute of Educational Technology at the Open University, U.K. Her doctoral research focuses upon academic social networking online and the construction of academic identity mediated by online tools.
E–mail: katy [dot] jordan [at] open [dot] ac [dot] uk
This work was carried out as part of a doctoral studentship from the Centre for Research in Education and Educational Technology at the Open University, U.K. The author would like to thank her supervisors, Professor Martin Weller and Dr. Canan Blake, and all the OU staff and students who participated in the study.
1. Bukvova, 2012, p. 351.
Omar Almousa, 2011. “Users’ classification and usage-pattern identification in academic social networks,” IEEE Jordan Conference on Applied Electrical Engineering and Computing Technologies (AEECT), pp. 1–6.
doi: http://dx.doi.org/10.1109/AEECT.2011.6132525, accessed 22 October 2014.
Albert-László Barabási, 2002. Linked: The new science of networks. Cambridge, Mass.: Perseus.
Judith Bell, 2005. Doing your research project: A guide for first-time researchers in education, health and social science. Fourth edition. Maidenhead: Open University Press.
Stephen P. Borgatti, Ajay Mehra, Daniel J. Brass and Giuseppe Labianca, 2009. “Network analysis in the social sciences,” Science, volume 323, number 5916, pp. 892–895.
doi: http://dx.doi.org/10.1126/science.1165821, accessed 22 October 2014.
danah boyd and Nicole B. Ellison, 2007. “Social network sites: Definition, history and scholarship,” Journal of Computer-Mediated Communication, volume 13, number 1, pp. 210–230.
doi: http://dx.doi.org/10.1111/j.1083-6101.2007.00393.x, accessed 22 October 2014.
Helena Bukvova, 2012. “A holistic approach to the analysis of online profiles,” Internet Research, volume 22, number 3, pp. 340–360.
doi: http://doi.acm.org/10.1145/2500098.2500111, accessed 22 October 2014.
Angelika Cosima Bullinger, Stefan H. Hallerstede, Uta Renken, Jens-Hendrik Soeldner and Kathrin M. Moeslein, 2010. “Towards research collaboration — A taxonomy of social research network sites,” AMCIS 2010 Proceedings, at http://aisel.aisnet.org/amcis2010/92/, accessed 22 October 2014.
CrunchBase, 2012. “Academia.edu company page,” at http://www.crunchbase.com/company/academia-edu, accessed 23 November 2012.
Gemma Edwards, 2010. “Mixed-method approaches to social network analysis,” National Centre for Research Methods, Discussion Paper, at http://eprints.ncrm.ac.uk/842/, accessed 24 October 2013.
Rip Empson, “Armed with new funding & a global mission, ResearchGate adds PayPal co-founder to board,” TechCrunch (1 March), at http://techcrunch.com/2012/03/01/armed-with-new-funding-a-global-mission-researchgate-adds-paypal-co-founder-to-board/, accessed 23 November 2012.
Christine Greenhow, 2009. “Social scholarship: Applying social networking technologies to research practices,” Knowledge Quest, volume 37, number 4, pp. 42–47.
Ben Hammersley, 2010. “With Mendeley, research papers get scrobbled,” Wired (1 February), at http://www.wired.co.uk/magazine/archive/2010/03/start/research-papers-get-scrobbled, accessed 23 November 2012.
Higher Education Statistics Agency (HESA), 2013. “HESA subject codes,” at http://www.hesa.ac.uk/content/view/102/143/1/2/, accessed 8 April 2013.
Sarah Huggett, 2010. “Social networking in academia,” Research Trends (March), at http://www.researchtrends.com/issue16-march-2010/research-trends-8/, accessed 23 November 2012.
Wei Jeng, Daqing He, Jiepu Jiang and Yang Zhang, 2012. “Groups in Mendeley: Owners’ descriptions and group outcomes,” ASIST 2012, at https://www.asis.org/asist2012/proceedings/Submissions/256.pdf, accessed 22 October 2014.
Kim Leeder, 2008. “Social networking with a brain: A critical review of academic sites” (10 December), at http://www.inthelibrarywiththeleadpipe.org/2008/social-networking-with-a-brain-a-critical-review-of-academic-sites/, accessed 23 November 2012.
Rensis Likert, 1932. “A technique for the measurement of attitudes,” Archives of Psychology, number 140.
Maria Menendez, Antonella de Angeli and Zeno Menestrina, 2012. “Exploring the virtual space of academia,” In: Julie Dugdale,Cédric Masclet, Maria Antonietta Grasso, Jean-François Boujot and Parina Hassanaly (editors). From research to practice in the design of cooperative systems: Results and open challenges: Proceedings of the 10th International Conference on the Design of Cooperative Systems, 30 May–1 June 2012, pp. 49–63.
Michael Nentwich and René König, 2012. Cyberscience 2.0: Research in the age of digital social networks. Frankfurt: Campus Verlag.
Jung Sun Oh and Wei Jeng, 2011. “Groups in academic social networking services: An exploration of their potential as a platform for multi-disciplinary collaboration,” 2011 IEEE Third International Conference on Privacy, Security, Risk and Trust (PASSAT) and 2011 IEEE Third Inernational Conference on Social Computing (SocialCom), pp. 245–248.
doi: http://dx.doi.org/10.1109/PASSAT/SocialCom.2011.202, accessed 22 October 2014.
Lee Rainie and Barry Wellman, 2012. Networked: The new social operating system. Cambridge, Mass.: MIT Press.
George Veletsianos and Royce Kimmons, 2013. “Scholars and faculty members’ lived experiences in online social networks,” Internet and Higher Education, volume 16, pp. 43–50.
doi: http://dx.doi.org/10.1016/j.iheduc.2012.01.004, accessed 22 October 2014.
Stanley Wasserman and Katherine Faust, 1994. Social network analysis: Methods and applications. Cambridge: Cambridge University Press.
Arlene Weintraub, 2012. “Social networks attempt to spark academic-university collaborations,” Nature Biotechnology, volume 30, number 10, pp. 901–903.
doi: http://dx.doi.org/10.1038/nbt1012-901, accessed 22 October 2014.
Xian-Ming Xu, Justin Zhan and Hai-tao Zhu, 2008. “Using social networks to organise researcher community,” In: Christopher C. Yang, Hsinchun Chen, Michael Chau, Kuiyu Chang, Sheau-Dong Lang, Patrick S. Chen, Raymond Hsieh, Daniel Zeng, Fei-Yue Wang, Kathleen Carley, Wenji Mao and Justin Zhan (editors). Intelligence and security informatics: IEEE ISI 2008 International Workshops: PAISI, PACCF, and SOCO 2008, Taipei, Taiwan, June 17, 2008. Proceedings. Lecture Notes in Computer Science, volume 5075. Berlin: Springer-Verlag, pp. 421–427.
doi: http://dx.doi.org/10.1007/978-3-540-69304-8_43, accessed 22 October 2014.
Holt Zaugg, Richard E. West, Isaku Tateishi and Daniel L. Randall, 2010. “Mendeley: Creating communities of scholarly inquiry through research collaboration,” TechTrends, volume 55, number 1, pp. 32–36.
doi: http://dx.doi.org/10.1007/s11528-011-0467-y, accessed 22 October 2014.
Received 21 November 2013; accepted 12 October 2014.
“Academics and their online networks: Exploring the role of academic social networking sites” by Katy Jordan is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Academics and their online networks: Exploring the role of academic social networking sites
by Katy Jordan.
First Monday, Volume 19, Number 11 - 3 November 2014