Social networking has accelerated changes in all fields of endeavour. Emergent professional social networks (PSNs) aim to aggregate scholars’ research work and to enhance collaboration. Professional social networks give scholars an opportunity to collaborate and disseminate research outputs quickly and avoid the delays that come with traditional publishing systems (Swanepoel & Scott 2018). Traditional publishing systems build strong connections with scholars to facilitate research production but have difficulty connecting with a larger audience (Beall 2013). Aggregators such as Academia.edu, ResearchGate and Google Scholar measure research outputs and reach a larger audience.

Professional social networks become more important as they change the way scholars teach, conduct research, share ideas, collaborate and disseminate research output (Nitza & Roman 2016). The problems associated with the adoption and use of PSNs and the underlying patterns between the benefits and challenges are often understudied. Little is known about how professional social networking is perceived by scholars in South African universities. Whilst there are benefits associated with adopting different types of PSNs, there are also challenges involved. This article reports on the benefits and challenges of adopting PSNs by South African higher education academics. These objectives were achieved by answering the following questions:

According to Carrigan (2019:16), academic PSNs are online platforms that allow academics and researchers to construct a public profile within a constrained system. With this definition in mind, any social network facilitating the sharing of a connection amongst scholars can be classified as professional. Platforms such as Academia.edu, ResearchGate and LinkedIn, amongst others, are usually classified as PSNs. Scholars use these platforms to communicate, share information and collaborate with other scholars. The concept of PSNs is usually misunderstood by isolating them from generic social networks (GSNs) such as Facebook and Twitter. In this article, professional social networking is not limited to a classification of specific platforms but rather perceived as an enabling feature of any social network platform. In other words, a social network platform that enables a scholar to perform professional duties is perceived and classified as professional. When debating over platforms such as Facebook and Twitter, many people think the platforms cannot be used as professional networks. When it comes to social network sites, the user decides its professionalism based on how it is used.

The concept of a social network was first introduced in 1997 with Six Degree (Boyd & Ellison 2008). By the year 2000, the Internet became popular with chatrooms with a reach of millions of online users. MySpace and LinkedIn were introduced in 2003. In 2006, Facebook and Twitter were introduced to the world. In 2008, ResearchGate and Academia.edu were introduced. From this time forward, the Internet started being bombarded with different social networks to the point that today it is challenging to know how many social network platforms are operational. Whilst Facebook and Twitter are designed to be used by anyone on the Internet, platforms like Academia.edu, ResearchGate are limited to specific targeted groups.

Professional social networks usually differ in the ways in which they function, but the construction of profiles is usually made up of the same information, such as first names, surnames, date of birth, gender and location. These details are often required for a user to subscribe. For an academic profile, a user, in addition to the above, is also requested to provide a biographical sketch, a list of publications, research experience and academic skills. In common with most social media networks, users can post content, comment on posted content, share the content (with anyone or just a target group), upload articles, perform one-on-one chats, use group chats and make video calls. These features are important for enhancing research and academic collaboration. Scholars usually subscribe to multiple platforms to cover for a lack of certain features from another platform. There is seldom a single platform that is inclusive of all features to meet the needs of scholars.

Some key drivers for adoption by scholars include awareness (Sheikh 2016), which is driven by the pressure coming from the world around scholars’ institutions (Azambuja & Nikolaeva 2019), self-promotion, acquisition of ideas, belonging to a network of researchers, interacting with peers (Meishar-Tal & Pieterse 2017), increasing visibility (Jeng, He & Jiang 2015), ease of use and connecting with other scholars without the restrictions that come with offline conferences and workshops (Wyse et al. 2016).

Professional social networks allow scholars to enhance their profiles, expose scientific voices to a larger audience, share ideas and publications quickly and access more research content (Yan & Zhang 2018). It can also be beneficial for a collaboration to start online and then convert to an offline collaboration (Althoff, Jindal & Leskovec 2017). Conversations in offline conferences can be tweeted to open the conversation to a larger audience. Professional social networks can assist scholars in disseminating publications (Palmer & Strickland 2017), reach a non-scientific audience and enable access to advanced citation and altimetrics that differ from traditional metrics (Li & Gillet 2013).

With these benefits in mind, some challenges forming barriers to adoption are also raised. These challenges include concerns around user interface (Bhardwaj 2017); the lack of several features which slow adoption and usage (Bhardwaj 2017); missing guidelines; the lack of training (Williams & Woodacre 2016); privacy issues; exposure to exploitation (Hoffmann, Lutz & Meckel 2014); plagiarism; copyright and legal issues (Ashraf & Mohamed 2016); the lack of time; the lack of confidence in computer skills (Osterrieder 2013); the lack of trust, clarity and integration of institutional policy (Coppock & Davis 2013); commercialisation of content; the lack of credibility; and the quality of content (Lupton 2014:3). The lack of awareness, time limitations and concerns related to copyright were raised in a qualitative study conducted by Swanepoel and Scott (2018) in South Africa.

With many tools available online, their approval and acceptance remain limited in the community of scholars. There is a need for comprehensive training which highlights the strengths and weaknesses of each platform to allow scholars to make the right choice. On Academia.edu, scholars can find, follow scholars, post publications, share publications and participate in discussions in a community of scholars (Academia.edu 2022), but they cannot reach a non-scholar community. These features are also provided by ResearchGate with the same limitations (Thelwall & Kousha 2015), except those publications are more open, whereas Academia.edu commercialises publications. On Mendeley, scholars can upload publications and contribute by providing comments on uploaded publications (Elsevier 2022). Just like Mendeley, Zotero facilitates the creation of groups in which scholars can connect, collaborate and access research papers published by other researchers (Chen et al. 2018) and also facilitate the management of bibliographies. Another interesting platform that enables collaboration is Office 365, currently known as Microsoft 365, where scholars can use Microsoft Word, Teams, SharePoint, Yammer and other inclusive features to collaborate on drafts of journal articles, books and conference proceedings, as well as chat and perform video calls (Mercurio 2018).

Scholars can, instead of using their Facebook profile as a research space, create a research page where they can post research-related activities and research outputs. Scholars can also build networks with their colleagues on Twitter by following those interested in their field. Whilst Twitter is mostly used in South Africa (Statcounter 2020), only a small percentage of South African scholars (6%) are using hashtags to share and discuss new and existing publications (Joubert & Costas 2019). LinkedIn is another interesting platform where scholars can write posts and articles, share videos, check accessibility by other scholars and expand professional networking (LinkedIn 2022).

WhatsApp and Google Scholar, even though not classified as social networks, are often used by scholars. Google is usually used by scholars to access their publications and their citations per year metrics (Martín-Martín et al. 2018). With the use of web crawlers, Google Scholar consolidates publications that have been made available on ResearchGate, Academia.edu, Facebook, Twitter, LinkedIn, Mendeley and Zotero, amongst others. It is necessary that this platform is not excluded from the list of PSNs, as it connects and consolidates professional content from other social network platforms to ensure the content is searchable by anyone on the Internet. WhatsApp, on the other hand, is a messaging application which facilitates social connections through group chats (Gon & Rawekar 2017), where scholars can discuss and collaborate on topics to enhance research collaboration and productivity (Nitza & Roman 2016).

At the University of South Africa, scholars prefer using free search engines instead of fee-based traditional systems for validating and measuring research impact (Adriaanse & Rensleigh 2017), which is then linked to promotions and pay incentives (Dettori, Norvell & Chapman 2019). There has been an increase of predatory publications enabled by weak policies of social networks (Mouton & Valentine 2017). Most social networks used for professional social networking lack filters to eliminate non-reviewed and predatory research publications (Gasparyan et al. 2017). Compromised profiles and unattended profiles can be used by scholars with bad intentions to increase uninspected publications on these platforms.

Scholars are probably consuming information, sharing less and interacting less with others. In the South African context, the research conducted has used qualitative methods, focusing on just a few scholars at a few universities. Such results are difficult to generalise to the population of scholars. This article evaluates the extent to which the adoption of PSNs by researchers at the majority of public South African universities is currently unknown.

This study adopted a quantitative method to test assumptions and then a qualitative method to understand the experiences of scholars. A qualitative approach was used to strengthen the results obtained from the quantitative approach. A sequential explanatory design was implemented where quantitative results preceded qualitative experiences. The data was consolidated from a survey, which was disseminated online to scholars in different public universities in South Africa. The survey focused on scholars’ adoption and usage of PSNs for collaboration and dissemination of research outputs. Likert-scale questions were used to investigate how many participants agreed or disagreed with predefined statements. All Likert-scale responses were factor analysed in the Statistical Package for the Social Sciences (SPSS) to eliminate unreliable data and to identify intercorrelated variables from a large set of variables. The data was validated and relied on Cronbach’s alpha test results, which were all above 0.80 (0.893 for adopters, 0.876 for non-adopters). Participants were scholars located at 17 public universities in South Africa including the University of KwaZulu-Natal, Cape Peninsula University of Technology, Central University of Technology, Durban University of Technology, North-West University, Rhodes University, Stellenbosch University, University of Fort Hare, University of Free State, University of Johannesburg, University of Limpopo, University of South Africa, University of Venda, University of Zululand, Tshwane University of Technology, University of the Western Cape, Vaal University of Technology and the University of the Witwatersrand, where ethical clearances and/or gatekeepers’ letters were obtained. Informed consent was electronically recorded for all 950 responses that were randomly obtained from these institutions.

In addition, after consenting, 10 interviews were conducted with expert scholars in the field to expand the results obtained from the survey. Participants were academic or research scholars based at public universities in South Africa. These were imported into NVivo for qualitative analysis to confirm and support the views in the questionnaire. With this in mind, the question around the number of interviews to rely on is usually raised by scientists and cannot be ignored. Patton (2002) and Marshall et al. (2013) proposed that a small sample size of interviewees is not sufficient but at the same time did not provide evidence for this argument. However, Creswell (2013) proposed a minimum sample size of between 5 and 25, whilst Hagaman and Wutich (2017) argued that 16 or fewer interviews are sufficient. Because there is no specific agreement amongst scientists on determining the sample size in qualitative studies, this research depended on the statement made by Ryan and Bernard (2003), who concluded that it is not about the quantity or the number of participants in qualitative studies but the relevance of the data. The data obtained from interviews was classified into different thematic groups based on perceived benefits (collaboration and dissemination), challenges (plagiarism, commercialisation of content, profiles, privacy), policies and support (missing support and preferred support).

This article focuses on the variables of the diffusion of innovation (DoI) (Rogers 2003; Khan & Khan 2018) and technology, organisation and environment (TOE) (Rosli, Siew & Yeow 2016) models by evaluating causes and effects using the push and pull factors model associated with PSNs’ adoption and usage.

In Figure 1, traditional networking methods are associated with the use of e-mails to enhance collaboration and research activities but also with offline systems of collaboration such as attending offline conferences and workshops. The diffusion and innovation (DoI) theory determined the level of adoption of PSNs and the types of PSNs. Benefits and challenges are also evaluated within this model. For a holistic overview, the DoI is combined with the TOE model. This allowed evaluation of the environment and organisation in which PSNs are adopted and used by scholars. The benefits and challenges identified in Figure 1 can pull scholars toward PSNs or push them away from PSNs. Perceived benefits have more chances to pull scholars toward PSNs whilst perceived challenges discourage and push scholars from adopting PSNs. Perceived challenges and benefits also determine the level of adoption and the types of PSNs being adopted. Furthermore, adopting PSNs does not mean scholars are using them. Scholars can default back to traditional networking systems if they discover that PSNs are not responding to their needs. This article is also not suggesting that traditional methods need to be removed from usage but rather that combining different systems can lead to an increase in research productivity at different public universities in South Africa.

This study used quantitative and qualitative approaches to identify factors linked to the adoption of PSNs at various public universities in South Africa. The findings were based solely on results received from scholars at public universities and excluded private universities, professional associations and colleges. These can be targeted by other investigators to broaden the results and by augmenting the minimum qualitative sample size used in the study. In a South African context, there are more challenges than benefits to the adoption of PSNs. As revealed, scholars at various universities have adopted PSNs and use them to an extent. Whilst the level of PSNs’ adoption is extremely driven by the perceived benefits, usage and continuity shrink as scholars find it challenging to navigate the different types of PSNs. This causes complete discontinuity or having a passive presence. At governance and institutional levels, there is a need for the implementation of policies that will guide scholars to navigate these online environments without compromising on their existing workload. There is a need for management to support the usage of these tools and integrate them with existing systems for effective usage and continuity. Misalignments between institutions’ processes, publishers and PSNs’ owners need to be resolved by all parties for better research productivity.