Keywords

1 Introduction

The use of social media platforms in the education field are on the rise [1], encouraging scholars to participate in online professional communities to enhance learning [2]. Twitter is used among academics, at all educational levels, as a teaching, learning, and professional development tool [3]. Students, teachers, professors, and other professionals use Twitter as a pedagogical tool for enhancing learning environments that promote engagement among users [1, 4]. It also provides participants with opportunities to collaborate, gather evidence, and reflect on their practice, with other professors and professionals outside their institutions and fields [5].

About 1 in 40 scholars are using Twitter for scholarly chat and self-promotion, but also for community building [6]. Educator-driven professional communities on Twitter, such as #AcademicTwitter, #PhDchat, #AcWri, and #AcademicChatter are gaining popularity among academics and students. Particularly, the hashtag #AcademicTwitter has emerged in the past years as a prominent indexing tool where thousands of tweets are sent every month. The hashtag #AcademicTwitter is used to share information, provide support, and engage in conversations regarding the world of academia. Despite the popularity of these social spaces, there is a lack of understanding of how users interact with one another [7]. And while studies have addressed the importance of professional communities and relationships for educators [8, 9], it is yet unclear how online professional communities shape these connections and relationships between educators [10].

Our study contributes to the online professional communities’ literature, by gaining insights into the patterns of interactions in the #AcademicTwitter hashtag. The aim of this investigation is to better understand the patterns of interactions of those using the #AcademicTwitter hashtag professional community. The following research questions guided this investigation:

  • RQ1: What are the demographics of the #AcademicTwitter users (i.e., gender, role, and field)?

  • RQ2: What sentiments were expressed in the tweets tag with the #AcademicTwitter hashtag?

  • RQ3: Who are the central participants in the #AcademicTwitter hashtag?

2 Literature Review

There is no denial that social media is now part of our daily lives. There is a range of different social media platforms (i.e., Facebook, Instagram, Snapchat, YouTube, LinkedIn, TikTok) and users are often actively participating in more than one particular social media outlet. One of these social media platforms is Twitter. According to the Pew Research Center [11], Twitter is used by 22% of adults in the United States. In addition to linking with accounts that they wish to follow; Twitter users tend to connect using hashtags. In Twitter, individuals use hashtags to reach a broader audience of individuals who share a similar domain of interest. This is particularly the case of users who seek to connect with a professional network of individuals. According to Romero-Hall [12], professional growth via social media is generated through the social sharing and refining of ideas in a network or community with a common domain.

Various researchers have explored the use of hashtags by different professional communities in Twitter to better understand the social nature of interactions, types of users, and content shared [13,14,15,16]. For example, Greenhalgh and Koehler [13] examined targeted and timely professional development after the terrorist attacks in Paris in November 2015 by analyzing tweets with the hashtag #educattentats. This hashtag served as a temporary affinity space to provide support for teachers preparing to address the incident with their students. However, unlike other hashtags use for professional communities in Twitter, this was a temporary space that was only used for 28 days. Rashid, McKechnnie and Gill [14] investigated advice that is given to newly qualified doctors as they start their career via the hashtag #TipsForNewDocs. The results showed that most tweets focused on professional development as well as knowledge sharing, of both tacit and know-how knowledge. There were also humorous tweets related to socialization. Gomez and Waters [17] explored a Twitter hashtag created by professionals in the Public Relations fields that served as a point of connection among educators, practitioners, students, and various other organizations. An analysis of the network properties and actor roles of the hashtag #PRProfs showed that conversations in this Twitter community are predominantly about sharing knowledge, teaching tips, and trends in the PR industry [17].

Another example of the exploration of hashtags to better understand a Twitter professional community is the research conducted by Kimmons and Veletsianos [18] in which the researchers collected tweets related the American of Educational Research Association (AERA) Annual Meeting to better understand academic Twitter use during, around, and between the annual conference both as a backchannel and general means of participation. Tweets with the hashtags #AERA14 and #AERA15 were collected and analyzed. The results served to compare participation patterns between two years of the conferences. One major finding by Kimmons and Veletsianos [18] was the difference in participation norms by students and professors.

There have been several investigations focused specifically on the use of the #edchat hashtag and its users. For example, Coleman, Rice, and Wright [19] collected post and survey teachers who used the #edchat hashtag to determine if the exchanges between teachers served as continuing education and merited credit. The results indicated that conversations between teachers using the #edchat hashtag were found to generate social capital and bind a professional community. Staudt Willet [15] also conducted an investigation focused on the #edchat hashtag. This author explored the types of tweets that users contributed to #edchat and the purposes observable in the tweets. The results indicated that based on the analysis of the #edchat tweets, posts were mostly on topic related to education and the practice of teaching. Yet, Staudt Willet [15] added that teachers were not using #edchat to its full potential, as tweets were missing important emotional elements that tend to shape relationships and too many times the hashtag was use more for self-promotion.

Researchers have address that the effective use of hashtags is determined by factors other than its affordances and design such as the users’ needs and desire, as well as social, cultural, economic, and political environments [16]. In an investigation comparing three hashtags (#NutricionMOOC, #EdTechMOOC, and #PhDChat), Veletsianos [16] observed general participation patterns on these hashtags, the types of user who contributed to the hashtags, and the content tags in the tweets with these hashtags. The results of this particular study showed that there are a variety of outcomes on the potential benefits of hashtags pending the contexts. Two of the hashtags in this investigation (#NutricionMOOC and #EdTechMOOC) primarily served as mediums for announcements and promotion rather than professional development and social connection. Although hashtags offer significant opportunities for professional development, teaching, and learning, they may or may not fulfill the need as expected [16].

The review of the literature gives some insights on how different hashtags in Twitter, related to professional communities, have at times served to create networks, ignite conversations, increase knowledge, and foster relationships. However, it is also clear from the literature that at times hashtags for professional online communities serve for shallow networks [16] or may provide temporary connections in a “just-in-time” format [13]. The creation of quality interactions in any kind of setting requires the investment of time, commitment, and the willingness to engage. Socialization, networking, and the creation of connections are complex processes [14].

As stated by Coleman, Rice, and Wright [19]: “social capital brings members of a group together in solidarity and coalesces a group with string bonds within a community.” Given the proper nurturing and attention, professional communities using hashtags in Twitter can provide social capital to those engage. For academics, evidence show that Twitter provides immediacy, reach, and scholarly engagement that is relax, professional, and at time humorous [20, 21]. In addition to the casual chatter, social media used by academics has shown to provide a sense of belonging to a community, cross country interactions, and additional learning resources and research collaborations [22]. Due to the value that Twitter communities provide to academic discourse and socialization, it is important to investigate different elements of this social media network.

3 Methodology

3.1 Data Collection and Cleaning

This study uses quantitative content analysis and social network analysis techniques to examine communication patterns and network properties of the #AcademicTwitter community on Twitter. Netlytic [23], a free cloud-based text and social networks analyzer developed by the Social Media Lab at Ryerson University in Toronto, was used to recollect tweets that have the hashtag #AcademicTwitter. We ran Netlytic software from March 1st to April 1st, 2019, to collect and analyze tweets. The raw dataset has 26,287 unique tweets and 15700 unique users that participated in the discussions.

3.2 Variables

For this investigation, we selected a random sample of 500 users to manually analyze their Twitter bio profiles, following three variables:

  1. 1.

    Gender (individual, organization, or other). “Other” refers to a user who did not provide a clear name, bio, or a photo that makes it harder to categorize as individual or organization.

  2. 2.

    Field: STEM, Social Sciences, Arts & Humanities, Business, Education, Professional Studies, and others.

  3. 3.

    Actor role: Originally 16 categories were identified: 1. Assistant Professor, 2. Associate Professor, 3. Professor, 4. Other professors, 5. Researchers, 6. Teachers, 7. Graduate students, 8. Undergraduate students, 9. Professionals in education, 10. Other professionals, 11. Individuals interested in education, 12. Universities/colleges, 13. Nonprofits, 14. Journalists, 15. Media, 16. Others (do not disclose their profession or job role; usually disclose hobbies or random quotes). After collecting and examining the results, we merged some categories as shown in the findings section.

The categorization was derived inductively, guided by grouping Twitter users by their profession, position, and field. The actor role categories are mutually exclusive. We selected the first role each participant disclosed in the bio for codification. Sometimes users can indicate multiple roles, but researchers always coded the first role or job they disclose. Two coders coded jointly the profiles and discussed to solve coding discrepancies to assure reliability. Sentimental and textual analysis of 26,287 unique tweets were performed as well.

3.3 Social Network Analysis (SNA)

The name network (who mentions whom) was considered for analysis, which revealed 15,052 directional ties among 5560 nodes (posters with ties). We used two centrality measures: in-degree and out-degree. In-degree centrality indicates the number of ties (e.g., messages) a node (user) receives from others. A high in-degree centrality shows the popularity of a user, which is actively mentioned by others. In contrast, a high out-degree indicates an active participant who has the purpose to disseminate information to the network. We used Netlytic software to calculate macro level measurements such as density, diameter, reciprocity, centralization, and modularity to reveal network structure.

4 Findings and Discussion

Researchers were interested in learning about the communication and network structure of the #AcademicTwitter community by identifying influential actors. Descriptive results are presented as follows: 82% of users were individuals, 11% organizations, and 7% others (i.e., researchers could not identify a categorization base on the bio). Figure 1 indicates the most recurrent users, showing the involvement of professionals (professionals working in education or professionals interested in education), graduate students, educators (Assistant, Associate, Professors, Lecturers), and researchers. Most of the actors belonged to the STEM (31%), Social Sciences (23%), and Arts and Humanities (13%) fields. Fourteen percent of the users did not disclose their field.

Table 1. Central participants by different centrality measures.
Full size table
Fig. 1.
figure 1

Most predominant roles of the participants in #AcademicTwitter.

Full size image

We found that most of the tweets were positive. A total of 4240 posts addressed positive feelings such as great (1256), good (805), love (599), excited (384), and happy (360). Only 584 tweets were negative, conveying feelings such bad (162), lonely (44), tired (44), dull (38), and nervous (37). Figure 2 shows the salient themes in the #AcademicTwitter community indicating discussion topics concerned to graduate students’ life and overall success of professionals in academia. We also discovered a call-to-action language that dominated the conversations such as give, make, check, learn, and find.

Fig. 2.
figure 2

Salient themes on #AcademicTwitter.

Full size image

The #AcademicTwitter community consisted of six main clusters with central actors that influenced the way information traveled through the network, as shown in Fig. 3. Clusters are a group of connected people which tend to communicate frequently with others in the group and typically do not communicate with users outside of the cluster. The first cluster in the #AcademicTwitter community is influenced by a media platform-social education content provider: @academicchatter (total degree: 447), the second cluster is a media platform, The Chronicle of Higher Education: @chronicle (total degree: 78), the third cluster, @ph_d_epression, another media platform, which at the time of analysis-not during data recollection-, was inactive (total degree: 167), the fourth cluster, another media platform: @humanbiojournal (total degree: 46), the fifth cluster was a graduate student: @hannahlebovits (total degree: 85), and the sixth cluster, the University of Guelph in Canada: @uofg (total degree: 149). All 6 clusters are shown in Figure; isolates are also visualized. Each cluster has a different color. Findings reveal that conversations were not centered on one specific and solid community, instead several micro communities were found.

Fig. 3.
figure 3

Main clusters in the #AcademicTwitter community.

Full size image

Table 1 indicates the top central participants by in-degree and out-degree centrality. In-degree users are tagged in posts and tend to be popular. In this example, the Twitter user @JuliaFtacek is tagged in the message with the @symbol: @JuliaFtacek I personally love #AcademicTwitter. I particularly enjoy connecting with other academics inside and outside my field. I’ve had many a good discussion about methodology for research and pedagogical methods for my classroom). Out-degree users post frequently (either tweeting, retweeting or tagging users) and show good awareness of others. For instance, RT @zra_research: Staying organized and time management are half the battle in pushing research projects forward. This is a great example. An interesting finding in the results is the active presence of @AcademicChatter as both an in-degree and out-degree user as shown in Table 1. @AcademicChatter is a social media education content provider for graduate students and academics.

The most recurrent posters (out-degree) were media outlets (e.g. @AcademicChatter and @ThePhdStory), other professionals (@HigherEDPR, user is a communication strategist for faculty and researchers), and professors in STEM disciplines (e.g. @Carlymdunn_mph). In-degree users (users mentioned in tweets) were mainly media outlets (@Phdforum, @AcademicChatter) and educational organizations such as @TutorsIndia. It is also worth noting the minimal presence of Twitter accounts of universities and colleges (only 15 profile users were found) in the conversations.

During the one-month period analyzed (March 1–April 1st, 2019), there was a peak of tweets sent during March 21st as illustrated in Fig. 4. This was because several out-degree users retweeted this tweet by RT @elizabethsiber: “#AcademicTwitter - did you know that there’s free, reliable software that will close-caption your powerpoint presentation?” Academics like to learn about free resources to improve our work, especially during times of creating material that are more accessible for students. Figure 4 includes the number of posts over time ranging from as low as 480 tweets per day to more than 1,500 tweets per day in the #AcademicTwitter community. An average of 600–700 tweets were posted per day, making #AcademicTwitter an important resource for academics.

Fig. 4.
figure 4

Number of posts over time.

Full size image

The #AcademicTwitter network primarily specializes in information and resource sharing but also users are constantly seeking for advice (e.g. Okay #academictwitter How do I write a paper for a collected volume without sounding like it’s my first ever undergraduate essay? For some reason, I seem to have totally lost my ability to, like, use words that make me sound, like, super smart). The #AcademicTwitter community involves users from a diverse set of professional backgrounds and fields, sharing interesting resources information.

Social network analysis is a useful technique that reveals how information and resources move through the network [24], identifying prominent actors in the network [24, 25]. The social network analysis performed to the #AcademicTwitter community indicated the following measurements: Diameter (44), Density (0.000180), Reciprocity (0.025680), Centralization (0.026280), and Modularity (0.923500). The #AcademicTwitter community is a wide network with a few central participants dominating the information flow, indicating a centralized network. Most users were disseminating or retweeting information, but engagement was not always predominant between network participants. Central actors in networked online communities are opinion leaders who affect others and control the information flow [26], which usually is led by a small group of dominant and engaged actors [27].

The first measurement examined was centralization (0.026280) which indicated that there were few influencers in the network. If values are closer to 1 than 0, indicates that a few central participants dominated the flow of information in the network. In addition, only 40% of the users tweeted more than one time, reafirming a centralized network. Density is another network measurement which examines how close participants are within a network. The #AcademicTwitter network had a density of 0.000180, which indicates that mostly no one was connected to others in the network (values closer to 1 are evidence of a close-knit community). Diameter calculates the longest distance between two actors in a network, which in the #AcademicTwitter community was 44, showing some actors with higher degrees of separation or connection and presenting a wide network. Reciprocity (0.025680) indicates if users engaged in two-way communication. Results showed that only 0.25% of the users participated in conversations, showing a predominant one-way information network. Lastly, modularity, determines if there are several small communities or a one singular community in the network. A higher modularity (more than 0.5) indicates divisions between communities as represented by clusters. The #AcademicTwitter network presented six different clusters and a modularity value of 0.923500, indicating the presence of micro-communities in this network.

Our results align with the work of Bruns and Burgess [28] and Gruzd, Wellman, and Takhteyev [29] which indicate that people use hashtags to discuss topics of shared interest. The #AcademicTwitter users participated (as in-degree or out-degree) by different reasons which can include connecting with other professors and professionals, sharing resources, seeking advice, self-professional branding, or just having a break from work. All actors involved in Twitter professional communities share a mutual aim which is to distribute information that will potentially impact the communications flow [30].

Our #AcademicTwitter study also supports previous studies (Xu et al. 2015) which found that information sharing and building relationships are the most important aspects of Twitter conversations in online communities. #AcademicTwitter has emerged during the past decade, building a community of educators interested in making connections and relationships, providing support, and sharing resources about teaching, research, service, and overall the academic life. However, there is still participation inequality in social media where 90% of social media users are “lurkers” who do not contribute to the communities [31]. Future studies could examine lurkers in professional communities such as #AcademicTwitter, which could help understand the motivations of their passiveness on social media platforms. In the same line, further studies could answer what motivates academics to engage in learning and professional communities and the benefits they receive. Previous studies [5] have found that online professional communities are a source of continuous professional development for academics, providing authentic and personalized opportunities for learning and support.

5 Conclusions

Our study used Twitter to analyze online professional communities related to academia and how participants were using Twitter and specifically #AcademicTwitter to support their professional development and learning. Our paper also examined communication patterns and user influence in the #AcademicTwitter community.

Social media platforms and online professional communities provide great opportunities for learning, guidance, and academic support. People from diverse backgrounds are turning to social interactions (e.g. online chat groups, discussion lists) to satisfy their needs no matter if they are personal or professional [5]. Our paper contributes to the literature of social media and education, providing insights in the user role and influence in the #AcademicTwitter professional online community. #AcademicTwitter is a growing and popular educational-driven community on Twitter that could attract other active users in the discussion (e.g. nonprofits or industries), to provide helpful information and opportunities to connect with scholars.

This study has some limitations that suggest avenues for further work. This is an exploratory study in the use of Twitter among educational communities. Future studies can take a more in-depth analysis to understand specific topics and content shared among academics in online professional communities. The current study does not involve the codification and further examinations of tweets (messages). Having few influential participants and a network characterized by low two-way conversations, does not mean that real collaboration wasn’t involved. In other words, further studies could analyze the collaborative nature of the conversations, and even message purpose (e.g. informative, educative, engagement, mobilization). This work has implications in the education field as it identifies the prominent users in the discussion of educational topics and issues on Twitter. Identifying these influential actors helps to provides opportunities for academics to engage with these prominent users for networking, support and collaborative opportunities.