Background:
Twitter is one of the most popular social media platforms inside and outside academia. Researchers use Twitter to search for and disseminate research, communicate, network, and engage with a broader audience. In the academic context, Twitter is most notably used to improve one’s visibility (Haustein, Bowman, Holmberg, Peters & Larivière, 2014; Holmberg, Bowman, Haustein & Peters, 2014; Van Noorden, 2014). Twitter activity has been receiving increasing attention by academia as a potential indicator of research impact, alongside other social media-based metrics (altmetrics). Several studies suggest that Twitter might reflect public outreach and science communication activities, not measured by traditional bibliometric indicators (Haustein et al., 2014).
Objective:
To further investigate Twitter use in academia as an extension of scholars’ scientific network and as a diffusion channel, this paper looks at the Twitter activity of a specific community: information science schools and their faculty members. More specifically, we aim to answerthe following questions: 1) How much do information science schools and their faculty members use Twitter? 2) How are the members of this community connected on Twitter? 3) What do they tweet about?
Methods:
A list of the 30 North American iSchools was obtained from the website ischools.org. Using the schools’ website, we retrieved the list of all faculty members, and then searched for their Twitter account. We found a Twitter account for the 30 iSchools and 267 (33%) of the 803 faculty members. Using the Twitter API, we retrieve the date of creation, the number of tweets, the tweets content1 and the list of followers for each account found.
Results:
Preliminary results show a large disparity in the Twitter presence and activity of iSchools, with faculty members’ presence per institution ranging from 7% to 84% and activity ranging from 56 tweets/year to 19,067 tweets/year. Figure 1 shows the highly skewed distribution of tweets and followers where, in a Pareto-esque fashion, 20% of Twitter users send more than 80% of the tweets (left) and 20% have more than 80% of the total number of followers (right). 1 Only the last 3,200 tweets for each account are available from the Twitter API.
We applied the Blondel et al. (2008) community detection algorithm to the follower-followee network of scholars and institutions, and found eight distinct communities, mostly linked to institutional affiliation. Despite those clusters, the network is characterized by a high density of interconnections, which is not surprising given that following someone on Twitter does not require a high level of engagement. Furthermore, the most active and followed Twitter users are not central in this network, indicating that they are followed by people outside of the iSchools community. This suggests that the Twitter network is more than a simple extension of the professional network.
Looking at the 50 most frequently used hashtags by the iSchools community (Figure 2), based on the 132,638 tweets collected (24% of all tweets), one can see that library and information science related topics are among the most discussed. However, other frequently used hashtags relate to the general political and social context, showing that the iSchools community members also use Twitter to discuss topics that are not necessarily related to their work.
Conclusion:
The aim of this work was to assess the Twitter presence and activity of iSchools and their faculty members in order to determine whether it is used by scholars to expand their network and as a diffusion channel. Our results show that the Twitter network is not limited to professional connections and despite the fact that only a minority of faculty members (about 34%) have a Twitter account, these users reach a large number of people from outside the scientific community. This highlights Twitter’s potential as a tool for public outreach. In order to provide a more thorough description of the iSchools faculty members’ activity on Twitter, and to see how Twitter affordances (e.g., mentions, retweets, hashtags) are used, further studies could look at tweets content more systematically. Such studies could help understand to what extent this type of social media activity actually reflects scientists’ public outreach.
References:
Blondel, V.D., Guillaume, J.-L., Lambiotte, R., Lefebvre, E. (2008) Fast unfolding of communities in large networks. Journal of Statistical Mechanics: Theory and Experiment (10), P10008.
Haustein, S., Bowman, T.D., Holmberg, K., Peters, I., and Larivière, V. (2014). Astrophysicists on Twitter: An in-depth analysis of tweeting and scientific publication behavior. Aslib Journal of Information Management, 66(3), 279–296.
Holmberg, K., Bowman, T.D., Haustein, S., and Peters, I. (2014). Astrophysicists’ conversational connections on twitter. PloS ONE, 9(8), e106086–e106086.
Van Noorden, R. (2014). Online collaboration: Scientists and the social network. Nature, 512
