Robinson-Garcia, N., Costas, R., Isett, K., Melkers, J. and Hicks, D. (2017). The unbearable emptiness of tweeting—About journal articles. PLOS ONE, 12(8), p.e0183551.

Underlying data to the study

This recent paper from Robinson-Garcia et al, part of a project looking at dissemination channels for dentistry in the US, has (ironically enough) gained considerable traction on Twitter:

As a low-barrier platform to interact with a broad audience Twitter has proved popular with social-media savvy academics as a channel to disseminate their research outputs. It’s also infested with automated accounts, the dreaded Twitter bot, spewing links into the ether, everything from pornography to cutting edge research.

Robot image from Research Data Leeds dataset
Robot image from Research Data Leeds dataset

It’s so easy to tweet a link to an article, in fact, by clicking a button on a journal or repository for example, that many real people are indistinguishable from robots and the paper finds that, at least in the field of dentistry, less than 10% of tweets exemplify “an ideal of curating and informing about the literature”.

“The bulk of tweets about dental papers were sent by accounts seemingly run by people but whose dental journal article tweeting could be easily automated”

It’s an interesting and valuable paper. However, the value of Twitter as a tool for disseminating research is not as badly undermined as the provocative title might suggest. No disrespect to the authors who clearly know a thing or two about promoting their work (as of 1pm on Friday 1st September it has a very healthy altmetric score of 448 – including 659 tweets from 606 users, with an upper bound of 1,316,619 followers).

To see the live score see

Metrics have a lot to answer for and the paper is about counting tweets as a potential indicator of reach and impact. What it’s NOT really about is tweeting about your research, which can be valuable if you do it properly, spend time developing your network and interacting with them and with your research in a meaningful way.

This is the type of interaction we hope to encourage via the Open Research Leeds Twitter account @OpenResLeeds, which rather than that 90% of noise, we aim to be amongst the (nearly) 10% of valuable dissemination channels and a node in various academic networks across the University of Leeds and beyond.

One initiative is to leverage altmetrics to disseminate research when the ‘green’ self-archived version of the manuscript is released from embargo from the White Rose repository. The colour coded altmetric ‘score’ that is embedded in all WRRO and Symplectic records can be used to identify how and where journal articles have been disseminated and Twitter can be used to amplify the impact of research outputs, by retweeting a Leeds based author, for example, or linking to an open version of a paper from a mainstream news article discussing the research. The actual score doesn’t really matter, it’s simply a convenient method to visualise the network.

We are keen to develop synergies with other Leeds based accounts, through reciprocal retweets for example, and have curated a list of nearly 700 accounts associated with the University of Leeds – lists are a feature of twitter that offer a great way of limiting ‘noise’ by focusing on a specific subset of users such as a research community. ‘Hashtags’ can also be employed to emphasise specific types of content, #openaccess, for instance or #JiscRDM which is a powerful method of building community and attracting subscribers to your network.

(#JiscRDM is promulgated by Jisc to foster a community around Research Data Management and is used at community events such as the Research Data Network –

So tweeting your research need not be unbearably empty, just don’t be a robot.

Further reading:

To Tweet or Not To Tweet –  an Academic Questions [blog] (by Dr Ben Britton)

What happens when you tweet an Open Access Paper [blog] (by Melissa Terras)

Network effects: on alternative metrics [blog] (by @ukcorr)

Social Media for Academics [book] (by Mark Carrigan)