267 — Tweet success? Scientific communication correlates with increased citations in Ecology and Conservation

Lamb, Gilbert, & Ford (10.7717/peerj.4564)

Read on 14 May 2018
#social-media  #twitter  #internet  #social  #society  #altmetric  #metrics  #citation  #academia  #conservation  #biology  #ecology  #scientific-communication 

Twitter! Social media! Please like and subscribe.

The Altmetric Attention Score is just what it says — an alternative metric that attempts to capture academic paper impact based upon things that are not citations. For example, an alt-metric includes Facebook post attention, Twitter hullabaloo, news articles that reference a paper, Wikipedia mentions, and many more such indirect “citations.”

In 2011, a paper was published (Altmetric of 1604, 276 cites) that suggested that metrics such as Twitter attention — captured in the Altmetric Attention Score — were adequate to predict citation counts and, by proxy, an article’s impact factor.

So that’s pretty neat. But recently, that correlation has weakened, at least in the ecology and conservation sciences. This could be due to any number of causes, but most likely it’s a sign — at least in my opinion — that there’s more social media chatter about ecology and conservation in general (and so proportionally less per cite).

Altmetrics are useful for this correlation because they aggregate quickly right after a paper is published (as everyone reads it for the first time), and then essentially plateau by the time the first papers come out that cite that document. This is because of the long timeline on academic paper publication, but it’s a convenient point for our analytical purposes.

This research found that Wikipedia and Twitter tweets most closely correlated with ecology paper cite counts, whereas Facebook posts and news articles had very little correlation.

Obviously, more social-media-savvy authors will manage to get their paper into the hands and brains of more people as a result of their larger social network; and so it’s unclear to me how much of this is causation — either a higher impact paper garners more attention, or in the reverse, a paper with more attention will garner more citations — and how much is just correlation.