I was quite bemused while browsing the net the other day (heading down a few rabbit warrens) to find myself at the Journal of Medical Internet Research (JMIR) where I noticed that, as well as giving citation figures for each article, the journal also provides ‘tweetation’ figures, showing the number of times the URL to the journal article has been tweeted.
Further investigation led to the discovery that the creator of the term ‘tweetation’, Gunther Eysenbach, published a study in the same journal in 2011 entitled ‘Can Tweets Predict Citations? Metrics of Social Impact Based on Twitter and Correlation with Traditional Metrics of Scientific Impact‘.
The study mined all tweets containing links to articles in the JMIR published between July 2008 and November 2011. The results showed an association between tweetations and citations, finding that highly tweeted articles were 11 times more likely to be highly cited than less-tweeted articles, thus enabling top-cited articles to be predicted from top-tweeted articles. Essentially the study argued that social impact measures based on tweets – the ‘twimpact’ factor (defined as ‘the cumulative number of tweetations within n days after publication’ ) can be used to complement traditional citation metrics.
The article’s current tweetation count is 1216, so evidently it has sparked a lot of interest in the twittersphere, as well as giving rise to some commentary and controversy.
Laura Costello, for example, argued in EdLab that the article has implications for the marketing of academic publications:
The results found a weekend “drop off” time for tweetations and the author suspects that there may be an ideal day of the week for maximum social publishing impact. This could be important for publishers and providers that have begun to feel long tail demand and have pushed to make articles available à la carte to unaffiliated scholars.
Phil Davis provided a critique of the study in a post for The Scholarly Kitchen, in which he made some salient points including the fact that, not only were many of the counted tweets not sent out by humans, article citations are not comparable with microblogging:
Unlike an article citation — which requires an author to produce a new piece of research, have it vetted by peers, and published in a journal, which is indexed by a reputable source that tallies citations — there are very low barriers to microblogging.
There is no doubt that twitter metrics provides some level of insight into research impact, although to what extent it does so, is unclear.
Whatever the significance of ‘tweetation’ and ‘twimpact factors’ for scholarly communication, at the very least we now have two new fantabulous words to add to the English lexicon.