José Luis Ortega, a web researcher from the Cybermetrics Lab at the Spanish National Research Council, recently investigated whether author participation on Twitter influences the volume of tweets about, and citations of, research articles.

To answer these questions, Ortega looked at 4166 articles from 200 authors; 76 were Twitter users and 124 were not. He found that the number of tweets about an article is strongly affected by author Twitter presence, with articles by Twitter users tweeted 33% more than those by non-users. This potential influence of author Twitter membership casts doubt over the suitability of the platform (e.g. number of tweets) as an objective measure for research evaluation. The number of followers that an author has on Twitter is also an important factor in increasing reach. However, the actual effect of follower number is relatively small; to achieve one extra mention on Twitter an author would need to increase follower numbers by three times.

An interesting relationship was uncovered between number of Twitter followers and article citations. Whereas there was no relationship between author presence on Twitter and number of citations, there was a weak relationship between number of followers and article citations. Ortega suggests that authors with a large number of followers on Twitter are able to share their research with a much wider audience, increasing the likelihood of citation in the future. Research impact may be increased to some degree by the intensity of dissemination, and the more channels that are used, the greater the likelihood of relevant people reading and later citing the article.

The author concluded that Twitter is a powerful method for disseminating research articles, and that wide dissemination increases the number of citations. However, Ortega poses the question – if dissemination alone can affect citation rate, then to what extent can citations be used as a reflection of research quality and novelty?

——————————————————–

Summary by Philippa Flemming, PhD from Aspire Scientific