Altmetrics, Twitter, and Engagement : A Content Analysis Study of Tweets to Humanities Articles

Abstract: Altmetrics is a new bibliometric subfield that uses data from online platforms and social and mainstream media to track attention to and impact of scholarly content. Qualitative research is needed to increase knowledge of altmetrics and of what they may indicate about scholarly publications’ impact and influence. Through analyzing the second-largest source of altmetric events, tweets to academic articles, this thesis extends our understanding of these metrics. Using content analysis and an adapted framework for categorizing acts by levels of engagement, I analyzed 1,200 tweets to four humanities articles, recording their engagement with the articles, sentiment, hashtag use, and @mentions. I also categorized 203 of the tweeting accounts. There is considerable inter-article variation among tweets to the four articles. Overall, however, the tweets evidence low engagement with the articles, the retweet percentage is high, and many of the tweets are traced to few accounts. Contradicting prior suggestions that tweets may be particularly suited to indicating public engagement with scholarly articles, a large proportion of the tweets appear to be from academics. Confirming findings of prior studies, the results indicate that academics use Twitter to filter scholarly literature. I conclude that a single altmetric score is too reductive and of negligible use without deeper qualitative analysis. Unlike some altmetrics indicators that may reveal impact, tweets to articles rather indicate attention, casting doubt on the claim that altmetrics as a whole measure impact. Despite these limitations, tweets to articles and altmetrics more broadly can provide meaningful information not conveyed by citations.