Altmetrics for the humanities: Comparing Goodreads reader ratings with citations to history books

Structured Abstract: Purpose: This paper assesses the value of Goodreads reader ratings as an altmetric research tool for measuring the wider impact of scholarly books published in the field of History. Design/methodology/approach: Book titles were extracted from the reference lists of articles that appeared in 604 history journals indexed in Scopus (2007-­‐2011). The titles were cleaned and matched using an API in WorldCat.org (for publisher information) as well as Goodreads (for reader ratings). A set of N=8,538 books was first filtered based on Dewey Decimal Classification class 900 'History and Geography', then a subset of 997 books with the highest citations and reader ratings (i.e. top 25%) was analyzed separately based on additional characteristics. Findings: A weak correlation (0.212) was found between citation counts and reader rating counts for the full dataset (N=8,538). An additional correlation for the subset of 997 books indicated a similar weak correlation (0.190). Further correlations between citations, reader ratings, written reviews, and library holdings indicate that a reader rating was more likely to be given to a book on Goodreads if the book was held in an international library, including both public and academic libraries. Originality/Value: Altmetrics research has focused almost exclusively on scientific journal articles appearing on social media services (e.g., Twitter, Facebook). In this paper we show the potential of Goodreads reader ratings to identify the impact of books beyond academia. As a unique altmetric tool Goodreads can allow scholarly authors from the social sciences and humanities to measure the wider impact of their books.