Exploring the content composition of online book reviews

Abstract

Today, anyone can perform an opinion-expressing form of literary criticism by writing online book reviews. Sellers and publishers recognised the strategic potential of such reviews, for example, to increase sales. However, despite the popularity and recognised importance of book reviews, only little is known about the actual content in detail. Drawing on a category system and manually annotated reviews, this study explores the content composition of book reviews. We disclose frequently used content-related book review components and perform a cluster analysis, exploring which components often occur together. Our results support literary scholars in investigating the digital phenomenon of literary criticism and the study illustrates a sample Computational Humanities project which can be transferred to other research endeavours.