Developing methods for analysing and evaluating literary engagement in digital contexts

  • Lang, Anouk (Principal Investigator)

Project: Research

Project Details


"Research into reading and the value that it holds for individuals has been an area of rising interest for scholars within a constellation of related fields, including book history, cultural studies, the sociology of culture and literary studies. However, until relatively recently, researchers have been largely restricted to gathering accounts of readers' experiences from sources with limited evidence such as marginalia or accounts in diaries and letters, which are time-consuming to gather, or from somewhat artificial contexts such as interviews, questionnaires and focus groups, where what Pierre Bourdieu has termed the legitimacy effect - which prompts individuals to report on what they consider they ought to have read, or ought to think, rather than what they have actually read or actually think - may hamper the investigation. The advent of social media platforms, and the unprecedented amount of user-generated data they produce, has opened up compelling new avenues for research into reading, an experience long considered personal and private which is increasingly coming into view as a cultural activity whose social dimensions are equally as complex and as vital to understand.

Not only has user-generated data from social media platforms given researchers very large amounts of data to work with, it has also added new forms of information, such as numerical ratings of books accompanying descriptive reviews, and information about readers' affinity with others depending on the number of books in their shared collections (a more complex version of Amazon's recommendation algorithm that suggests books that users may wish to buy by comparing their purchasing or browsing activity to those of others). With these new forms of data, however, has come a parallel obligation for scholars to develop robust methodologies for their analysis, including the capacity to make use of the rich metadata - the temporal and geospatial information attached to some of the information produced by users - on social network sites. In the commercial world, such information is already exploited by marketers through such techniques as sentiment analysis - automated analysis that can identify whether a tweet is broadly negative or positive - but such techniques are too crude for understanding the complex interplay between texts and readers as meaning and value are negotiated. This is the gap that this project seeks to fill. By capturing information from two platforms used by individuals to report on their reading and to connect with other readers - Twitter and the book-collection website LibraryThing - it will investigate what this born-digital data can tell us about readers' interactions with texts by taking samples of this data, and exploring and refining the methods that can be used to deal with it, particularly as it contains both numerical and descriptive elements. It will apply techniques of computational analysis to this data to identify significant patterns in relation to three areas. First, thematic and topical information: what do readers choose to foreground when they go to the trouble of broadcasting their thoughts on Twitter or LibraryThing? Second, chronological trends: do particular terms become prominent in online interactions at particular times, and how might these trends relate to events such as the publication of books which achieve mass popularity such as the Harry Potter volumes or Fifty Shades of Grey? Third, geographical influences: do socio-economic factors or the existence of book-related institutions such as bookstores and libraries play a role in influencing the spatial distribution of online interactions around books? By using these disparate methodologies in concert, and finding ways to productively combine them, the project will both advance knowledge about the intrinsic value texts represent to readers, and point to ways that this research can be carried out with ever-larger bodies of data in the future."
Effective start/end date1/01/1431/07/14


  • AHRC (Arts and Humanities Research Council): £16,983.00


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.