Digital and Quantitative Methods for the Study of Screen Cultures

Digital and Quantitative Methods for the Study of Screen Cultures

An overview of "Digital and Quantitative Methods for the Study of Screen Cultures", convened by Dr Alan O'Leary and fourth in the Sadler Seminar series "Remapping World Cinemas for the Digital Age".

Sadler Seminar Overview: Digital and Quantitative Methods for the Study of Screen Cultures, 05 May 2017

This fourth session in the Sadler Seminar series, ‘Remapping World Cinemas for the Digital Age’, was methodological in emphasis. We considered the achievements and potential as well as some of the problems with digital and quantitative approaches to the study of film and audiences, and to new and social media.

Allison Cooper (Bowdoin College, USA) discussed her experience of building and annotating archives of film clips using the online digital management system Omeka. Tagging was a big theme – how tags are defined, how tagging can be done effectively, and how the crowdsourcing approach to tagging such as that adopted by Netflix has the potential to offer super-granulated descriptions of texts. Marco Cucco (Università della Svizzera italiana) discussed the payoffs of the economic study of cinema – and some of its frustrations. He mentioned the loneliness of the cinema-economics scholar (at least of Italian cinema) and the fact that so much useful viewing data is never made available by the likes of Netflix and Amazon. Heather Ford (Media, Leeds) talked about the importance of ‘cosmopolitan’ methodologies and the need to use using qualitative alongside quantitative approaches. She described doing qualitative interviews based on data visualizations which worked as a spur to dialogue and could then be made more accurate by interviewees. Huw Jones (currently University of York, soon Southampton) considered the kind of questions that can be answered using box office and admissions statistics. He showed how data helped the large MeCETES project answer its ambitious research questions. Nick Robinson (Politics and International Studies, Leeds) talked about a large project on the militarization of social media and the need to approach datasets with a qualitative sensibility. He emphasised how qualitative writing was essential in order to identify categories for the quantitative analysis of video material.

The session was convened and chaired by myself, Alan O’Leary (Centre for World Cinemas and Digital Cultures, Leeds). I organised the event because I’m looking for ways to radically expand the corpora we use as film scholars (in my own work, with particular reference to Italian cinema and history). My interest is part of a widespread desire for methods of greater reach that has been articulated in recent years in film studies. Major forums like the Society for Cinema and Media Studies (SCMS) annual conference have hosted discussions and workshops devoted to investigating how film studies can benefit from quantitative and computer-based approaches. Statistical analysis has increasingly been employed ‘to answer questions about the economics of the film industry, about patterns in the style and form of motion pictures, about audiences’ behaviours and attitudes, and about how we understand and experience the cinema’ (Redfern 2014). Identifying and counting the incidences of particular elements within film texts (a useful technique to trace the development of a particular genre over time, for example) has increasingly been done with the help of computers. Software has been developed to generate data (and databases) on shot length and shot scale (e.g., Shot Logger, CineMetrics, or to allow the annotation of digital copies of films (e.g., Lignes de temps, Omeka), and so on. And of course, web resources like the Internet Movie Database (IMDb ― founded by Leeds alumnus Col Needham) are a potential trove of user-generated data.

Methods of quantitative study, mapping and the employment of evolutionary models championed by Franco Moretti in the study of literature also have much to offer film and screen studies (see Dudley Andrew’s article ‘An Atlas of World Cinema’, 2005). One advantage of Moretti’s ‘distant reading’ methods is to allow a ‘focus on units that are much smaller or much larger than the text: devices, themes, tropes—or genres and systems’ (Moretti 2013; Moretti himself has applied some of these methods in ‘Planet Hollywood’, a not altogether persuasive article from 2001 on genre and national markets). The employment of digital, statistical and ‘distant reading’ methods is also a response to the return of questions of the ‘longue durée’ in the discipline of history (Armitage 2012; Armitage and Guldi 2014), being designed to better access changes and continuities over longer periods. But needless to say, there’s no need to disown traditional competences in film studies: the ideal is ‘a blending of expertise from previously antipathetic disciplines’ (Butler 2014). A holistic approach (quantitative and qualitative) is what we might aim for, which adopts empirical, theoretical, historical, and cultural perspectives, dealing with production, aesthetics, reception and discourse. The use of mixed, or as Heather Ford called them in the seminar, cosmopolitan methods should facilitate the ‘toggling’ of scales from the very largest to the very smallest in order to ‘ground quantitative generalizations in the concrete particulars of microhistorical studies’ (Maltby 2006). These ruminations are programmatic and abstract: the realities of using data and digital are subject to ethics, access to tech and expertise, and most of all, to the careful honing of research questions. The Sadler session brought all these issues to light.

Generously funded by the Leeds Humanities Research Institute and the AHRC Open World Research Initiative project "Cross-Language Dynamics: Reshaping Community"