Innovation in Research Methods: Investigative Social Research
On the 21st and 22nd November at the University of Liverpool in London, NCRM held an innovation forum to explore the embryonic field of “investigative social research” and the methods that underpin it. A dynamic and frequently high impact contemporary field, investigative social research encompasses work by non-governmental organisation/civil society researchers, data and investigative journalists, open source investigators, lawyers and independent researchers alongside social scientists of all kinds, from anthropologists, criminologists, epidemiologists, geographers, historians and sociologists through to those involved in accounting, economics and financial studies as well as data science. This emerging global field is characterised by the breadth of output it produces: often fast-circulating studies, news stories, reports, trackers and apps which attract global public attention. Researchers in the field make heavy use of: “new data technologies and analytics and other means of intellectual cross pollination, exchanging ideas and sometimes working and writing together, side by side, across borders, and genres each of them with different perspectives, backgrounds, interests, professional expertise, not to mention internationally and culturally diverse geographic and economic circumstances” (Lewis 2018: 23).
Professor Susan Banducci, University of Exeter, discusses the importance of flows of information in her talk on the ‘Politics of Openness’ at the Innovation Forum, © Terence Heng 2019
In this context, investigative researchers are developing approaches which, as Ruppert and Savage observe in the context of their work on investigations into the MPs expenses scandal, “engage with new forms of data and analytic techniques, undertake rich empirical analysis as well as develop new resources for understanding [the world and what happens in it]” (Ruppert and Savage 2009: 17). In so doing, researchers are contributing to the development of distinctive new “ways of knowing” (ibid.). No longer the preserve of universities and academic disciplines, this opening up and reworking of “the methods and practices that researchers and analysts use to make sense of data” and do useful things with it (Arribas-Bel & Reades 2018: 5) is happening across disciplinary, sectoral and geographical boundaries (with collaborations spanning the globe and involving researchers from countries in the Global South as much as the North). Investigative social research, as the forum showed, is often data intensive, digitally enabled, highly collaborative and impactful and gives rise to its own distinctive “politics of method” (Savage & Burrows 2007, Gray 2019).
The purpose of the forum was to explore the methods that enable researchers in this field to pursue topics of social, political and economic import. Whether it is the investigations into the Panama and Paradise Papers, the identification of the Salisbury Novichok poisoners, the documenting of NATO airstrikes as well as civilian deaths across the Middle East, the verification of political violence in Africa through crowd-sourced video and photographic imagery, the tracking and tracing of those infected by COVID-19 during the current pandemic or in quickly contextualising and framing emerging news stories by drawing on new datasets and analytical techniques, investigative social research certainly engages with important aspects of our lives.
Following an opening keynote by David de Roure, which highlighted the possibilities for cross-sectoral and inter-disciplinary research in the digital era, the forum focused on four major areas of both innovation and debate within investigative research:
1. Following the Money
After “All the President’s Men” - the dramatisation of the journalistic investigation leading to the fall of Richard Nixon - the injunction to “follow the money” entered the popular lexicon. Spoken by the secretive whistleblower Deepthroat in William Goldman’s screenplay, the injunction was that The Washington Post journalists Carl Bernstein and Bob Woodward investigating what had happened should follow Nixon’s hidden campaign finances in order to uncover the scandal. Since then, “following the money” has become an effective way of rendering the hidden activities of the powerful visible, as most recently demonstrated by the Panama and Paradise Paper leaks. Following the money has thus become a significant investigative research method. In this section of the workshop, investigative researchers were therefore invited to discuss the data, methods and techniques they employ in the course of tracing financial transactions and structures with thought provoking presentations by Richard Whittel of Corporate Watch with David Whyte, University of Liverpool, on their collaborations, and Jean Shaoul, University of Manchester, on locating and decoding financial and accountancy data for research purposes.
2. The Needle and the Haystack: Investigating at Scale
The volume of digital information produced on an hourly basis is daunting but investigative researchers are drawn to this vast sea of information in order to discover concrete things about the world in rigorous ways. This session examined some of the methods used to do this.
Milena Marin opened the session by discussing Amnesty International’s Decoders Project, which she leads, focusing specifically on their Strike Tracker study. Involving an eclectic but powerful mix of: globally crowd-sourced satellite image analysis; painstaking field research; the analysis of data made available by the US (and other members of the military coalition it led in Syria against ISIS) and collaborations with other NGOs such as Airwars, this project tracked the damage done to Raqqa by coalition forces from the built environment to civilian deaths. A major global news story in 2019, the analyses produced by the project changed how the world viewed the conflict.
John Burn Murdoch from the Financial Times’ award-winning Data Journalism Team then discussed the techniques the FT employs in order to respond to emerging news items with analysis that both contextualises and helps readers grasp the implications of emerging stories quickly and insightfully from Brexit to COVID-19. As John explained, the members of the team draw on a diverse array of datasets to produce their analyses, and social research data plays an important role in that work.
Finally, Anna Fitzmaurice from the Turing Institute discussed the importance of undertaking creatively critical research in the emerging field of data science. While the gendered character of contemporary data science has been much discussed, few had attempted to use data science methods and techniques in order to investigate the underlying problem – something the project she was currently involved in was seeking to rectify, repurposing datasets and methods designed for other ends in the process.
3. Going Here, There and Everywhere: The Possibilities of Geospatial Data
Nothing may seem more obvious than the observation that things happen somewhere. However, investigating where things actually happened and, from there, working out how to more accurately characterise what took place, can be an immensely informative exercise. Whether this involves locating people, objects, or processes, the results are powerful. Reflecting on investigative research involving geo-spatial data of many different kinds, this section of the workshop emphasised the centrality of location to understanding the contemporary world and what takes place within it.
The session opened with Suzanne Vanhooymissen from the BBC’s award-winning Africa Eye Investigations team who discussed their investigation of political violence in Sudan in 2019 and the methodical work which underpinned the resulting documentary, ‘Sudan’s Livestream Massacre’. Drawing on images, video and metadata from more than 300 smartphones, the team pieced together the events of one of the bloodiest days in the conflict. While the digital data was critical, however, Suzanne was at pains to underline the rigorous behind-the-scenes work that underpinned the verification of the data they received. She also noted that the team had to work hard to build contacts with protestors in Sudan in order to receive the data in the first place. Trust and interpersonal connections were thus as important to the work of the team as data analysis. This was emphasised further by Chris Woods, Director of Airwars UK, an NGO involved in “tracking, assessing and archiving international military actions and related civilian harm claims in conflict zones such as Iraq, Syria and Libya”, which also works “with militaries, where practicable, to help improve understanding of public civilian harm allegations – with the aim of reducing battlefield casualties”.
Focusing on the role of “open source intelligence” in their work, data recovered from various digital sites across the world, Chris discussed how investigative research can “dispel the fog of war”, providing a much clearer picture of what actually takes place in conflict zones. Using the example of their joint work with Amnesty on the assault on Raqqa, like Suzanne, Chris pointed to the significance of “human intelligence” alongside digital evidence of many kinds, not just as sources of information but also as analysts.
Finally, Jon Reades from the school of Geography at Kings College London, discussed ways in which geospatial data can best be managed when working in fields where versions of software are continuously shifting, providing an overview of what he called ‘the full stack’, a suite of tools essential to ensuring the integrity of the research process when faced with a deluge of data.
4. When and How to Open Things Out
In a world where information is everywhere and researchers are accessing rather than generating it per se, questions of ethics take on a rather different cast. Investigative researchers do tend to pride themselves on conducting ethical research (see for example: Gray & Lämmerhirt 2019) but here as elsewhere those norms take different forms and are embedded in different practices, for instance when striking a balance between ensuring trust among and acknowledging the contributions of collaborating partners, protecting the anonymity of sources but also publishing work in the public interest. This session examined how ethical standards are translated into practice and what digital ethical standards mean.
The session opened with a talk by Meirion Jones, Director and founder of The Bureau of Investigative Journalism (TBIJ), which examined the work of TBIJ with a particular focus on collaboration and impact. Among the many issues Meirion raised, including the important point that research requires funding which non-academic researchers can find hard to consistently guarantee over time, a key aspect of his point was that the scale of the stories journalists are currently tackling means they cannot work alone – collaboration is critical. This is not just about bringing together information; it is also about analysing it, with work on particular stories undertaken by consortia of journalists undertaking distributed work across the globe. Confidentiality, secrecy but also maintaining integrity and ethical standards – that is, finding agreed paths between the closed and open aspects of the investigative process – are major issues within this distributed work. Moreover, finding ways of working together while upholding standards of journalistic probity was not just an issue across countries but within them. With resources for local journalism disappearing, Meirion suggested that crowd-journalism, providing a platform for local investigative journalists, was going to be an increasingly important strand in their work.
Meirion was followed by Ella McPherson, Cambridge University, who talked about “Open Source Investigations and the Technology-Driven Knowledge Controversy in Human Rights Fact-Finding”. Drawing on her experience of setting up The Whistle, a site that provides “digital tools to helpfully, efficiently and safely report and verify human rights issues”, Ella pointed out the importance of maintaining a critical stance towards new investigative methods and methodologies. As Ella noted, we have to situate our new tools socially, culturally, historically and politically, and understand that they create blindspots as well enabling us to see more clearly in some areas because new tools are good at some things but not all, implying trade-offs. A balanced approach recognising the possibilities and limits of new methods and forms of data was thus needed. This was also a theme in the final talk by Susan Banducci of the University of Exeter on “The Politics of Openness”, where she drew attention to the ways in which new methods and forms of data increase knowledge but also where they create overload and establish spaces for disinformation, often involving the mispurposing of bona fide research for partisan purposes,.
The speakers in these sessions discussed their work with an audience that was itself interdisciplinary and cross-sectoral in character and involved in investigative research itself. In the course of those discussions, particular issues raised across all of the themes included: accessing, managing and linking heterogeneous datasets both large and small, and methods for combining human and technological resources in investigative, analytical and verification work. A vibrant area of contemporary research, NCRM will work with researchers involved in investigations as part of sharing and opening access to its cutting edge.
The presenters were:
David De Roure (Oxford E-Research centre): The New Landscapes of Data and Research
Richard Whittle (Corporate Watch)/David Whyte (University of Liverpool): Using Open Financial Data to Research Corporate Power
Jean Shaoul (University of Manchester): Financial Data Sources and Their Uses
Milena Marin (Amnesty International): Amnesty International: On Big Data, Small Tasks and Massive Engagement
John Burn-Murdoch (The FT): Explorations with Open Data
Anna Fitzmaurice (Alan Turing Institute): Women in Data Science and AI: A Social Data Science Case Study
Suzanne Vanhooymissen (BBC Africa Eye Investigations): Investigating Sudan’s Livestream Massacre
Chris Woods (Airwars UK): The Airwars Approach: How Geolocation and Open Source Intelligence Can Help Disperse the Fog of War
Jon Reades (KCL): The Full Stack: Designing a Robust and Replicable Geospatial Work Stream
Meirion Jones (The Bureau of Investigative Journalism): The Work of The Bureau of Investigative Journalism: Collaboration and Impact
Ella MacPherson (Cambridge University): Open Source Investigations and the Technology-Driven Knowledge Controversy in Human Rights Fact-Finding
Susan Banducci (University of Exeter): The Politics of Openness
Arribas-Bel, D. & Reades, J. (2018) “Geography and Computers: Past, Present, and Future”. Geography Compass, 12(10), e12403, https://doi.org/10.1111/gec3.12403
Gray, J. (2019) ‘Data Witnessing: Attending to Injustice with Data in Amnesty International’s Decoders Project’, Information, Communication & Society, 22(7): 971-991
Gray, J., & Lämmerhirt, D. (2019) ‘Making Data Public? The Open Data Index as Participatory Device, pp. 174-188 in Daly, A., Devitt, S.K. & Mann, M. (eds.) Good Data, Amsterdam: Institute of Network Cultures
Lewis, C. (2018) ‘Tear Down These Walls: Innovations in Collaborative Accountability Research and Reporting’, pp. 5-25 in Sambrook, R.(ed.) Global Teamwork: The Rise of Collaboration in Investigative Journalism, Oxford: Reuters Institute for the Study of Journalism, Department of Politics and International Relations, University of Oxford
Ruppert, E. & Savage, M. (2009) ‘New Populations: Scoping Paper on Digital Transactional Data’, CRESC Working Paper no. 74, Manchester: CRESC
Sambrook, R.(ed.) (2018) Global Teamwork: The Rise of Collaboration in Investigative Journalism, Oxford: Reuters Institute for the Study of Journalism, Department of Politics and International Relations, University of Oxford
Savage, M. & Burrows, R. (2007) ‘The Coming Crisis of Empirical Sociology’, Sociology, 41(5): 885-899