An epistemological experiment: issue mapping, data journalism and the public understanding of complex issues

Slides for talk at Utrecht Data School, Utrecht University, 27th October 2014. Further details at: http://jonathangray.org/2014/10/22/digital-methods-data-journalism-utrecht/

Transcript

1. An epistemological experiment: issue mapping, data journalism and the public understanding of complex issues 27th October 2014, Utrecht Data School, Utrecht University Liliana Bounegru | lilianabounegru.org | @bb_liliana! Jonathan Gray | jonathangray.org | @jwyg
2. Gray, J., Bounegru, L. & Chambers, L. (2012) The Data Journalism Handbook.! Available at: http://datajournalismhandbook.org/
3. Rogers, R. (2013) Digital Methods.
 Cambridge, MA: MIT Press.
4. “An epistemological experiment”
5. social science researchers leading digital newsrooms new ways of covering complex issues = +
6. social science! researchers leading digital newsrooms new ways of covering complex issues = +
7. Using digital traces to study the social: actor-network theory, issue mapping, digital methods
8. – Bruno Latour, Reassembling the Social: An Introduction to Actor-Network Theory (2005) “[T]here is nothing specific to social order; (…) there is no social dimension of any sort, no social ‘context’, no distinct domain of reality to which the label ‘social’ or ‘society’ could be attributed; (…) no ‘social force’ is available to ‘explain’ the residual features other domains cannot account for (…) and (…) society, far from being the context ‘in which’ everything is framed, should rather be constructed as one of the many connecting elements circulating in tiny conduits”
9. “The social is visible only by the traces it leaves..” – Bruno Latour, Reassembling the Social: An Introduction to Actor-Network Theory (2005)
10. “The interest of electronic media lies in the fact that every interaction that passes through them leaves traces..” – Bruno Latour & Tommaso Venturini, “The Social Fabric: Digital Traces and Quali-quantitative Methods” (2009)
11. Digital methods are “methods of the medium” designed to repurpose digital objects such as tags, likes, links and hashtags to study issues. – Digital Methods Initiative,
 University of Amsterdam
12. Three examples:! 1. climate change negotiations! 2. rise of the far right in Europe! 3. health worker migration
13. Example #1:! mapping dominant topics and countries in UN climate negotiations
14. EMAPS (2014) “Climaps: A Global Issue Atlas of Climate Change Adaptation”! Available at: http://climaps.org/
15. “Adaptation turn”
16. – Leo Hickman, “Can carbon offsetting ever be truly green?”,
 The Guardian, 3rd September 2008. “In what seems like a flash, the climate-change debate has lurched from talk of mitigation to one of adaptation.”
17. The Atlantic (2014) “The UN's New Focus: Surviving, Not Stopping, Climate Change”.
 Available at: http://www.theatlantic.com/international/archive/2014/04/the-uns-new-focus-surviving-not- stopping-climate-change/359929/
18. Can the shift from mitigation to adaptation be observed in the UNFCCC negotiations?
19. Venturini, T., Baya-laffite, N., Cointet, J., Gray, I., Zabban, V., & De Pryck, K. (2014) “Three Maps and Three Misunderstandings : A Digital Mapping of Climate Diplomacy.” Big Data and Society, 2014, 1(1). 
 Available at: http://medialab.sciences-po.fr/publications/misunderstandings/
20. Findings! Mitigation is more dominant - the majority of the clusters are about mitigation. Mitigation is much more diverse and distributed. Adaptation is a much more tightly clustered topic and highly connected to other topics.
21. Venturini, T., Baya-laffite, N., Cointet, J., Gray, I., Zabban, V., & De Pryck, K. (2014) “Three Maps and Three Misunderstandings : A Digital Mapping of Climate Diplomacy.” Big Data and Society, 2014, 1(1). 
 Available at: http://medialab.sciences-po.fr/publications/misunderstandings/
22. Venturini, T., Baya-laffite, N., Cointet, J., Gray, I., Zabban, V., & De Pryck, K. (2014) “Three Maps and Three Misunderstandings : A Digital Mapping of Climate Diplomacy.” Big Data and Society, 2014, 1(1). 
 Available at: http://medialab.sciences-po.fr/publications/misunderstandings/
23. Findings! Both adaptation and mitigation are highly visible in negotiations.
 
 Adaptation financing has been central to climate negotiations from the outset. There is a noticeable shift towards adaptation during the period we examined.
24. Venturini, T., Baya-laffite, N., Cointet, J., Gray, I., Zabban, V., & De Pryck, K. (2014) “Three Maps and Three Misunderstandings : A Digital Mapping of Climate Diplomacy.” Big Data and Society, 2014, 1(1). 
 Available at: http://medialab.sciences-po.fr/publications/misunderstandings/
25. Which countries intervene most in UN climate negotiations and how do these interventions evolve over time?
26. Graphing the number of interventions in the negotiations of the 21 most active countries based on daily summaries from the Earth Negotiations Bulletin (ENB)
27. Venturini, T., Baya-laffite, N., Cointet, J., Gray, I., Zabban, V., & De Pryck, K. (2014) “Three Maps and Three Misunderstandings : A Digital Mapping of Climate Diplomacy.” Big Data and Society, 2014, 1(1). 
 Available at: http://medialab.sciences-po.fr/publications/misunderstandings/
28. Findings! Notable stability in presence and intervention of countries. Notable exceptions include Bolivia and Philippines who are becoming more prominent in recent negotiations. Most active are China (representing G77), United States and Europe. Countries tend to be more active when they host the negotiations.
29. Venturini, T., Baya-laffite, N., Cointet, J., Gray, I., Zabban, V., & De Pryck, K. (2014) “Three Maps and Three Misunderstandings : A Digital Mapping of Climate Diplomacy.” Big Data and Society, 2014, 1(1). 
 Available at: http://medialab.sciences-po.fr/publications/misunderstandings/
30. Wired Italia (2014) “Cambiamenti del clima: 20 anni di conferenze”. March 2014. No. 60.
31. Wired Italia (2014) “Cambiamenti del clima: 20 anni di conferenze”. March 2014. No. 60.
32. Wired Italia (2014) “Cambiamenti del clima: 20 anni di conferenze”. March 2014. No. 60.
33. Wired Italia (2014) “Beautiful Information, in mostra le migliori infografiche di Wired”.
 Available at: http://www.wired.it/attualita/media/2014/03/04/beautiful-information-infografiche-wired/
34. Wired Italia (2014) “Beautiful Information, in mostra le migliori infografiche di Wired”.
 Available at: http://www.wired.it/attualita/media/2014/03/04/beautiful-information-infografiche-wired/
35. Example #2:! mapping the rise of the far right in Europe with the web and social media
36. The Guardian (2013) “The rise of far right parties across Europe is a chilling echo of the 1930s”.
 Available at: http://www.theguardian.com/commentisfree/2013/nov/15/far-right-threat-europe-integration
37. Huffington Post (2014) “Sudden Rise of Far Right Groups in EU Parliament Rings Alarm Bells Across Europe”. Available at: http://www.huffingtonpost.com/elinadav-heymann/sudden-rise-of-far-right- _b_5512961.html
38. New York Times (2014) “Populist Party Gaining Muscle to Push Britain to the Right”.
 Available at: http://www.nytimes.com/2014/05/08/world/europe/populist-party-gaining-muscle-to-push- britain-to-the-right.html
39. What are the recruitment methods
 of far right groups?
40. Are current recruitment counter-measures proving effective?
41. What kinds of issues are most active amongst far right groups?
42. How are far right extremist groups connected to populist right and other right wing groups?
43. Profiles for 13 European countries.
44. 1. List of links per country 2. Analyse links between them 3. Study issues and actors
45. Findings
 New issues (e.g. environment, anti- globalisation and rights), principles and recruitment techniques. 
 Counter-measures are outdated. ! Islamophobia is located primarily in the North.
46. Greece: blood and soil and organic markets
47. Rogers, R. et al (2013) “Right-Wing Formations in Europe and Their Counter-Measures: An Online Mapping”. Digital Methods Initiative. https://wiki.digitalmethods.net/Dmi/RightWingPopulismStudy
49. Hungary: horse and yurt recruitment festivals
50. Rogers, R. et al (2013) “Right-Wing Formations in Europe and Their Counter-Measures: An Online Mapping”. Digital Methods Initiative. https://wiki.digitalmethods.net/Dmi/RightWingPopulismStudy
52. Taking back the yurt?
53. Counter-Jihadist groups on social media
54. The Guardian (2012) “Far-right anti-Muslim network on rise globally as Breivik trial opens”. Available at: http://www.theguardian.com/world/2012/apr/14/breivik-trial-norway-mass-murderer
55. Hope Not Hate (2012) “Counter-Jihad Report”.
 Available at: http://www.hopenothate.org.uk/counter-jihad/
56. Are different Counter-Jihadist groups in Europe connected? If so how?
57. Digital Methods Initiative. “Counter-Jihadist Networks: Mapping the Connections Between Facebook Groups in Europe.”
58. Digital Methods Initiative. “Counter-Jihadist Networks: Mapping the Connections Between Facebook Groups in Europe.”
59. Findings
 
 Facebook is an important medium for extremist groups. ! Three main clusters based on geographical proximity. ! European Counter-Jihadist groups are networked and transnational.
60. Digital Methods Initiative. “Counter-Jihadist Networks: Mapping the Connections Between Facebook Groups in Europe.”
61. Who are the new leaders?
63. Findings! ! Offline leaders are active on Facebook. ! There are also new emerging online leaders. ! New technique for identifying online leaders.
64. Example #3:! who’s talking about health worker migration?
65. “care drain”
66. Mills et al (2008). “Should active recruitment of health workers from ! sub-Saharan Africa be viewed as a crime?”. Lancet 2008; 371: 685–88.
67. The Guardian (2011). “Assessing the cause and effect of health worker migration”.! Available at: http://www.theguardian.com/global-health-workers/health-workers-move-from-area-of-origin
68. – Sue George, “Assessing the cause and effect of health
 worker migration”, The Guardian, 18th January 2011. “Health worker migration is a big issue – in 2005, it was widely reported that there were more Malawian doctors in Manchester than Malawi. Now, it seems, there are more Ethiopian doctors in Chicago than in Ethiopia.”
69. – Sue George, “Assessing the cause and effect of health
 worker migration”, The Guardian, 18th January 2011. “[a 2010 global code of practice] sets out guiding principles and voluntary international standards for recruitment of health workers, to increase the consistency of national policies and discourage unethical practices”
70. Which actors in the UK health sector are talking about the migration of health workers?
71. Rogers, R., Sanchez Querubin, N. & Kril, A. (2015) Ageing Places: A Digital Issue Mapping.
 Amsterdam: University of Amsterdam Press.
72. Findings for UK! Notable difference between sectors. Most vocal are government and academia.
 
 Less discussion amongst healthcare NGOs. Next to no discussion in private sector.
73. Who is recruiting Polish health workers?
74. Findings for Poland! ! Demand for Polish care workers greater outside country than within. ! Based on recruitment postings, Poland looks to be susceptible to care drain.
75. Rogers, R., Sanchez Querubin, N. & Kril, A. (2015) Ageing Places: A Digital Issue Mapping.
 Amsterdam: University of Amsterdam Press.
76. Digital tools and methods
77. Digital Methods Initiative (2014) DMI Tools.! Available at: http://tools.digitalmethods.net
78. Sciences Po Media Lab (2014) “Tools we develop and tools we use”! Available at: http://tools.medialab.sciences-po.fr/
79. “Netvizz is a tool that extracts data from different sections of the Facebook platform (personal profile, groups, pages) for research purposes.”
80. Rieder, B. (2013). Studying Facebook via data extraction: the Netvizz application. In WebSci '13 Proceedings of the 5th Annual ACM Web Science Conference (pp. 346-355). New York: ACM.
81. Netvizz: https://tools.digitalmethods.net/netvizz/facebook/netvizz/
82. “The Twitter Capture and Analysis Toolset
 (DMI-TCAT) captures tweets and allows for multiple analyses (hashtags, mentions, users, search, ...).”
83. Borra, E. & Rieder, B. (2014) “Programmed method: developing a toolset for capturing and analyzing tweets”. Aslib Journal of Information Management. Vol. 66 No. 3: 262-278.
84. TCAT: https://wiki.digitalmethods.net/Dmi/ToolDmiTcat
85. TCAT: https://wiki.digitalmethods.net/Dmi/ToolDmiTcat
86. “The Issue Crawler is web network location and visualisation software. It consists of crawlers, analysis engines and visualisation modules.”
87. “A software tool that locates and visualizes networks on the web”
88. Issue Crawler: https://wiki.digitalmethods.net/Dmi/ToolIssueCrawler
89. “Gephi is an interactive visualization and exploration platform for all kinds of networks and complex systems, dynamic and hierarchical graphs.”
90. Gephi: http://gephi.org
91. Which of these techniques might be of interest in journalism?
92. i. Co-occurrence analysis to identify themes ! ii. Network analysis to identify actors and sources ! iii. Hyperlink analysis to explore “politics of association” ! iv. Resonance analysis to identify source partisanship
93. social science researchers leading! digital! newsrooms new ways of covering complex issues = +
94. Emerging practises and epistemologies of data journalism
95. Geeks vs. pundits: ! The clash of two epistemological cultures
96. "Nate Silver says this is a 73.6 percent chance that the president is going to win? Nobody in that campaign thinks they have a 73 percent chance — they think they have a 50.1 percent chance of winning. And you talk to the Romney people, it’s the same thing. . . . Anybody that thinks that this race is anything but a toss-up right now is such an ideologue, they should be kept away from typewriters, computers, laptops and microphones for the next 10 days, because they're jokes." (Joe Scarborough, MSNBC, 2012)
97. “I am Nate Silver, the Lord and God of the Algorithm!” (Jon Stewart, 2012)
98. The life of data in the newsroom
99. “Objects, too, have agency” – Bruno Latour, Reassembling the Social: An Introduction to Actor-Network Theory (2005)
100. What journalistic practices, values and visions are articulated around the use of data as raw material for reporting? How are traditional journalistic practices, values and norms, transformed?
101. The politics of quantitative methods in the newsroom
102. Where do journalists’ attachments to particular forms of quantitative analysis come from? How are these commitments articulated? How do they shape the process of knowledge production and its outcomes? What quantitative methods are being left out? (the question of alternative histories)
103. Rieder, B (2013) “Interactive visualization and exploration of network data with Gephi”. Presentation from DMI Summer School. Available at: http://www.slideshare.net/digitalmethods/gephi-rieder-23834788
104. social science researchers leading digital newsrooms new ways of! covering! complex issues = +
105. How might digital and tools methods for issue mapping be used in journalism to improve coverage of complex issues?
106. Opportunities and challenges
107. Paper for Tow Center for Digital Journalism
108. Interviews and discussions with journalists in Europe and US about issue mapping and digital tools and methods
110. Some preliminary findings and reflections…
111. Some different potential uses of digital methods in journalism: ! • Story discovery (news desk/projects) • Internal reference resource (news desk/projects) • Preparation for live coverage (news desk/projects) • Quick/easy tools for journalists (news desk/projects) • Presentational device (interactive/graphics) • Interactive news “toys” for exploration (interactive)
112. Some challenges to using digital methods in journalism: ! • Time, resource and budget constraints • Resistance to change in (especially bigger) newsrooms, hard to introduce new tools/methods • Social scientists often want to capture complexity, journalists often want to simplify • Tension between traditional journalistic values (recency, human interest, etc) and research values • Rendering complexity readable and accessible to broader publics - not just issue experts/researchers • Keeping interactive projects about current events up to date • Not just tool provision, but also training • Transparency of tools and interpretation of results • Some of tools are complex to install and no web version available • Speed of using tools as events unfold • Efficiency of these methods compared to others • Hard to find stories in data
113. Some opportunities for using of digital methods in journalism: ! • Interest in using tools from interactive teams • More newsrooms have been experimenting with related approaches, still at very early stage • Potential for researchers working with journalists (rather than issue experts) to help with selection, filtering, framing and narration • Introducing robust methodology around use of social media data in newsrooms • Identifying human sources for interviews. • New forms of analysis and verification of sources. • Hyperlink analysis and web data currently very rarely used in journalism • Low uptake of text-mining and scientometrics tools and methods • Input/feedback from journalists could feed into existing software development • New web versions of existing tools (e.g. Gephi) • Using social media and web as data, not just content (to look at relationships and interactions)
114. Next steps: • Preliminary report for Tow Center • Embedded experiments in newsrooms • Pilot around Paris 2015 climate negotiations • Toolkit and handbook for journalists
115. Thank You! Liliana Bounegru | lilianabounegru.org | @bb_liliana Jonathan Gray | jonathangray.org | @jwyg Sciences Po médialab http://www.medialab.sciences-po.fr/ ! Sciences Po médialab - Tools http://tools.medialab.sciences-po.fr/ Digital Methods Initiative https://digitalmethods.net
 
 Digital Methods Initiative - Tools https://tools.digitalmethods.net