Interview on new edition of Data Journalism Handbook

As Liliana Bounegru and I are preparing the manuscript of the Data Journalism Handbook: Towards A Critical Data Practice (forthcoming, Amsterdam University Press), here is the full transcript of an interview we did about the book with the European Journalism Centre in December 2018, selections of which were published here and here.

Why another Data Journalism Handbook?

We were really lucky with the previous edition, which seemed a good fit for the moment in 2012. It has been widely translated and adopted as a core text on data journalism courses and trainings around the world. We designed it to age well, providing not just practical guidance on how to work with data, but also a diverse snapshot of the hopes and practices of data journalists at that particular moment in time.

The field of data journalism has travelled a long way since 2012. Not just because of more sophisticated technologies, but also because the social, cultural, political and economic settings of the field have changed. We’ve seen not only major initiatives like the Snowden leaks and the Panama Papers, but also debates and controversies around the role of data, platforms and digital technologies in society. Lots of tough questions have been raised about what data journalism is, who it is for and what it might do in digital societies.

Rather than just having a narrower focus on data practices, we take a broader look at these questions and consider what we might learn from them. In parallel to doing the first edition of the handbook, we’ve also gone through our graduate studies and into universities where we’ve been researching these kinds of questions about the societal implications of digital data, methods and platforms. The new edition of the book is our attempt to make sense of the field of data journalism and its changing role in the contemporary world, thinking along with a diverse mix of practitioners and researchers.

A lot has changed since 2012 in the data journalism landscape, what does it mean for the book and for you as editors?

To use a metaphor: there was a time when the number of readily available books in the world was small enough that it was not considered crazy for a librarian to try to build what they could consider a pretty comprehensive collection of books. Their acquisitions policy could just be: “whatever we can get our hands on”. In 2012, it might have not have seemed so crazy to, for example, have a go at making a list of data journalists and their projects from around the world. One very practical consequence for us as editors is that we cannot kid ourselves about our partiality: can only hope to cover a comparatively small number of the projects, practices and themes in this ever-growing field, and it was a difficult job to decide what to focus on in the book.

Also, as we’ve mentioned above, there have been many debates and controversies about the role and status of data journalism, which we’ve sought unpack and address in the book, as well as examining its relation to other developments. Since 2012 we have seen the rise of questions about the societal implications of technology precipitated by actors, events and issues as diverse as Snowden, Trump, Brexit, Bolsonaro, Xi Jinping, the Syrian civil war, Cambridge Analytica, Gamergate, #metoo, #IdleNoMore, Black Lives Matter, strikes and walkouts from tech workers, Uber riots, the European refugee crisis, Facebook’s News Feed algorithms, “fake news” and misinformation, the Gab platform and the rise of far-right populism and extremism. Several chapters in the book suggest how data journalists might attend to and position themselves in relation to such phenomena.

In 2012, there were not so many resources for learning data journalism. Now there are plenty more. How will this book be different from other data journalism resources out there?

There are indeed now so many useful resources (and there were too in 2012) – from practical tutorials and MOOCs on tools, code, visualisations, design and machine learning techniques, to “behind the scenes” blog posts and commentaries. Such resources will continue to evolve and are “best served fresh” from a variety of different sources.

The new edition of the book takes a step back from this more “hands on” material, and provides space to reflect on different aspects of contemporary data journalism practice. We have felt such a need for such a book when teaching data journalism on several courses – and in particular to complement more practical materials with practitioner perspectives as well as research perspectives from fields such as science and technology studies, media studies, internet studies and journalism studies. This provides a richer context for introducing data journalists’ particular blend of methods and approaches for working with data – examining not just the “how” but also the “what”, “where”, “when”, “why” and “for whom”.

How did you personally fall into data journalism? What fascinates you about it?

We both have a professional background in this area, and met at a data journalism conference which Liliana organised in Amsterdam in 2010. At that time Jonathan was working at an NGO called Open Knowledge International and particularly interested in data visualisations and the politics of public data (also going back to early 20th century experiments such as the Isotype Institute), at the same time as pursuing his graduate studies in London. Liliana was leading data journalism activities at the European Journalism Centre, at the same time as undertaking graduate studies at the Digital Methods Initiative at the University of Amsterdam (where she later became Managing Director).

We were both fascinated by emerging practices around how the world is accounted for through data – not just through established institutions such as national statistics bodies, but also through experimental and inventive modes of storytelling and participation with data in journalism and civil society. Such data practices would not just produce accounts of the world through data, but they’d also have to assemble publics for their work. We were also interested in how such projects would aspire to attend to collective life at scale, including environmental, social, political, economic and cultural dynamics beyond the to-and-fro of talking heads.

What is your personal motivation to work on the Data Journalism Handbook again?

There are so many things that we’ve enjoyed about this project – not least the fact that it has given us an opportunity to spend time immersed in this field learning about recent developments around the world, as well as trying to curate an interesting blend of perspectives from our contributors. We hope that the book inspires other to explore this field, and to look beyond the most well-known examples, to recognise emerging work and also the various actors and forms of labour involved in making projects succeed.

We also hope the book contributes to more conversations and collaborations between researchers, practitioners and students. So many data journalists also moonlight as instructors at universities, where they may encounter others to collaborate with. There are so many interesting perspectives on data journalism emerging in recent research. Researchers are also doing their own experiments in telling stories with data. We hope that some of the chapters in this book may inspire data journalists and researchers to experiment together and learn from each other, including around some of the challenges we mention in the introduction to the book.

In editing such a book, you have to make choices. How did you go about selecting the contributions? Which criteria did you use?

We’re not going to lie: this was a really, really difficult process as there is such an abundance of interesting projects, practices and perspectives that we wanted to include. Thus we felt compelled to do as much homework as we could. We spent time with as much material as we could – including trawling through all past entries to the Data Journalism Awards for the past several years; looking through as many outlets and portfolios as we could find; looking at links from the #ddj hashtag on Twitter; asking everyone we had contact with for suggestions for themes, contributors and projects. As we went we accumulated reams of field notes on what the handbook should attend to.

In social research there is this idea of the “snowball interview” where you ask interviewees for suggestions for others to interview. In a way our process could be described as a kind of curated “snowball editorial”, in that we tried to absorb all of the leads we could, and find all of the paths we could, before sending invites. We also approached the book in several waves, where we waited for material to come in, before considered what else we needed based on our evolving (and personal) map of the field. Throughout the process our mantra to ourselves was to consider how we were giving voice to a blend of perspectives, geographies, themes and genders.

Whom do you have in mind as the readers of this book?

We started by mapping out who was using the previous book as a starting point.

I think we ended up with eight main kinds of readers, but the three main ones appeared to be: data journalism students on programmes around the world (where it is adopted as a key reference point); practitioners looking for ideas or perspectives on data journalism (including but not limited to journalists); and researchers from a variety of backgrounds interested in making sense of the field.

We were also influenced by our own students, teaching activities and guest lectures to classes in Amsterdam, Berkeley, Copenhagen, London, New York, Paris, Utrecht and beyond. Graduate students on the data journalism programme at King’s College London played a role as imagined readers of the new edition with its focus on “critical data practice”, as this is an approach which was piloted on that course.

How did you ensure diversity in the book? What have been the challenges in including more “unheard” voices in data journalism (women, people of colour, from the Global South) in the book and how did you deal with them?

As mentioned above, our approach to the editorial was giving voice to a diversity of perspectives, geographies, themes and genders. Hence for the first stage in the editorial we were particularly looking for perspectives from women contributors and chapters giving perspectives from the global south. We were also keen to highlight different perspectives, methods and themes beyond those which were already prominent – including recognising not just established figures in the field, but also other forms of labour in this field (such as organising, training and social media).

Several of the chapters also explicitly address questions around inclusion, participation, voice and the politics of data journalism – including on gender and feminism (D’Ignazio; Vaca); colonialism (Young and Callison), giving voice to the marginalised (Constantaras), and “data sovereignty” for indigenous communities (Kukutai and Walter). Anita Say Chan’s chapter also addresses the “myth of digital universalism”: the notion that digital innovation flows from creative “centres” to derivative “peripheries”, calling for renewed attention to a greater diversity of sites and practices.

These are small steps, and there is much work to be done to improve diversity amongst both data journalism practitioners and researchers. As a caveat we’re also aware that we’re based in Europe, mainly mainly operating in English and other European languages and very much dependent on our friends and colleagues in the global south to help us to identify other voices in their communities. The book provides an inevitably partial picture, but we hope to inspire others spend more time learning about data journalism projects and practices around the world.

How is the book “collaborative”?

As we say in the introduction the book was intended to be a “collective experiment in accounting for data journalism practices and a collective invitation to explore how such practices may be modified”. We’ve been inspired by our friend and colleague Bruno Latour (one of the pioneers of science and technology studies and cofounder of the médialab at Sciences Po), who has been experimenting with the book as a format, including how various publics are invited to engage and give input around it.

The final version of the book will have been the result of many discussions, workshops, calls, meetings and many (!) emails with our contributors and others. The result is not just one frame (as one might expect in a textbook), but many different perspectives on the field of data journalism. We certainly had our own ideas about what we set out to do with the book, but its final form is a collective accomplishment.

Additionally we hope that the online beta will provide an opportunity to gather feedback on these chapters as well as the overall shape of the book, which may result in further discussions and modifications to the chapters before the print version is published.

What did you learn in the editorial process for the new edition of the Data Journalism Handbook?

If our editorial for the first book started with a collaborative “book sprint” in London, the second book has started with our curated “snowball editorial”, along with online discussions and an online beta version to test out the chapters before the final version is published. We will also be testing the full book in the classroom before the final version is published.

One of the questions that has accompanied us throughout the process is: what can we do with the format of the book? What are the advantages of this particular form, and how can we play to its strengths as one format amongst others? We could have contributed to a continuously updated forum with different discussion threads, a wiki, discussions around a hashtag, a collection of blog posts, or a series of workshops or events around the world – so why a book? And why a book that has a single published form rather than, say, a living book?

One of the nice things about a book, is actually that the contribution of a chapter requires a sustained period of care and reflection in constructing a narrative, an argument or an experience. We hope that the process has also provided space for some of our contributors an opportunity to articulate their thoughts in a way which they might not have otherwise have done. It is still a learning process for all of us, and we’re not done with the new edition yet – so perhaps we can get back to you!

What does ‘Towards a Critical Data Practice’ mean and why is it so important we take this angle in data journalism today?

We take “critical data practice” to be the task of modifying data practices in light of critical reflection (which we have previously written about here and here, inspired by artificial intelligence researcher Philip E. Agre’s notion of “critical technical practice” ).

Critical reflection may come from many sources and traditions. In the context of academic research this may include from fields such as critical theory, political economy, new media studies, science and technology studies, gender studies and critical race studies. Beyond academia, critique may be inspired by public discourse, social movements and activist groups. Such critical reflection may involve reflecting on the politics, power relations and broader social, political, cultural and economic settings of data practices.

As we discuss in the introduction to the book, there are many reasons why such critical questions about data journalism arise – and we see it as our task in the book to stay with these questions and consider how might suggest changes to how things are done. Sometimes this can involve asking simple questions like: “Where does data come from?”, “Who is data journalism for?” or “What kinds of participation, citizenship or relations does data journalism enable?”. As we show throughout the book, data journalists are responding inventively to such questions in many different ways – including through responding to a lack of data by making their own; investigating data, platforms and algorithms; or experimenting with visual or participation formats for inviting publics to make sense with data.

What are the 12 challenges for data journalism? And why do you leave them to the readers as open questions in the introduction?

The “twelve challenges for critical data practice” in the introduction are our own proposal for areas for experimentation that have emerged in the course of the editorial process and our associated research. They are intended in the spirit of “grand challenges”: large or different challenges which may require interdisciplinary teams, time and resources to address. They are areas that may benefit from collaborations between data journalists and other actors – whether universities, public institutions or civil society groups.

They are open-ended as they are intended to inspire experimentation, and many of them would be suitable as prompts for group projects on data journalism courses. Some of these things which data journalists are interested in, but may not always have the time or resources for, such as archiving their work or finding suitable ways to account for its impact. Some of them build on practices that are already described in the book: such as accounting for platforms and algorithms in a way which treats them not just as technical artefacts but as socio-technical systems which both shape and are shaped by cultural and social practices.

What was the most interesting/surprising thing you learned about data journalism while working on the book?

Drawing on researchers such as Susan Leigh Star, one of our starting points for the book was looking at data practices not just in terms of representations but also in terms of relations. In other words how a story about, say, segregation in US cities is not only an articulation of facts about demography and geography, but also how it assembles relations between census data, statistical operations, visual formats, code libraries, narrative techniques, editorial practices and social media audiences.

We’ve found this a fascinating lens – understanding data journalism as a kind of creative, curatorial craft or choreography in bringing different elements together in a project. It has led us to surface stories about the alliances, mobilisations, collaborations and networks that data journalism has built on, tapped into, or given rise to – such as noticing trees in cities, showing connections between products and colonialism, gathering communities to map land conflicts, supporting citizen journalists to help marginalised groups in China and many other things.

When will the next chapters be released and how? What about the actual book? Where can I buy it?

The full version of the Data Journalism Handbook will be released on Amsterdam University Press. It will be an open access book, thanks to support from the Netherlands Organisation for Scientific Research (NWO). Print copies will also be available for purchase, of course, distributed by Amsterdam University Press in Europe and by the University of Chicago Press in the United States. Translations will also be coming soon, as with the previous edition.

We know it might be difficult to say, but what’s your favourite contribution to the Handbook? And why?

We’re afraid this is the one question that we can’t answer: we couldn’t possibly pick favourites (not least because not all of the pieces have been finalised; at least that is what we can say to get us off the hook!). But we can mention a few things that might particularly stand out in the new edition in contrast to the previous book. The “doing issues with data” section which opens the book, acts as a kind of concise “taster menu”, and we’re particularly pleased with how this has turned out. Two of the themes that we were really keen to explore when we set out with background research for the second edition also have their own dedicated sections: “assembling data” and “investigating data, platforms and algorithms”.

As well as accounts from practitioners, we also feel really lucky to have benefited from a great series of reflections from researchers. For the online beta this includes renowned historian and philosopher of science Helen Verran with a piece on how data journalists might tell stories not just with numbers but also about numbers, and how they are made; Helen Kennedy et al’s chapter examines everyday engagements with visualisations; Chris Anderson looks at how data journalists might benefit from a deeper engagement with the development of their field; Wiebke Loosen highlights the importance of awards as a way to consolidate and institutionalise data journalism practices, as well as to understand what is considered valuable; and Nikki Usher offers a challenge to keep play and interactivity in the foreground.

What’s coming next for you (personally, professionally)?

At the moment Liliana is currently working on climate change misinformation at the University of Oxford, and Jonathan is working on his book on “data worlds” at King’s College London. Besides this, we are also continuing to experiment and collaborate around themes in the book at the Public Data Lab.

This includes, for example, encounters between data journalism and digital methods research, as discussed in Richard Rogers’s chapter. We are also exploring fact-checking through the lens of Noortje Marres’s work on public facts. To understand more about different approaches to doing participation with data, we’ve been working on a collaboration with the UN Foundation around citizen-generated data. Building on previous research projects, we’re very interested in telling stories with networks in different fields, and our lab is developing a new open-source, web-based tool for the visual analysis of networks. Finally we’re doing an updated version of our recent Field Guide to “Fake News” and Other Information Disorders, which has recipes for investigating not just misinformation but also the digital infrastructures and platforms through which it circulates and the digital cultures in which it resonates.

Some people say data journalists will eventually become regular journalists with good data literacy. What do you think of this view?

As you say, many people have emphasised the point that data journalism is still a form of journalism, and that perhaps the “data” part will become less important as data journalism practices become more widespread and more normalised. At the same time, we think it is worth considering when, why, how and for whom the distinction might matter, or not.

To use an analogy: when might it be useful to consider “photojournalism” as a distinct set of journalistic practices? There may be some cases where specificity is desirable (e.g. awards, associations, events focusing on specific kinds of practices) and some cases where it is desirable to be recognised as part of a broader field (e.g. legal protections afforded to journalists, normalisation and acceptance of particular kinds of work).

The value of such labels and distinctions cannot be established in the abstract, one must look at what they do in different settings. Along with other researchers, we’re interested in mapping the diversity of practices that have been associated with the “data journalism” label, and how this is used to advance, recognise and institutionalise different kinds of practices.

Data Journalists use a large variety of tools. How did you decide which tools the Handbook should discuss?

The handbook does not focus on tools as such. Specific tools enter the story insofar as they are relevant in accounting for different projects or practices. So this made it less difficult from an editorial perspective as we haven’t had to decide about tools, but instead have invited contributors to talk about different projects, practices, methods and settings – and tools have entered into these stories.

This is not, of course, to underestimate the importance of taking tools seriously. As our colleague Bruno Latour puts it: “Change the instruments, and you will change the entire social theory that goes with them.” The choice of whether to use a spreadsheet, a relational database, a network, a machine learning algorithm or a visualisation tool to work with data can shape what you notice and attend to, and the way you think about the phenomena you are investigating. We’re currently working on a research agenda at the Public Data Lab examining and comparing network practices in different fields – including their narrative capacities in journalism. Several chapters in the book look at techniques, and some look at the use of particular tools. But tools as such have not been the editorial starting point.

How can data help journalism fight the use of algorithms that support massive fakes?

We’re working on several projects on “fake news”, misinformation, manipulation and various forms of cultural production – and how journalists can respond. Quite a bit of attention has been paid to “deep fakes”, and their implications.

While this is a fast-evolving area, there are many ways that (data) journalists can respond. One of the areas that we’re particularly interested in is how to report on not just content, but also its social life and how people share, interact and make sense with it, including the platforms and infrastructures through which it circulates. We’ve provided several “recipes” for digital methods investigations into “fake news” in our recent Field Guide to “Fake News”, which has led to collaborations with BuzzFeed News and others. There’s a dedicated section in the Data Journalism Handbook on investigating data, platforms and algorithms.

Will the Handbook explain more about R and D3?

There are already many great resources online to learn how to use programming languages and libraries like R and D3. So instead we focus on how these have been used in the context of different data journalism projects to tell stories and to facilitate participation around different issues. As such the book is more focused on practices, projects, approaches, settings and issues, rather than on tools as such.

*What are the differences between the first Handbook and the second one? Or *why another Data Journalism Handbook?

We were really lucky with the previous edition, which seemed a good fit for the moment in 2012. It has been widely translated and adopted as a core text on data journalism courses and trainings around the world. We designed it to age well, providing not just practical guidance on how to work with data, but also a diverse snapshot of the hopes and practices of data journalists at that particular moment in time.

The field of data journalism has travelled a long way since 2012. Not just because of more sophisticated technologies, but also because the social, cultural, political and economic settings of the field have changed. We’ve seen not only major initiatives like the Snowden leaks and the Panama Papers, but also debates and controversies around the role of data, platforms and digital technologies in society. Lots of tough questions have been raised about what data journalism is, who it is for and what it might do in digital societies.

Rather than just having a narrower focus on data practices, we take a broader look at these questions and consider what we might learn from them. In parallel to doing the first edition of the handbook, we’ve also gone through our graduate studies and into universities where we’ve been researching these kinds of questions about the societal implications of digital data, methods and platforms. The new edition of the book is our attempt to make sense of the field of data journalism and its changing role in the contemporary world, thinking along with a diverse mix of practitioners and researchers.

Does have the book information about data journalism in Central América?

We could have organised the book geographically (e.g. by continent) but instead we opted to organise it by different themes (e.g. “assembling data”, “organising data journalism”).

There are indeed projects from Central America in the book, including one of the opening chapters included in the online beta, “From Coffee to Colonialism”, which features work in Guatemala and Honduras.

How much has data journalism changed since the first book? Or A lot has changed since 2012 in the data journalism landscape, what does it mean for the book and for you as editors?

To use a metaphor: there was a time when the number of readily available books in the world was small enough that it was not considered crazy for a librarian to try to build what they could consider a pretty comprehensive collection of books. Their acquisitions policy could just be: “whatever we can get our hands on”. In 2012, it might have not have seemed so crazy to, for example, have a go at making a list of data journalists and their projects from around the world. One very practical consequence for us as editors is that we cannot kid ourselves about our partiality: can only hope to cover a comparatively small number of the projects, practices and themes in this ever-growing field, and it was a difficult job to decide what to focus on in the book.

Also, as we’ve mentioned above, there have been many debates and controversies about the role and status of data journalism, which we’ve sought unpack and address in the book, as well as examining its relation to other developments. Since 2012 we have seen the rise of questions about the societal implications of technology precipitated by actors, events and issues as diverse as Snowden, Trump, Brexit, Bolsonaro, Xi Jinping, the Syrian civil war, Cambridge Analytica, Gamergate, #metoo, #IdleNoMore, Black Lives Matter, strikes and walkouts from tech workers, Uber riots, the European refugee crisis, Facebook’s News Feed algorithms, “fake news” and misinformation, the Gab platform and the rise of far-right populism and extremism. Several chapters in the book suggest how data journalists might attend to and position themselves in relation to such phenomena.

How will the Handbook help human rights reporting?

There are many chapters which are relevant to reporting on human rights issues. As well as touching on human rights-relevant themes, some of the techniques and approaches (e.g. assembling data, experiencing data, organising data projects) may be of interest to those reporting on human rights with data.

Back to posts