Data-Driven Journalism

Covid-19 has brought data and statistics to the centre of daily life like never before: everything we do in response to the virus at individual, organisational and societal levels depends literally on what the numbers tell us. In March 2020, as the exponential virus loomed large into an existential threat, an influx of numbers that would normally stay within the domain of specialist expertise suddenly occupied the physical and cultural space of the lockdown family. “Scary” concepts – R-naught, infection rate, transmission rate, death rate, excess deaths, false positive, false negative, relative risk, absolute risk, random sampling, statistical modelling and so on – abounded everywhere, from the TV and computer screens to husband-wife or even parent-child musings.

With that came a statistical chaos that continues, albeit to a lesser extent, today. Associated with such numbers is an unprecedented level of complexity and uncertainty due to the novelty of the virus. But as they become so crucial, Covid-19 data have been subject to a rather fierce battle between different frames and narratives, in which scientists dispute with each other as well as compete – not always successfully – with religion, culture and, most importantly, politics. Amidst much public confusion, anxiety and fear, numerical misinformation and disinformation seem to be everywhere on social media.

What does all this mean to journalism, a profession that is rarely commended for their ability to engage and deal with numbers? What are the major challenges? How have journalists performed in in questioning, scrutinising and communicating Covid-19 data, including debunking statistical “lies and damn lies”? What methods, techniques and platforms do they use to obtain, unpack, portray and deliver Covid-19 data and statistics to help people make sense of the pandemic? To what extent can their work change hearts or alter minds? What can the media learn from the first year of news reporting of the virus?

On 4 December, 2020, twenty-four senior science, health and data journalists, scientists, statisticians and media scholars gathered for to reflect on the above at a one-day conference jointly organised by Bournemouth University, the Royal Statistical Society and the Association of British Science Writers. The event, called Coronavirus, Statistical Chaos and the News Symposium, was a platform for journalists and scholars to share ideas and exchange experience around the above questions. This report summarises key issues, challenges and lessons learnt from the first year of British news reporting of the pandemic.

The rise of user-generated content (UCG) as a source of information in the journalistic lifecycle is driving the need for automated methods to detect, filter, contextualise and verify citizen reports of breaking news events. In this position paper we outline the technological challenges in incorporating UCG into news reporting and describe our proposed framework for exploiting UGC from social media for location-based event detection and filtering to reduce the workload of journalists covering breaking and ongoing news events. News organisations increasingly rely on manually curated UGC. Manual monitoring, filtering, verification and curation of UGC, however , is a time and effort consuming task, and our proposed framework takes a first step in addressing many of the issues surrounding these processes.

For journalism the phenomena of ‘big data’ and an increasingly data-driven society are doubly relevant: First, it is a topic worth covering so that the related developments and their consequences are made understandable and debatable for the public. Second, the ‘computational turn’ has already begun to affect practices of news production and is giving rise to novel ways of identifying and telling stories. What we observe as a result is the emergence of a new journalistic sub-field often described as ‘computational/data journalism’. This study focuses on the output of data journalism; by using a classic 'handmade' standardised content analysis methodology it aims to contribute to a better understanding of its reporting styles. The sample consists of all the pieces that were nominated for the Data Journalism Award (DJA) – an award issued annually by the Global Editors Network – from 2013 to 2015 (n= 179). Our categories of analysis look at, amongst other aspects, data sources and types, visualisation strategies, interactive features, topics, and types of media outlets nominated. Results show that over 40 percent of the data-driven pieces were published on the websites of (daily or weekly) newspapers and nearly 20 percent came from non-profit organisations for investigative journalism like ProPublica. Almost half of the cases cover a political topic, and social, economic as well as health and science issues appear frequently too. Financial data and geodata are the types of data used most often and most of the data relates to a national context. More than two-thirds of the projects use data from an official source like Eurostat. Further analyses regard the differences between 2013, 2014 and 2015 and look deeper into visualisation strategies and interactive features.

DOWNLOAD LINK: https://revistadaanpoll.emnuvens.com.br/revista/article/view/1307

Abstract: The concept and practice of data literacy is approached in order to characterize it as an instantiation of the  transdisciplinary of literacy (s) in a more general sense, and, based on this characterization, questions and hypotheses are presented  f a research project focused on data journalism. We place the discussion in the context of the digitized datafication of social activity, arguing that it affects the formulation of opinions and arguments based on journalistic texts that characterize the exercise  f citizenship in the public sphere. Our bibliographical research has shown that, although the conceptualizations of data literacy are predominantly functional and/or skills-based, there are also proposals based on the ideological/sociocultural model of literacy. We  onclude by hypothesizing that, in data journalism practices, there would be a tendency for qualitative interpretations to be subordinate to quantitative knowledge, but it may be possible to create situations in which disciplinary shortcomings show the  ransdisciplinary nature of the texts and the events for the producers and readers themselves. Keywords: Literacy; Transdisciplinary;  atafication; Data Literacy; Journalism

Resumo: Aborda-se o conceito e a prática do letramento de dados, visando caracterizá-lo como instanciação da transdisciplinaridade do s) letramento(s) em sentido mais geral, apresentando-se, a partir dessa caracterização, perguntas e hipóteses de uma proposta de pesquisa focada no jornalismo de dados. Situamos a discussão no contexto da dadificação digitalizada da atividade social, argumentando  ue ela afeta a formulação de opiniões e argumentos baseados em textos jornalísticos que caracteriza o exercício da cidadania na esfera  ública. Nossa pesquisa bibliográfica evidenciou que, embora predominem concepções de letramento de dados de natureza funcional e/ou  aseada em habilidades (disciplinares), há também propostas baseadas no modelo ideológico/sociocultural do letramento. Concluímos apresentando a hipótese de que, nas práticas de jornalismo de dados, haveria uma tendência à subordinação das interpretações  ualitativas aos saberes quantitativos, mas que se podem criar situações em que as lacunas disciplinares evidenciam o caráter  ransdisciplinar dos textos e dos eventos para os próprios produtores e leitores. Palavras-chave: Letramento; Transdisciplinaridade;  adificação; Letramento de Dados; Jornalismo

Die Entstehung des Datenjournalismus kann als Antwort des Journalismus auf die „Datafizierung“ der Gesellschaft verstanden werden: In der Auseinandersetzung mit dem Phänomen „Big Data“ entwickelt der Journalismus neue Wege, Geschichten (in Daten) zu identifizieren und (mit Daten) zu erzählen. Von Journalismusforschern wie -praktikern wird das dabei entstehende Berichterstattungsmuster häufig als die Zukunft des Journalismus angesehen, vor allem aber als genuin für den Onlinejournalismus. Aus Sicht der Forschung ist jedoch noch nicht einmal der Status Quo des ohnehin sehr dynamischen Datenjournalismus geklärt: Noch wissen wir wenig insbesondere über das, was den Datenjournalismus als Berichterstattungsmuster ausmacht, nämlich die datenjournalistischen Beiträge sowie die spezifischen inhaltlichen und darstellerischen Elemente, aus denen sie bestehen und die sie von anderen Formen der Berichterstattung abgrenzen. Der Schwerpunkt dieser Studie liegt daher auf den Produkten des Datenjournalismus: Mithilfe einer standardisierten Inhaltsanalyse untersuchen wir datenjournalistische Projekte, die man als „Goldstandard“ datengetriebener Berichterstattung betrachten kann – Projekte, die von 2013 bis 2015 (N = 179) fur einen der Data Journalism Awards (DJA) nominiert wurden, einen jährlich vom Global Editors Network in unterschiedlichen Kategorien vergebenen Preis. Untersucht wurden unter anderem die genutzten Datenarten und -quellen, Visualisierungsstrategien, Interaktionsmöglichkeiten, die behandelten Themen sowie die Medienangebote, von denen die Beiträge stammen.
Die Ergebnisse zeigen unter anderem, dass uber 40 Prozent der datengetriebenen Arbeiten auf Webseiten von Tages- oder Wochenzeitungen veröffentlicht wurden und fast 20 Prozent von Non-Profit Organisationen für investigativen Journalismus wie ProPublica stammen.
Fast die Hälfte der Fälle behandelt ein politisches Thema. Auch Gesellschafts- und Wirtschafts- sowie Gesundheits- und Wissenschaftsthemen kommen vermehrt vor. Finanz- und Geodaten sind die meistgenutzten Datenarten, wobei sich der Großteil der Datensätze auf eine nationale Ebene bezieht. In mehr als zwei Drittel der Projekte werden die Daten einer offiziellen Quelle wie zum Beispiel Eurostat verwendet. In weiteren Analysen werden die Unterschiede zwischen den Jahren 2013, 2014 und 2015 herausgearbeitet und Visualisierungsstrategien sowie Interaktionsmöglichkeiten untersucht.