Students make the news by producing original data journalism

In the framework of the master’s course, Critical Data Studies, students had the opportunity to work with the data journalists Paul Ronga, of Tamedia, and Cécile Denayrouse, of the affiliated Tribune de Genève, in order to develop articles for publication in Le Matin Dimanche, 24 Heures and La Tribune de Genève.Their articles published on July 4th and today, address pressing issues with broad relevance: how Covid-19 has changed spending patterns and the carbon footprint associated with having a baby. Preparing these articles under the guidance of Ronga and Denayrouse allowed the students to explore what it means to work with social data.

Public outreach is a core objective of the Critical Data Studies course. Its interdisciplinary teaching team includes Christine Choirat, from the Swiss Data Science Center, Selim Krichane, GameLab UNIL-EPFL co-founder and coordinator of the TILT courses, Semion Siderenko, a machine learning engineer who served as TA and Charlotte Mazel-Cabasse, who has a background in Science and Technology Studies and directs the UNIL-EPFL dhcenter. Together, they devised a pedagogical model that blends analytical perspectives from data science and social and human sciences with hands-on engagement beyond EPFL. Students spent the spring semester of the year-long course doing a group project to learn how data is used in different concrete settings, in dialogue with outside partners. These partnerships covered a wide range of institutional contexts, from museums to start-ups, and they focused on diverse themes, including railway networks, social-media propaganda, and domestic violence.

Two groups of students opted to do projects on data journalism, under the guidance of Denayrouse and Ronga. Both groups included a mix of different disciplinary perspectives, including one UNIL student in digital humanities (this is one of several TILT courses in EPFL’s Social and Human Sciences curriculum that are also open to UNIL students).

The first group – Ghali Chraibi, Manon Michel, and Guillaume Parchet – developed a calculator to allow parents to assess the carbon footprint of a baby. “The goal,” Michel explains, “is not to say, ‘you should not have a baby because it is bad for the environment’ but rather to see the choices that one can make to reduce that impact.” Rather than presenting a single optimal model for eco-friendly parenting, the group offered a tool that allows people to evaluate the relative ecological impact of discrete choices, for example whether to breast-feed or use formula. They also contextualized individual choices in a broader social context in order to acknowledge factors that may limit access to some lower-carbon options such as washable diapers.

The second group – Axel Matthey, Axelle Piguet, Jan Briachetti, and Matthias Zeller – analyzed how Covid-19 has changed the way people use money. They were able to take as their baseline a previous survey of spending habits conducted before the pandemic by the Tribune de Genève and several other Tamedia outlets. They then published a new survey last spring in order to gather a second data set indicating how patterns have changed over the course of the past year. Their recent article presents an assessment of the results that highlight shifting attitudes towards money but also other behavioral changes such as increases in smoking and drinking.

Ronga and Denayrouse guided the two project groups through the wide world of data journalism. Mazel-Cabasse explains that ideally, the students would have been able to immerse themselves in a newspaper’s editorial room for a few days in order to see first-hand the organizational and intellectual processes behind journalism. Instead, “all that work to understand an environment was only channeled through the mediation of Paul and Cécile.” Their core task was to help the students grasp data journalists’ responsibilities towards the general public. Denayrouse emphasizes that the students had to learn to “seduce the reader and write to be read.” This was a big intellectual leap to make from academic writing for an audience of like-minded specialists. The students “had to adopt an entirely different way of thinking.”

The students welcomed this change of pace. Both Chraibi and Matthey emphasize the value of leaving the “inward-oriented academic community” in order to try to reach a wider public. Piguet concurs that journalism offered “a different way of seeing data.” She observes that “normally when we are using data, there is really a partition between the public and the data whereas here…this is data for the public, interpretations for the public, not for colleagues.”

Ronga similarly reflects, “this relationship is very enriching for us because it involves an entirely different approach. As a journalist, you are normally working under severe time constraints. Here, we were in a different temporality, and it was also a different perspective.” Because the students prepared a single article over a full semester, they were able to delve into more detail in their background research than a journalist would normally be able to do, but this also entailed pitfalls. Parchet recalls that “as soon as we started to dig into something, we could go very deep very quickly…there are many factors, for example geographic considerations about where a diaper comes from, that can quickly make our analysis too complex.” Thus, the students had to learn to summarize a multilayered context for a general audience while remaining faithful to the underlying data. In short, they had to learn to practice the craft of data journalism.

Opportunities to blend a technical formation with journalistic experience are still fairly rare in Switzerland, where there is not yet widespread training for data journalists (although Ronga and Denayrouse do offer workshops). Ronga notes that most current practitioners in the field have improvised their own “hybrid” training in journalism and computer programming, and they form “a niche profession” within the world of Swiss journalism. Denayrouse laments that data literacy is not more common among journalists who are not specialists in this domain. Many journalists “have trouble leaving their comfort zone” which means that data journalists remain “on the margin.” Mazel-Cabasse observes that this is a broader problem that can be seen across other areas of society:

"Today we effectively have a system where people who do data are on one side and people who don’t are on the other, and one can expect that in the future there is going to have to be more hybridization …What Paul and Cécile are doing today for these students is helping them to understand how to engage with people who do not come from a data background…there is something absolutely fundamental at stake here: figuring out how we will be able to achieve an intersection between people who already have a data culture and those who do not, yet."