‘Shining a Light Where There Are Shadows’: Latin American Outlets Innovating With Data

3 months ago 30

From mapping the impact of organized crime to diving into the behavior of extractive companies, and from digging into embezzlement and public contracts to investigating the impact of climate change, data journalism is helping outlets across Latin America carry out innovative projects that reveal the stories hidden behind large volumes of data.

It has enabled new formats, spurred on creative storytelling, and sparked creative cross-border collaborative projects such as NarcoFiles and Digital Mercenaries. It has also led to the development of new tools such as the “Funes” algorithm from Ojo Público, a Peruvian outlet that uses data journalism to explore hidden realities.

“We really understand that the data is not something floating in space, but rather that it must be integrated with the story.” — La Data Cuenta founder Hassel Fallas 

According to Sandra Crucianelli, the coordinator of the data intelligence unit at Infobae in Argentina, investigative data-driven reporting in the region is helping reporters “shine a light where there are shadows.”

And while there are challenges — in terms of resources, training, and access to information — the results are impressive. “We are going through a boom in independent journalism of the highest quality,” says Hassel Fallas, a data analyst and founder of La Data Cuenta from Costa Rica.

The Art Of Building Stories With Data

Across the continent, there are innovative success stories: In Brazil, InfoAmazonia — an outlet that specializes in using data to report on isolated areas in the Amazon — has used data visualizations and mapping to publish reports about deforestation, drug trafficking groups that threaten communities, and guerrilla fighters that recruit Indigenous young people for their ranks.

In Peru, the team at Convoca has carried out a huge data analysis for its series Excess Without Punishment. This multi-part exposé intertwined data journalism, on-the-ground reporting, and freedom of information requests to explore the behavior of the extractive industries in the country and the impact on the lives of residents of the Andes and the Amazon.

In Paraguay, El Surtidor took to the streets for data collection: using sensors to record the body temperatures of food delivery drivers for their investigation What it’s Like Working in One of the Hottest Cities in South America.

One of the strengths of journalists working across this region is making stories about data relatable. “We really understand that the data is not something floating in space, but rather that it must be integrated with the story,” Fallas says.

And combining narratives with data, while maintaining a strong focus on the storytelling, is a particular specialty across the region, she adds.

“In Latin America we humanize investigations that are related to data. We do not overwhelm people with a very large amount of information, we know how to prioritize data analysis, and what is relevant to the investigation,” Fallas explains. “We don’t focus as much on the form as on the substance.”

Alongside Crucianelli and Fallas, GIJN spoke to Luis Enrique Pérez Pinto, project coordinator of Convoca’s Deep Data project in Peru; Daniela Guazo, a data journalist at El Universal in Mexico; and Tai Nalon, executive director of Aos Fatos in Brazil about innovative projects they have worked on and the challenges they face producing data journalism from the continent.

From Investigating the Military Dictatorship to Scouring Public Databases: Finding Stories in the Data at Infobae

A map of the different districts of Buenos Aires, from a secret document in which the military junta set out key infrastructure like distilleries, electrical plants, and water supply stations in the months before they lost power. Image: Screenshot from the Infobae investigation

Decades after the fall of the military dictatorship in Argentina, the team at Infobae was able to dig into the clandestine diktats of the country’s former leaders — downloading 7,000 documents that had never been processed and analyzed in their entirety.

The resulting investigation — The Secret Decrees of the Dictatorship — revealed the arrest warrants ordered by the executive branch, details of deportations and news censorship, plus figures on the arms trade. Analyzing the data, reporters found that the year 1977 saw the highest number of secret decrees, at 1,212, with the military junta justifying the secrecy under the auspices of “consolidating peace and preserving the interests of the Republic.”

The work, led by Crucianelli and Mariel Fitz Patrick, involved qualitative and quantitative research. “What we did was process them in a database and look for patterns, repetition, and different aspects that allowed us to know details about the loans that had been taken during the dictatorship, the books that had been banned, the people who had been prohibited from leaving the country, and other lines of work,” explains Crucianelli.

The project won them the National Award from the Forum of Argentine Journalists (FOPEA) in the category of investigative journalism, and showed the value of digging into the secrets of the past for investigative stories.

Data remains central to Infobae’s production, and with two weekly data stories, they are among the top producers of data-led reporting in the region.

To maintain this publishing rate, the team systematically searches a large list of national and international databases for new material, and take an “aggressive” approach to requesting information.

While open data has given journalists access to sensitive information, such as public spending records, budget figures, official statements, and salary details, “there are outstanding debts when it comes to governments,” notes Crucianelli, particularly in terms of quality and data being published “in a timely manner.”

Using Data to Investigate Corruption, Embezzlement, and the Environment at Convoca 

An interactive map in which readers can explore how public funds have been spent in relation to helping people exposed to toxic heavy metals. Image: Screenshot of the investigation by Convoca

One of the first big data projects of the Peruvian outlet Convoca was the regional blockbuster Lava Jato — translated as Operation Carwash — the result of collaborative work by journalists from Latin America to investigate one of the largest corruption cases in the region in recent times.

The investigation, which delved into a cross-border bribery scheme centered on Brazilian companies, saw Convoca journalists carry out data analysis to see cost overruns and to allow reporters to follow the money.

Since then, the outlet has used data to explore environmental topics, such as in The Illegal Route of Peruvian Seaweed, an investigation with Connectas and the International Center of Journalists (ICFJ), and Toxic Legacy, where reporters delved into the impact of toxic heavy metals on children and other people exposed to substances like lead and mercury.

Recently, they have been using data to develop new formats, creating Games For News, which they say is one of the first interactive video game platforms of public interest in South America.

It was used to create the videogame Truths vs. Lies of the Truth Commission, in which users try to identify if speeches purportedly from the final report of the country’s Truth and Reconciliation Commission — related to the country’s internal armed conflict — are true or false.

Pérez Pinto says improving the user experience is key to attracting audiences, and a commitment to which the media should pay more attention. He highlighted the examples of data units in other regions that bring together projects with advanced visualizations, interactive elements, and impressive designs as inspirational.

But he says, for all of these projects, being able to access data is vital and reporters are increasingly having to grapple with “bureaucratic barriers.”

“Then as journalists we have to go to the Transparency Court, and sometimes, they don’t give you a reason,” for granting or denying your request for data, he notes.

Data Systematization and Databases: the Case of El Universal  

El Universal created interactive maps to explore the crisis of missing people in Mexico and Colombia. Image: Screenshot

In Mexico, reporters have been using data to help with in-depth investigations into one of the country’s most challenging problems: organized crime.

An early success was the project Disappeared, a collaborative multimedia project designed to tell the stories of the victims of organized crime and human trafficking in Mexico and Colombia, and detail some of the stories of the more than 25,000 people who have gone missing in the last decade. The Mexican daily newspaper El Universal joined with Colombia’s El Tiempo to carry out this project that won the Ortega y Gasset Award for the best multimedia coverage in 2016.

In Mexico, Country of the Cartels, the team created a system to analyze and review a leak of emails containing intelligence reports from the Mexican Secretary of National Defense. “Those documents told you which criminal organizations were in different states and municipalities,” explains Daniela Guazo. Using the data, the team was able to reveal that the country has more than 80 organized crime groups and around 16 criminal gangs, data which they used to create an interactive map.

More recently, in collaboration with Connectas, the team worked on Predators in the Classroom, an investigation into child abuse in schools based on data requested from 32 state agencies across the country. After receiving the information, processing and analyzing it, the team was able to create a database with more than 3,000 alleged attacks on minors within schools over a period of 10 years. The investigation won the King of Spain Award for International Cooperation and Humanitarian Action this year.

“We are moving forward, not yet on a scale like media outlets in the United States, Europe… but according to our context,” Guazo says.

But investigative data reporting comes at a price. Not only is technology and equipment expensive, but personnel are critical as well.

“Data journalism costs and costs dearly,” notes Guazo, adding that reporters are also battling a complex panorama of low salaries, threats from organized crime, and even push back from governments.

Using Data to Explore Topics that Matter to New Audiences: La Data Cuenta

A graphic exploring how temperatures have increase in Costa Rica, published as part of the climate change project. Image: Screenshot from La Data Cuenta

La Data Cuenta literally means “the data matters.” And since it was launched the platform has developed a name for itself exploring topics like climate change, migration, human rights, and gender inequality through data.

One project — This Is How Climate Change Affects You — used data, graphics, and research to show how climate change impacts Costa Rica, where the team is based. It involved analysis and visualization of data, and allowed the outlet to touch on a topic that matters to younger readers.

“There are countries where a government promotes open data or transparency, and suddenly another government comes in, with another ideology, and locks down the information, or makes it more difficult.” — Hassel Fallas

“That project was key for several things [including] to connect with the audience that was following us, mainly young people between 18 and 25 years old,” says Fallas. “It was an important project because it allowed us to explain the issue of climate change from its different angles and phases, with data, stories, and examples from everyday life.”

Meanwhile, their climate finance investigation Show Me the Money! unraveled the way in which the global financing system for climate change operates, and how the system is — according to the reporting — trapping the region with “an excessive dependence on climate credits.”

“This investigation opened the door for us to make alliances with other outlets such as the Center for Investigative Journalism in Puerto Rico,” Fallas says. She adds that the outlet is currently focusing its energies on regionalization — seeking to create content that is relevant beyond Costa Rica — while maintaining their focus on climate change and gender, two topics that were not high on the media agenda previously.

But, she said, government changes have made getting access to data more difficult, with political ideologies influencing reporters’ ability to get their hands on the relevant information. “There are countries where a government promotes open data or transparency, and suddenly another government comes in, with another ideology, and locks down the information, or makes it more difficult to obtain it,” she said.

How Brazil’s Aos Fatos is Using Language Models to Tackle Disinformation 

Radar Aos Fatos, a multiplatform disinformation monitor that analyzes thousands of posts on digital platforms such as WhatsApp. Image: Screenshot

Aos Fatos — an investigative and fact-checking organization that uses technology to tackle misinformation — was created in 2015. Data and technology have always been key to their strategy, and that is why, from the beginning, the team structured their content management system so that they could organize their reports by topics, sources used, and common terms to allow them to integrate different projects.

“Obviously, without data there is no data journalism.” — Tai Nalon, co-founder and executive director of Aos Fatos  

This has allowed them to develop projects such as the chatbot Fátima, which operates with a large language model and maps the entire Aos Fatos archive to respond to the audience. “Hi!” Fátima tells readers, “I am here to help you verify if information is true or not. To interact with me, ask a question or send a text message that you would like to have verified.”

The multidisciplinary team also created Radar Aos Fatos, a multiplatform disinformation monitor that analyzes thousands of posts on digital platforms such as WhatsApp, Facebook, Instagram, and YouTube in search of common linguistic patterns in disinformation campaigns. That is, it automatically processes data from the platforms and generates another database, filtered according to the topic, with which its journalists can work.

Radar won the team the Gabo Award in 2020, where judges said the project “makes outstanding use of artificial intelligence to detect and display trends and patterns of misinformation in social networks.”

“We are very proud of it, but it depends on the data provided by the platforms,” says Nalon, the co-founder of Aos Fatos. “As they increasingly restrict access to their data, it becomes more difficult to keep it up and running. Obviously, without data there is no data journalism.”

She is concerned about the impact of AI on journalism and the possibility that data journalism becomes more economically and technologically dependent on large technology companies based abroad, and AI models developed from a global north perspective.

Accessing reliable data has also become more difficult across the board, said Nalon, with newsrooms increasingly smaller and with fewer professionals available to demand transparency and use mechanisms such as Freedom of Information legislation to promote a culture of public data.

Innovative reporting projects, she says, will always carry an investment risk, something that is “contrary to the corporate culture of our times, which favors immediate rewards.”

But there is still room for optimism. For his part, Pérez Pinto suggests a future in which universities collaborate with news outlets by providing training in visualizations, the creation of news applications, and interactive games. Collaboration across disciplines, he hopes, can help spread knowledge and increase the scope of investigations.

And in Argentina, Crucianelli is also quietly optimistic. “Data journalism is going to grow in the coming years much more than it has grown up to now,” she says. “The challenge is for the media to find the necessary resources.”



Lucero Hernández García is a freelance journalist and digital consultant from Mexico, and GIJN en español collaborator. She has a masters degree in Communication and Digital Media, with a specialism in multimedia production. She runs workshops and teaches data, visualization, digital tools, and online journalism to university students. Her work has been published by IJNet, and she has received scholarships from Cosecha Roja, Sembramedia, and the Thomson Reuters Foundation.

Read Entire Article