Project description

Since the “war against the drugs“ started in Mexico in 2006, the country has faced increasing violence. The homicides, the disappearance of people, and the findings of clandestine graves have become symbols of this violence, but authorities have failed to accurately inform the public how many graves there are in the country, how many people have been found there, and who they are.

With the use of public records and work on the field, this investigation reveals how, between 2006 and 2016, the finding of clandestine graves in Mexico spread across the country. The investigation also shows the lack of organization and shared standards by the authorities, which has repercussions for the families of the people that have disappeared, and the organizations doing their own searches.

What makes this project innovative?

The project compiles data that reveals what the authorities know and that provides a baseline for the magnitude of the humanitarian crisis. Government and human rights organizations have worked to document the graves; this project represents the most comprehensive public analysis at a national-level to date. It presents the appearance of clandestine graves and the number of bodies located in them with a national interactive map that shows time, geography and magnitude of the findings. The map takes advantage of powerful new geospatial technologies and the design represents a new variation on the “bubble map” format that encodings changes across time using color. The investigation combines narrative journalism and data visualization in sophisticated ways. The text and map complement each other by showing both the big picture and close-up examples of specific cases. The publication also includes 24 state-level maps with greater detail. For each state, we include more specific data like remnants and fragments of bones and craniums that are specific to each state’s public records. And we show the challenges of the FOIL process and provide detail on each state’s data collection.

What was the impact of your project? How did you measure it?

For the publication, we partnered with 15 national, regional and local media outlets in Mexico. By the end of the first week, 50 media outlets in Mexico and other countries mentioned the story and/or embedded our map. This map and the numbers that our investigation reveal serve as a reference for media and families looking for the disappeared relatives. In some states, families have used the investigation to demand more transparency to local authorities. The Mexican government modified the official number of clandestine graves when just few days after Andrés Manuel López Obrador took the presidency. In December 2018, Alejandro Encinas, the new Under-Secretary for Human Rights, recognized that there are more than 2,000 graves in Mexico at the International Human Rights Commission. Nevertheless, in February 2019, he went back to government’s original numbers, saying that there are more than 1,100, a number closer to those reported by the National Commission of Human Rights in September 2018. In December 2018, the team was invited to write an op-ed for the Washington Post, and on December 13th, the English version of the investigation and the national map was published in partnership with The Intercept (https://theintercept.com/2018/12/13/mexico-drug-war-mass-graves/). The investigation was also summarized in Italian and German. The project has had significant reach. Because the visualization could be embedded anywhere, some publications put it on every page of their website; we call these “passive” views as opposed to “active” views where the visualization is a primary part of the content. To date, the maps have had almost 200,000 unique “active” views and over a million unique “passive” views. We have been invited to present the methodology, findings and the maps to the Argentine Forensic Anthropology Team, the National Search Commission, the International Committee of the Red Cross, the International Commission on Missing Persons, as well as different Mexican and US universities like the University of Chicago along with research groups.

Source and methodology

To collect the data, the team sent FOIL requests to the 32 state attorney's offices in the country and PGR (General Attorney). 24 states recognized the findings of clandestine graves. In the requests, we asked for: 1) number of clandestine graves, when and where were they found, 2) number of bodies, what studies have been practiced for identification, 3) number of remnants, what studies have been practiced for identification, 4) how many people have been identified and how many people have not. The public records received presented problems like denials, incomplete information, inconsistencies, and contradictions, and we found out that state authorities do not use the same categories. In the cases where it was necessary, the team sent follow-up requests to clarify information and to avoid duplication. In total, we submitted around 200 requests and appeals (recursos de revisión). We put together the information in a complete database organized from the minimum information such as the number of graves and bodies found at a municipal level per year, to the more specific data like remnants, craniums and fragments, and number of people that have been identified. For the maps, we chose to visualize the number of graves and bodies found at the municipal level per year. (Municipalities in Mexico are similar administrative units to counties in the US and UK.) The maps were designed to be embedded and shared. Publishers need good visuals to accompany their reporting, and when they embed, it provides exposure for the project and its findings. At least 14 publications have embedded one of the maps. The mapping and data visualization were also central to social media images and motion graphics that helped spread the findings across platforms. We decided to work mostly with the data presented by state governments and to contrast it with the federal level information provided by PGR. We also relied on open geospatial data from the National Institute of Statistics and Geography (INEGI). The design decisions in the final editorial product were possible because of the on-going conversations that combined the on-the-ground experience covering the consequences of violence in Mexico and with the meaning and challenges of the data itself. Questions such as whether the data could or should be shown on a national map were the result of a dialogue between these disciplines and reporting experience.

Technologies Used

The data processing web presentation is powered by open source technologies and is itself open source (https://github.com/eads/fosas-gatsbyv1). We use Google Sheets, GNU Make, Python processing scripts, and PostgreSQL / PostGIS for data management. Google Sheets made data entry accessible and easy to check. A reproducible data loading and analysis pipeline combines the database created by the reporting team with open geospatial data published by the National Institute of Statistics and Geography (INEGI). The pipeline let us quickly iterate and validate our analysis. Web publishing is powered by GatsbyJS in order to create pages and components that load quickly and work well on mobile devices with limited bandwidth and/or processing power (over 70% of traffic came from mobile, most of it from Android devices). The front-end makes extensive use of React and MapboxGL to visualize the data and provide interactivity. We use NPR’s Pym.js to provide responsive embedding.

Project members

Team: Alejandra Guillén, Mago Torres, Marcela Turati, David Eads, Erika Lozano, Paloma Robles and Aranzazú Ayala. Contributors: Juan Carlos Solís, Alejandra Xanic, Ignacio Rodríguez Reyna, Gilberto Lastra, Mayra Torres, Pedro Zamora, Carlos Juárez, Daniela Rea, Carlos Quintero, Rodrigo Caballero, Mónica González Islas, Pedro Pardo, Félix Márquez, Queso, Rafael del Río, Ruth Muñiz.

Link

Additional links

Followers

Click Follow to keep up with the evolution of this project:
you will receive a notification anytime the project leader updates the project page.