Skip to Content, Navigation, or Footer.
Support independent student journalism. Support independent student journalism. Support independent student journalism.
The Dartmouth
June 17, 2024 | Latest Issue
The Dartmouth

Spotlight on The Early Warning Project: A Force of Statistical Risk Assessment Change

One writer investigates how the Early Warning Project is a force for statistical change in the field of genocide prevention.

Earlier this month, the Early Warning Project released their ninth annual Statistical Risk Assessment for Countries at Risk for Mass Killing. The project, a collaboration between the Dickey Center for International Understanding and the Simon-Skjodt Center for the Prevention of Genocide at the United States Holocaust Memorial Museum, seeks to identify the early warning signs of genocide and mass atrocities, through data and research to hopefully prevent them, according to government professor and chair Ben Valentino. 

Development of the project began in 2011 during Valentino’s fellowship at the Simon-Skjodt Center. Valentino, who also serves as the faculty director of the War and Peace Studies Fellows Program at the Dickey Center, has been producing statistical models and forecasts of genocide and mass atrocities against civilians since the mid-1990s. He had previously worked for the United States government, developing similar risk assessment models for the State department, foreign governments and non-governmental organizations. 

Valentino said his work for the U.S. government quickly became of interest to the Holocaust Museum’s Simon-Skjodt Center.

“It made sense that the public should have access to something like the model that I was doing for the U.S. government,” Valentino said.

The collaboration resulted in the creation of the Early Warning Project and, with help and funding from the Dickey Center, was publicly released in 2015. 

The Dickey Center’s support has been instrumental in collecting new data for the Early Warning Project, as well as student support, according to Valentino. 

“We had students here at Dartmouth at the Dali Lab produce a prototype webpage for us that became this webpage,” said Valentino. “Over the years I’ve had student assistants who work there.” 

While there are currently no student assistants working on the project, Prescott Herzog ’25 attested to the Dickey Center’s focus on supporting global issues like the issue of genocide and mass killings. 

“The Dickey Center’s main idea is that the world’s issues are your issues,” Herzog said. 

The Early Warning Project utilizes 38 different variables to produce forecasts for every country in the world with a population of over 500,000 people. These forecasts indicate what each country’s annual risk of having a mass violent event against civilians is. 

Ashleigh Landau, research assistant at the Simon-Skjodt Center, oversees the daily work of the project by managing consultants and fellows. She highlighted that there were two main pillars of the project: one being quantitative and the other being qualitative. 

“The project consists of quantitative work, and that’s the statistical risk assessment with the report that just came out, but we also do qualitative work … where we actually do a deep dive into a specific country and really try to understand the risks for mass atrocities and the on-the-ground dynamics,” she said. 

One of the issues encountered early on in the project was the selection of the variables that would be used for the statistical model itself. Data can be difficult to find and to collect, making the selection of variables for the project very slim. Landau stated that the project utilized “The Fundamentals of Genocide Prevention,” a book by Scott Strauss, as a reference for deciding which variables were necessary to consider.

“This book is so great because [Strauss] basically presents the most widely acknowledged risk factors for genocide and atrocities,” said Landau.

Despite occasional  challenges with the model, Valentino stated that the goal of the project is firm: “to try to give decision makers, both in government and in non-governmental organizations and intergovernmental organizations like the UN, some warning that particular countries might be at risk.”

Landau corroborated this, adding that the hope is for the project to be “one additional tool for [decision makers] … to prevent atrocities.” 

Furthermore, Valentino identified three conditions that the data required to be included within the model. 

“In order to make it into our model, we need data that goes back all the way to 1955,” he said. “We need it to be annual, so it needs to be updated every year since then. It needs to be for every single country around the world. That’s a pretty high demand.” 

However, the variables chosen for the EWP provide great insight into possible triggers or warnings of mass violence. Valentino noted that one example of an important variable is GDP, as poorer countries are at a greater risk of mass violence than those with a higher GDP.

Another variable utilized is infant mortality, which Valentino identified as one of the strongest predictors of mass atrocities.

“That’s not because infants are dying in mass atrocities, for the most part,” he said. “Instead, we think it’s likely an indicator of a government that’s not investing very much in the welfare of its people.”

Within this broad goal, the EWP aims to promote two specific actions from organizations. First, they hope organizations will invest more resources into countries at greater risk to discover the issues within these nations. Second, they want organizations  consider intervention plans to prevent genocide and atrocities. 

“The idea is hopefully to get [involved] earlier because we know from studying atrocities that it's so much better to get [involved] as early as possible to prevent them from occurring in the first place,” Landau said.

Valentino emphasized that the model has been used by multiple groups, including the United States government, foreign governments and non-governmental organizations, as one of many inputs for research and data. One explicit example of the use of the Early Warning Project is in the 2019 Global Fragility Act. 

“[The act] devoted two billion dollars to try to prevent conflict in the developing world, and the model was explicitly noted as one of the inputs into identifying which countries would get the most attention,” Valentino said.

The work of the Dickey Center and the Simon-Skjodt Center has been monumental in the field of genocide prevention and is a testament to the change that can be enacted through research and a commitment to progress.

“The Dickey Center is one of the best parts of Dartmouth that I have joined,” Rujuta Pandit ’24 said. “They are really good at providing students with knowledge about how to access opportunities.”

The Early Warning Project is proof of the importance of statistical risk assessment, as seen through its impact locally and globally.