For 26 years until 2009, Sri Lanka was embroiled in a devastating civil war that ravaged the landscape, ruined the economy, and left an estimated 100,000 people dead. Princeton University student Sridutt Nimmagadda ’18 learned about the conflict after watching a British documentary about it several years ago and was shocked by atrocities that the government committed.
“It had such an effect on me,” said Nimmagadda, 22. “I was horrified. It made me think about the brutality of war.”
He started looking more closely at the conflict when he learned that the Tamil people, the insurgents, originally came from India, where Nimmagadda himself was born, to work at rubber, tea and coffee plantations in Sri Lanka. Tamils battled with the Sri Lankan government, which was backed by the majority Sinhalese ethnic group, for an independent Tamil state until the rebels were defeated in 2009.
In his senior year, Nimmagadda decided to do a deep dive on conflicts like the Sri Lankan civil war by specifically looking at raw data compiled by the United Nations and other organizations and analyzing whether there was a correlation between the length of a conflict and the state committing human rights violations against its citizens. Can the use of political atrocities by a cruel government win the war and shorten a conflict?
This provocative question was at center of Nimmagadda’s senior thesis, which recently won the Independent Work Prize at the end-of-the-school-year poster session at the Center for Statistics and Machine Learning, where he earned a certificate. Nimmagadda, a graduate of the Woodrow Wilson School of Public and International Affairs, also earned a certificate on South Asian studies.
On first glance, the answer seems to be yes, Nimmagadda said. It’s been well documented by human rights organizations that despotic leaders have effectively used war crimes, such as mass killings and rapes, to squash dissent. After sifting through reams of data though, he found that assumption is actually not correct. The more the government engages in atrocities and causes more deaths, the conflict’s duration actually increases. But other variables also come into play he found out. A government that has consolidated its influence among various centers of power within a country and has strong domestic and international backing can quickly bring a civil war to an end. And a government with weak political institutions and interference from outside power centers generally engages in conflicts that go on longer.
For the project, Nimmagadda looked at 31 civil wars and 176 observations of human rights violations in those conflicts between 1995 and 2007. He also studied in closer detail the current Syrian war that was sparked by the 2011 Arab Spring and the Sri Lankan civil war from 2006 to 2009 when the conflict resumed after a ceasefire. He used a variety of statistical techniques, such as hypothesis testing, graphical representation, and Cox Regression modeling to analyze the data.
“My argument is that political atrocities are not enough, the strength of political institutions matter,” he said, pointing to the flare up of the Sri Lankan civil war in the mid-Aughts.
“In the case of Sri Lanka, the country has very strong political institutions. The government consolidated power specifically in the military.”
Because of this factor, including the government successfully labeling the Tamils as a terrorist group and shutting out human rights observers out of the country, the conflict was relatively short, he said. Syria is an opposite case. Foreign-backed rebels and the unpopular and weak government have ground that conflict to essentially a stalemate.
“The courses at the Center for Statistics and Machine Learning really helped me out and gave me the skill set to complete my study,” he said. “This thesis has a lot of data. I had to create my own data set from five to six sources, bring them together, and manipulate them in different ways to finally come out with my conclusion.”
“The main takeaway from this project is anyone can interpret numbers, but the true value of a project like this is how to deal with data,” he added. “It sounds trivial, but it isn’t. You are dealing with data from different sources. Sometimes you are presented with missing data, or data inconsistent between different groups. Making sense of it and knowing how numbers came about will help you arrive at a more robust conclusion.”
Melissa M. Lee, assistant professor of politics and international affairs in the Department of Politics and the Woodrow Wilson School, served as Nimmagadda’s adviser. He said Lee was integral in making his project a success because she provided value input on the data.
After graduating, Nimmagadda is turning his focus from the Sri Lankan civil war to long-held, happier interests. The NFL drafted him to be a junior rotational program analyst, a position that starts in about a month. He was chosen as one of seven people from over 2,000 aspirants for the prestigious job, which gives young professionals experience in working in the organization’s various departments - from broadcasting, finance, corporate strategy, to business development.
Nimmagadda was eager to join the NFL because he’s a “huge” football fan. He roots for his hometown team, the Seattle Seahawks.
“I've been a Seahawks fan since I was a little kid. It's hard to avoid when you grow up in Seattle, but they did go to the Super Bowl in 2005 and I suppose that's when I began caring to some degree,” he said. “But my fandom didn't really explode until I saw the 2014 NFC Championship against the (Green Bay) Packers. I still maintain that game was one of the most inspirational events I've ever witnessed, ever.”
As Nimmagadda launches his career, with an eye towards corporate strategy in media and entertainment, he wants to continue to utilize the statistical tool kit he developed studying the Sri Lankan civil war and other conflicts.
“I'm passionate about the applications of data analytics,” he said. “Thinking about numbers is very important to me."