Using Twitter to Predict Gang Violence

Columbia's Desmond Patton says law enforcement officials should analyze social media posts about grief and stress to prevent violent crime among young people.

By
Desmond Patton
May 30, 2019

New York City has experienced historic lows in community-based violence over the last two decades but a recent article in The New York Times suggests that parts of Brooklyn are seeing spikes in violence and some believe social media is the culprit. The days of large hierarchically structured gangs are long gone and in our advanced technological age behavior has shifted to “tit-for-tat” violence or petty arguments that may start on platforms like Facebook or Instagram, a behavior known as internet banging or cyberbanging.

Police now follow hashtags, organized attacks and grievances that are embedded in conversations, emojis and memes. I have found, however, that those types of engagements are difficult to understand without critical context, an understanding of the hyper-local language and are usually a last resort for the youth involved in gangs. Overwhelmingly, we find young people are looking for help and support as they deal with complex trauma and grief.

A crucial finding from my lab is that young people post expressions of trauma, grief or stress before turning to more aggressive or threatening posts. If your law enforcement focus is only on the most threatening social media posts you’ve really missed an opportunity to prevent aggressive talk before it becomes violent. 


Desmond Patton is an Associate Professor at Columbia's School of Social Work and department of Sociology. He the founding director of SAFElab, a research initiative at Columbia University focused on examining the ways in which youth of color navigate violence on and offline. This column is editorially independent of Columbia News.