Gang Violence, Crime, and Substance Use on Twitter: A Snapshot of Gang Communications in Detroit

Introduction: Gangs are complicated and associated with homicides and victimization related to gun violence, crime and drugs ( Swahn et.al. 2010). Gang-related homicides account for roughly 12 percent of all homicides annually ( NYGS, 2011). Emerging literature suggest that social media acts as a conduit for violence exchanges, and those interactions have great potential for spillover into community settings. There are little to no studies that investigate the use of Twitter specifically to promote gang behaviors and activities, particularly in urban areas.  The purpose of this study is to enhance our understanding of the role of social media as a potential vector for violence and substance use in urban areas.

Method: We performed an advanced keyword search of Tweets from Detroit gangs collected using an application programming interface (API). Our process involved four key steps. First, we produced key terms related to gang violence, crime and substance use. We reviewed sites like streetgangslang.com, worldstarhip.com, and YouTube. Second, we identified gang members by using common Detroit gang Twitter names. We then ran the identified Twitter names through an open source intelligence tool to find “ follows” and followers and created an excel spreadsheet to store and analyze the data. Third, we searched our key terms in order to remove unaffiliated links that are not connected to our search. During this stage suspected gang members who do not use gang-naming conventions were revealed as well as other previously unknown variants of a respective gang Twitter names. Also, we developed a core group of gangs and gang member Twitter names. Lastly, we performed queries using a retrieval mechanism that defines the relationship between groups of words. We used words like AND, OR and NOT and parentheses to retrieve specific posts in Twitter. If we did not see evidence of key terms associated with violence, crime and substance use we pruned those respective Twitter names from the data set.

Results: We analyzed 8.5 million tweets by known Detroit gang members from January 2013- March 2014. Of these tweets, 355,514 ( .04%) were relevant to our search. We identified four distinct ways gangs communicated on Twitter in the area of violence, crime and substance use: (1) grieving the loss of a loved one ( 79,971 tweets); (2) “ beefing” or online arguing with rival gangs ( 267,221 tweets);  (3) displaying or discussing firearms (e.g glock) ( 3,551 tweets); and displaying or discussing  illicit substances and alcohol ( e.g. “ dirty sprite, marijuana, Xanax, estasy, Patron, Grey Goose) ( 29,635 tweets).

Conclusions: Understanding how gang members communicate on Twitter has important research and practice implications for documenting how gangs communicate on social media and developing appropriate prevention and intervention strategies that utilize Twitter. For example, practictioners may consider social media as an assessment tool to: (1) engage gang involved youth in discussions around psychological (e.g., PTSD) and physical health; (2) track the escalation of violence that occurs on social media; and (3) provide a space for youth to describe negative sequalae associated with gang violence exposure.