A human-machine partnered approach for identifying social media signals of elevated traumatic grief in Chicago gang territories.

Forrest Stuart,Alicia Riley,Hossein Pourreza

doi:10.1371/journal.pone.0236625

Forrest Stuart, Alicia Riley + Show 1 more

Open Access

https://doi.org/10.1371/journal.pone.0236625

Copy DOI

Abstract

There is a critical need to improve trauma-informed services in structurally marginalized communities impacted by violence and its associated traumatic grief. For community residents, particularly gang-associated youth, repeated exposure to traumatic grief causes serious adverse effects that may include negative health outcomes, delinquency, and future violent offenses. The recent proliferation of digital social media platforms, such as Twitter, provide a novel and largely underutilized resource for responding to these issues, particularly among these difficult-to-reach communities. In this paper, we explore the potential for using a human-machine partnered approach, wherein qualitative fieldwork and domain expertise is combined with a computational linguistic analysis of Twitter content among 18 gang territories/neighborhoods on Chicago’s South Side. We first employ in-depth interviews and observations to identify common patterns by which residents in gang territories/neighborhoods express traumatic grief on social media. We leverage these qualitative findings, supplemented by domain expertise and computational techniques, to gather both traumatic grief- and gang-related tweets from Twitter. We next utilize supervised machine learning to construct a binary classification algorithm to eliminate irrelevant tweets that may have been gathered by our automated query and extraction techniques. Last, we confirm the validity, or ground truth, of our computational findings by enlisting additional domain expertise and further qualitative analyses of the specific traumatic events discussed in our sample of Twitter content. Using this approach, we find that social media provides useful signals for identifying moments of increased collective traumatic grief among residents in gang territories/neighborhoods. This is the first study to leverage Twitter to systematically ground the collective online articulations of traumatic grief in traumatic offline events occurring in violence-impacted communities. The results of this study will be useful for developing more effective tools—including trauma-informed intervention applications—for community organizations, violence prevention initiatives, and other public health efforts.

Highlights

Exposure to community violence and its associated trauma is a pressing concern
These statistics are even more striking in major cities, such as Chicago, where 30% of youth have witnessed a homicide and where, by the age of 15, more than 70% have witnessed at least one serious assault [2]. These numbers are elevated among gang-associated youth and their neighborhood peers, who are overrepresented as victims [3]. Such exposure to violence often results in traumatic grief—defined as the combination of trauma and loss that results from the sudden death of a friend or loved one and that is especially likely in the aftermath of a homicide [4]
Applying our collaborative approach to a sample of tweets about traumatic grief associated with 18 different Chicago gang territories/neighborhoods over a five-year period (2012–2016), we find that Twitter contains useful and accurate signals for identifying elevated levels of collective traumatic grief, as measured through the increased frequency of tweets about traumatic grief uploaded from a given territory/neighborhood

Summary

Introduction

Exposure to community violence and its associated trauma is a pressing concern. In the United States, an estimated 38% of adolescents have witnessed community violence [1]. To understand the extent to which expressions of collective traumatic grief on Twitter correspond to, or match, offline traumatic events and mental states, under the supervision of domain expert annotators, the research team examined the text field of every tweet that comprised each spike in Twitter activity to identify the names of the specific individual subjects to whom the “RIP,” “Long Live,” and “Rest Up” keywords were applied. To determine the prevalence of these modes of online expression among the 18 gang territories/neighborhoods in our analytic sample, our domain expert annotators performed a content analysis of the text fields of the individual tweets contained in the 91 spikes in the collective traumatic grief tweets identified earlier as positive matches.

Conclusion and discussion

Findings

Limitations and suggestions for future research