Whence the Action? the Persistence and Aggravation of Violent Crime at Addresses, Streets, and Neighborhoods

• Published date: 01 November 2023
• DOI: http://doi.org/10.1177/00224278221112839
• Author: Dan O'Brien,Alex Ciomek
• Date: 01 November 2023
• Subject Matter: Articles

Whence the Action?

the Persistence and

Aggravation of

Violent Crime at

Addresses, Streets,

and Neighborhoods

Dan O’Brien

1,3

and Alex Ciomek

2,3

Abstract

Objectives: Researchers have long studied the persistence of violence and

aggravation from disorder to violence in communities. Recently this

work has begun to consider how these phenomena might operate simulta-

neously at multiple geographic scales. We examine the role of neighbor-

hoods, streets, and addresses in these phenomena, presenting and

assessing a f‌ive-part typology for cross-scale interactions. Methods: We cal-

culated six measures of physical disorder, social disorder, and violent crime

from administrative records for all parcels (i.e., addresses) in Boston, MA,

for 2011–2016. Multilevel models used these measures to predict public

violence and gun-related events in the following year at all three geograph-

ical scales and with cross-scale interactions. Results: Persistence was com-

mon at all scales. Aggravation from disorder to crime was greatest for

1

School of Public Policy and Urban Affairs, Northeastern University, Boston, MA, USA

2

Department of Sociology, Harvard University, Cambridge, MA, USA

3

Boston Area Research Initiative, Northeastern & Harvard Universities, Boston, MA, USA

Corresponding Author:

Dan O’Brien, School of Public Policy and Urban Affairs, Northeastern University, Boston,

MA, USA.

Email: d.obrien@neu.edu

Article

Journal of Research in Crime and

Delinquency

2023, Vol. 60(6) 819–855

© The Author(s) 2022

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/00224278221112839

journals.sagepub.com/home/jrc

addresses. Nearly all signif‌icant cross-level interactions involved addresses.

The most common interactions were reinforced persistence, when persis-

tence of violence at an address was reinforced by violence in the street

or neighborhood; and mediated persistence, when persistence at a higher

geographic scale operated through addresses with disorder. Conclusions:

The study suggests that action is greatest at addresses, but streets and

neighborhoods offer critical context. It also provides a framework for

future work assessing the complementarity of communities and places.

Keywords

hotspots, communities and Crime, broken windows, urban crime,

quantitative research < Research Methods

Introduction

Criminologists studying communities have long concerned themselves with

the geographic persistence of crime and the aggravation of low-level issues,

like physical and social disorder, to more serious problems, like violence.

Historically, studies on these subjects focused on neighborhoods; high

crime neighborhoods would continue to experience high crime (e.g.,

Sampson, 2012), and neighborhoods with disorder were vulnerable to

increases in serious crime (Wilson & Kelling, 1982). Since the 1990s,

however, the emergence of a criminology of place, which focuses on the

dynamics of streets and addresses (Weisburd, Groff, & Yang, 2012), has

complicated this picture. This is a trend that has rapidly accelerated in

recent years, with studies examining the persistence of crime (e.g., Braga,

Hureau, & Papachristos, 2011; e.g., O’Brien & Winship, 2017) and path-

ways from disorder to crime (Wheeler, 2017) at these more localized

scales. Our understanding of how these multiple geographical scales drive

these processes, however, remains inchoate, because few if any studies

have examined addresses, streets, and neighborhoods simultaneously.

Consequently, although there is much discussion of how each geographical

scale “matters”for understanding crime (e.g., Sampson, 2012; Weisburd

et al., 2016), there is little examination of how they work in concert to

effect outcomes.

The current study examines the complementary roles of addresses,

streets, and neighborhoods in driving persistence and aggravation of

crime, def‌ined respectively as the continuation of crime and the increased

820 Journal of Research in Crime and Delinquency 60(6)

severity of crime over time at a given location. Using a longitudinal database

of records generated by 311 and 911 systems in Boston, MA, we test how

multiple forms of physical disorder, social disorder, and violent crime

predict year-on-year change in violent crime. A series of multilevel

models enables us to test two overarching research questions that compre-

hensively capture the complementarity between the three geographical

scales. (1) Which independent effects of persistence and aggravation are rel-

evant at each scale? For instance, is the presence of a certain type of disorder

predictive of violent crime for a whole neighborhood, or only for the streets

or addresses where it is located? (2) What are the cross-scale interactions

between these effects? For example, is disorder at an address more likely

to evolve into violent crime if disorder or crime is prevalent on the street

or in the surrounding neighborhood? These cross-scale interactions have

not been tested to date and have the potential to be diverse in nature. For

this reason, we also develop and present a framework for studying and

describing them.

In sum, the study makes not one but two contributions to the study of

crime in communities. First, it describes the interdependent effects that

addresses, streets, and neighborhoods have in the persistence and aggrava-

tion of crime, thereby advancing and synthesizing multiple nascent lines

of inquiry. Second, it presents a novel framework for studying and describ-

ing cross-scale interactions in driving patterns of crime and disorder.

Though applied here to the specif‌ic test case of persistence and aggravation,

this framework is generalizable to other criminogenic processes that might

operate at multiple geographical scales of interest.Before proceeding to

the data and analysis, the following sections describe current evidence for

persistence and aggravation at the multiple geographical scales; and lay

out the framework for the cross-scale interactions we might expect to see.

Persistence and Aggravation of Crime at Different

Geographical Scales

Research on the persistence and aggravation of crime and disorder have

long, intertwined histories as they are two of the more fundamental ways

of understanding the distribution of crime across time and space. Much of

this work has been oriented around neighborhoods, specif‌ically considering

how community-wide rates of crime and disorder lead to future levels of

crime (Bursik & Grasmick, 1993; Sampson, 2012; Sampson, Raudenbush,

& Earls, 1997; Shaw & McKay, 1942/1969; Wilson, 1987). More recently,

proponents of a criminology of place have highlighted the extent to which

O’Brien and Ciomek 821