Variability in the Use of Misdemeanor Arrests by Police Agencies From 1990 to 2013: An Application of Group-Based Trajectory Modeling

• Author: Cynthia Lum,Heather Vovak
• DOI: 10.1177/0887403417731598
• Date: 01 July 2018
• Published date: 01 July 2018
• Subject Matter: Articles

https://doi.org/10.1177/0887403417731598

Criminal Justice Policy Review

2018, Vol. 29(6-7) 536 –560

© The Author(s) 2017

Reprints and permissions:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/0887403417731598

journals.sagepub.com/home/cjp

Article

Variability in the Use of

Misdemeanor Arrests by

Police Agencies From 1990

to 2013: An Application of

Group-Based Trajectory

Modeling

Cynthia Lum1 and Heather Vovak1

Abstract

Arrest for minor offenses has become one tool that some police departments

employ to fight crime and disorder in their jurisdictions. Dubbed by some as “broken

windows” or “zero tolerance” policing, a few police agencies in the 1990s and 2000s

notably and significantly increased their use of arrest for such misdemeanors, such

as New York City and Baltimore. But was this the case for other law enforcement

agencies in the United States? Our analysis is the first to examine long-term trends in

the use of misdemeanor arrests in a sample of U.S. law enforcement agencies using

group-based trajectory modeling. Results show that police agencies have distinct

longitudinal patterns of use of arrests for minor crimes from 1990 to 2013; some

agencies significantly increased their use of arrests for minor crimes while others

did not. Further analysis of possible explanations for agency membership in any

given longitudinal trajectory found that agencies with similar patterns in their use of

misdemeanor arrests were not similar on demographic or crime characteristics. This

finding suggests that the decision to increase the use of arrest for minor offenses may

have been a policy choice by agencies influenced by factors not detected here.

Keywords

policing, misdemeanor, arrests, broken windows, zero tolerance, trajectory analysis

1George Mason University, Fairfax, VA, USA

Corresponding Author:

Cynthia Lum, Department of Criminology, Law and Society, Center for Evidence-Based Crime Policy,

George Mason University, 4400 University Drive, MS 6D12, Fairfax, VA 22030, USA.

Email: clum@gmu.edu

731598CJPXXX10.1177/0887403417731598Criminal Justice Policy ReviewLum and Vovak

research-article2017

Lum and Vovak 537

Introduction

Arrest for minor, nonviolent offenses in the United States has become one tool that

some police departments and officers employ to fight crime and disorder in their juris-

dictions. While arresting individuals for such offenses as vandalism, drug abuse viola-

tions, gambling, public drunkenness, vagrancy, or loitering is often discretionary,

some agencies have proactively chosen to reduce their discretion in favor of the more

legalistic arrest option for these offenses. In aggregate, we see a trend of increased use

of misdemeanor arrests in the United States starting in the 1980s for large agencies,

after both a period of increase in violence and the adoption of mandatory arrest poli-

cies for domestic violence. Even when excluding misdemeanor domestic violence and

assaults, the total number of misdemeanor arrests increased among larger agencies,

only to begin decreasing in the late 2000s.1

One possible explanation for this increase was the embracing by police officials of

a theory about the relationship between minor offenses and serious crimes known as

broken windows. In 1982, James Q. Wilson and George Kelling hypothesized in their

now often-cited article in The Atlantic that disorders and public nuisances could be

metaphorically viewed as “broken windows.” Such broken windows—when left

unattended—could attract further disorder and more serious crime because they sig-

naled a lack of care, concern, and guardianship. In their words, “disorder and crime

are usually inextricably linked” (Wilson & Kelling, 1982, p. 31). Consequentially,

they argued, unattended broken windows can lead to a “breakdown of community

controls” (p. 31) and could make a place “vulnerable to criminal invasion” (p. 32). By

fixing broken windows quickly, police and communities could prevent crime.

However, while their theoretical assertion was clear (see also Sousa & Kelling,

2006), Wilson and Kelling were more ambiguous about what the police might do

about broken windows. For example, they promoted the police as having a central role

in order maintenance to shore up informal community controls, but they did not pre-

scribe any one particular approach. They suggested that police officers enforce rules

about disorderly conduct, but described enforcement as not just arresting an offender

but moving an offender along, such as ejecting someone from a bus or location. Wilson

and Kelling also warned about the potential problems of unfair and harsh treatment by

the police in their order maintenance role.

Since the essay’s publication, the broken windows hypothesis has been interpreted

by law enforcement officials and scholars in different ways (see discussions in Braga,

Welsh, & Schnell, 2015; Sousa & Kelling, 2006). For example, policing disorder and

other minor offenses can take the form of aggressive arrest and order maintenance

programs, including “zero tolerance” policing. The zero tolerance approach reduces

officer discretion in the use of arrest for minor violations such as urinating in public,

loitering, jumping subway turnstiles, obstructing traffic, littering, possessing small

quantities of drugs, public intoxication, street gambling, and the like. The idea is that

individuals who commit such offenses might also be responsible for, or attract, more

serious crimes, which can be averted by arrest. However, Braga et al. (2015) empha-

sized that some police agencies have interpreted broken windows to mean focusing on