A Longitudinal Study of Police Differential Response Team Impact on Crime in Houston: A Test of Broken Windows Thesis

• Published date: 01 June 2023
• DOI: http://doi.org/10.1177/10986111221092947
• Author: Jihong Zhao,Yan Zhang
• Date: 01 June 2023

Article

Police Quarterly

2023, Vol. 26(2) 143–168

© The Author(s) 2022

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/10986111221092947

journals.sagepub.com/home/pqx

A Longitudinal Study of Police

Differential Response Team

Impact on Crime in Houston:

A Test of Broken Windows

Thesis

Jihong Zhao, and Yan Zhang

Abstract

The purpose of this study is to partially test the broken windows thesis, and to evaluate

the effect of Houston’s Differential Response Team (DRT) in particular at the patrol

division level using crime-related calls for service data (CFS) in the city. Monthly DRT

assignment data over a period of 13 years in f‌ive different space-time patterns were

utilized to estimate the effect of targeted disorder enforcement from 2005 to 2017.

Our f‌indings suggest that the DRT overall did not produce signif‌icant effects on crime-

related calls, with only one out of f‌ive space-time patterns showing demonstrable

benef‌it. We highlighted the underlying features generating differential rates of crime

across geographic areas as an explanation in the discussion section. The unique feature

of this study was the focus on routine police activities as opposed to arrests as the key

indicator of law enforcement intervention in the management of crime.

Keywords

broken windows thesis, differential response team in Houston Police Department,

proactive policing, time series analysis

Department of Criminal Justice and Criminology, Sam Houston State University, Huntsville, TX, USA

Corresponding Author:

Jihong Zhao, Department of Criminal Justice and Criminology, Sam Houston State University, Huntsville, TX

77341, USA.

Email: jzhao@shsu.edu

Introduction

To a large extent, the broken windows thesis has served as the primary catalyst for most

of the recent initiatives to move from traditional policing to disorder-focused policing

(Wilson & Kelling, 1982;Rosenfeld et al., 2007).

1,2

The theory holds that there are

clear associations among visible disorder, public fear and crime in a community.

3

The

postulated remedy for reducing crime and improving the quality of life of citizens is

believed to be to increase proactive police operations by targeting disorder phenomena

such as abandoned buildings, prostitution, street loitering, and fear of crime (for a

review, Skogan, 2015). The thesis holds that local police agencies should aggressively

suppress disorder phenomena; such visible signs of disorder are seen as the harbinger of

the fear of crime among citizens and the cause of serious crime events (e.g., Bratton,

1998). Since its inception the proponents of the broken windows thesis have held to the

belief that disorder policing can signif‌icantly reduce fear and crime, as was initially

evidenced in some empirical studies across the country—most notably in New York

City (e.g., Kelling & Sousa, 2001;Rosenfeld et al., 2007). However, critics of the thesis

quickly pointed out that most empirical f‌indings regarding crime reduction outcomes

did not demonstrate the presence of the hypothesized link between police disorder

enforcement and crime reduction; this relationship remains largely a mystery to this day

(Greenburg, 2014;Harcourt, 2001;Harcourt & Ludwig, 2006;O’Brian et al., 2019).

The purpose of this study is to evaluate the effect of the proactive policing of

disorder on crime-related calls for service (CFS) in Houston, Texas. A principal feature

of the organizational structure of the Houston Police Department is the strategic de-

ployment of a Differential Response Team (DRT) at the patrol division whose task is to

target and take action against social and physical disorder (without prioritizing arrests).

More specif‌ically, we investigate the spatial and temporal patterns of DRT activities and

crime-related calls in the Northeast Patrol Division, one known for its persistent signs

of disorder and elevated crime rates. Time series analysis of monthly DRT assignment

data over a period of 13 years in f‌ive different space-time patterns are utilized to

document the effect of disorder enforcement from 2005 through 2017.

Rationale for the Current Study

The current study makes at least two important contributions to the existing literature on

policing disorder. First, Kelling (2015: 628) has observed with insight that the task of

“f‌ixing of broken windows”goes far beyond the act of increasing misdemeanor arrests.

He notes in this regard: the common misconception “… that broken windows policing

is based on police making numerous arrests. This was far from what Wilson, Coles, or I

thought or put forward.”Prior research has relied upon police misdemeanor arrest data

to serve as a proxy for broken windows policing (for a review: Braga et al., 2015). Here,

we go beyond arrests and assess a variety of proactive police activities against disorder

to delve more deeply into the concept of disorder enforcement. The longitudinal time

and space analysis of DRT activities allows us to identify “natural locations”that arise

144 Police Quarterly 26(2)