The Seasons They Are a Changin’

• Author: Evan T. Sorg,Cory P. Haberman,Jerry H. Ratcliffe
• DOI: 10.1177/0022427818758375
• Date: 01 May 2018
• Published date: 01 May 2018
• Subject Matter: Articles

Article

The Seasons They

Are a Changin’:

Testing for Seasonal

Effects of Potentially

Criminogenic Places

on Street Robbery

Cory P. Haberman

1

, Evan T. Sorg

2

,

and Jerry H. Ratcliffe

3

Abstract

Purpose: To examine a component of crime pattern theory by exploring

whether the spatial predictors of crime vary across seasons. Methods: The

relationships among potentially criminogenic places and illicit markets and

seasonal census block robbery counts in Philadelphia, PA, were explored

using simultaneously estimated negative binomial regression models. The

equality of predictors’ effects on street robbery across seasons was subse-

quently tested using Wald’s tests. Results: While many facilities and illicit

markets were positively associated with street robbery, there were few

seasonal differences in their effects. Only the effect of high schools during

the fall was greater than during the winter and summer as hypothesized.

1

University of Cincinnati, Cincinnati, OH, USA

2

Rowan University, Glassboro, NJ, USA

3

Temple University, Philadelphia, PA, USA

Corresponding Author:

Cory P. Haberman, University of Cincinnati, 2160 McMicken Circle, 660 H Dyer Hall,

Cincinnati, OH 45221, USA.

Email: cory.haberman@uc.edu

Journal of Research in Crime and

Delinquency

2018, Vol. 55(3) 425-459

ªThe Author(s) 2018

Reprints and permission:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/0022427818758375

journals.sagepub.com/home/jrc

Conclusions: The results suggest areas with facilities and illicit markets that

are used consistently across the year experience the highest street robbery

levels regardless of the season.

Keywords

crime pattern theory, routine activities theory, facilities, seasonality

Rational crime control policies should be based on sound theoretical expla-

nations of the causes of crime (Mears 2007; Taylor 2015:20-21). Recently,

environmental criminology has underpinned many important innovations in

crime control policy (Cullen 2011). For example, opportunity reduction via

problem-oriented policing in hot spots is regarded as one of the most effec-

tive modern policing tactics (Braga, Papachristos, and Hureau 2014). A key

axiom of environmental criminology is that crime is concentrated in space

and time, thus it follows that spatial-temporal concentrations have been

assumed to be important for understanding and addressing crime problems

(e.g., see Clarke and Eck 2005, Steps 25-26). A large body of research has

consistently shown that crime is concentrated in space (Weisburd 2015), but

much less is known about spatial and temporal crime patterns. Most envi-

ronmental criminology research to date has been atemporal (Haberman,

Sorg, and Ratcliffe 2017). This research tends to demonstrate a dispropor-

tionate level of crime is concentrated in few places or explain those con-

centrations by regressing crime levels from a single year on characteristics

of the built and social environment. By design, these studies assume the

observed effects are temporally invariant despite environmental criminol-

ogy’s axiom that space and time are important for understanding crime

patterns.

The few studies that have examined spatial-temporal crime patterns to

date have focused on within-day spatial-temporal patterns and generally

found that crime concentrates in busy places regardless of time of day

(e.g., see Bernasco, Ruiter, and Block 2017; Haberman and Ratcliffe

2015). A large body of research dating back to the 1800s, however, has

argued that crime is seasonal, and environmental criminology theories have

become the widely accepted explanation of seasonal crime patterns. None-

theless, we demonstrate below that existing research has not adequately

tested whether the predictors of spatial crime patterns vary across seasons

as predicted by environmental criminology theories. Given the importance

of environmental criminology for current and future (unknown) crime

426 Journal of Research in Crime and Delinquency 55(3)

control policy, science that tests environmental criminology’s core yet

untested hypotheses is imperative for further developing the theory as well

as its implications for crime control policy.

Therefore, this study used data on potentially criminogenic facilities and

illicit markets to test whether the effects of these predictors on geographic

street robbery levels are different across seasons in Philadelphia (net of

sociodemographics). Specifically, we outline hypotheses of how different

potentially criminogenic facilities might link to geographic street robbery

levels across different seasons according to environmental criminology. We

then use simultaneously estimated negative binomial regression models and

equality of coefficient tests to directly examine whether some predictors

have greater effects during some seasons than others. Except for high

schools, the effects of potentially criminogenic facilities and illicit markets

on street robbery were not seasonal. Thus, we conclude by discussing the

importance of these findings for understanding spatial-temporal crime pat-

terns using environmental criminology theories.

Literature Review

Theories of Seasonal Crime Patterns

The existing studies of seasonal crime patterns generally note that the

literature on the topic dates to at least the 1800s (usually by citing Quetelet

1842). Researchers have long focused on explaining how and why some

crimes peak at some times of the year (Baumer and Wright 1996; Cohn

1990). Early research mostly posited that humans’ psychological response

to higher temperatures could explain seasonal violence patterns (Rotton and

Cohn 1999). Early laboratory studies found that humans responded more

aggressively as ambient temperature increased (Baron and Bell 1976).

Researchers later debated whether the re lationship between temperature

and crime levels is linear (general affect model) versus curvilinear (negative

affect escape model; Anderson and DeNeve 1992; Bell 1992; Cohn and

Rotton 1997; Rotton and Cohn 1999), but correlations have been found

between ambient temperature and aggregate violence levels (e.g., see Field

1992; Lab and Hirschel 1988; Rotton and Frey 1985).

Nonetheless, researchers eventually started to question the utility of

hypotheses stressing the importance of humans’ response to temperatures

for explaining patterns of other crime types (e.g., property crime; Hipp et al.

2004). Many researchers started to argue that routine activities theory

(RAT) could better explain the relationship between temperature and all

Haberman et al. 427