Proactive Criminal Thinking and Deviant Identity as Mediators of the Peer Influence Effect

• Date: 01 July 2017
• Published date: 01 July 2017
• Author: Glenn D. Walters
• DOI: 10.1177/1541204016636436
• Subject Matter: Articles

Article

Proactive Criminal Thinking

and Deviant Identity as

Mediators of the Peer

Influence Effect

Glenn D. Walters

1

Abstract

The purpose of this study was to test the moral model of criminal lifestyle development with data

from the 1,725-member (918 boys and 807 girls) National Youth Survey. It was hypothesized that

peer delinquency would predict proactive criminal thinking but not deviant identity as part of a four-

variable chain running from peer delinquency to participant delinquency. Consistent with this

hypothesis, the pathway running from peer delinquency to proactive criminal thinking to deviant

identity to participant delinquency was significant but the pathway running from peer delinquency to

deviant identity to proactive criminal thinking to participant delinquency was not. Deviant identity

nonetheless predicted proactive criminal thinking and delinquency. These results support a major

pathway in the moral model and indicate that while deviant identity plays a role in antisocial

development, it is as a cause and effect of proactive criminal thinking rather than as an effect of

delinquent peer associations.

Keywords

delinquent peer associations, proactive criminal thinking, deviant identity, longitudinal, criminal

pathways

Criminological theory is a rich source of hypotheses on criminal behavior. In an effort to capitalize

on this, Walters (2016c) recently introduced an integrated theory of criminal behavior that rests

heavily on existing criminological theory—social learning theory (Akers, 1998; Sutherland, 1947),

general strain theory (Agnew, 1992), labeling theory (Becker, 1963; Lemert, 1951), and the general

theory of crime (Gottfredson & Hirschi, 1990), in particular. The integrated theory also shares a

great deal in common with life-course developmental theories of crime (Farrington, 2010) through

its emphasis on pathways to delinquency. In presenting his theory, Walters (2016c) emphasized

two interrelated and overlapping models, referred to as the control and moral models of criminal

lifestyle development. Both models highlight constructs from major criminological theories. The

control model, for instance, features low self-control (Gottfredson & Hirschi, 1990), general strain

1

Department of Criminal Justice, Kutztown University, Kutztown, PA, USA

Corresponding Author:

Glenn D. Walters, Department of Criminal Justice, Kutztown University, Kutztown, PA 19530, USA.

Email: walters@kutztown.edu

Youth Violence and JuvenileJustice

2017, Vol. 15(3) 281-298

ªThe Author(s) 2016

Reprints and permission:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/1541204016636436

journals.sagepub.com/home/yvj

(Agnew, 1992), and the peer selection effect (Gottfredson & Hirschi, 1990), with reactive (impul-

sive, irresponsible, and emotional) criminal thinking serving as a principal mediator of pathways

linking these constructs. Walters (2015a) recently tested several of the control model pathways and

discovered that low self-contr ol led to peer rejection (general s train) and peer rejection led to

delinquent peer associations by way of both delinquency (peer selection effect) and reactive criminal

thinking.

Complementing the control model is a moral model of criminal lifestyle development that

features callous/unemotional traits instead of low self-control as the core construct and includes

labeling and identify transformation (Lemert, 1951), cumulative disadvantage (Sampson & Laub,

1993), and peer influence (Sutherland, 1947) as central concepts and proactive (planned, calculated,

and unemotional) criminal thinking as the principal mediator of relationships. Callous/unemotional

traits were originally derived from Facet 2 of Hare’s (2003) Psychopathy Checklist–Revised

(PCL-R) but have since been studied in juvenile populations by Frick and colleagues (Frick, Ray,

Thompson, & Kahn, 2014; Roose, Bijttebier, Decoene, Claes, & Frick, 2010). Research indicates

that compared to children and adolescents with below average to average scores on measures of

callous/unemotional traits, those who score high on these traits are more involved in serious crime

(Frick et al., 2014). Children who exhibit callous/unemotional traits are also more likely to offend in

groups (Goldweber, Dmitrieva, Ca uffman, Piquero, & Steinberg, 2011 ; Thornton et al., 2015),

affiliate with gangs (Thornton et al., 2015), assume leadership roles in a criminal group or gang

(Kerr, Van Zalk, & Stattin, 2012; Thornton et al., 2015), and engage in offenses that are planned and

premeditated (Lawing, Frick, & Cruise, 2010). These findings fit well with the strategic and calcu-

lated nature of proactive criminal thinking and instrumental criminal behavior.

The moral model of criminal lifestyle development is shown in Figure 1. Several of the individual

pathways in this model have already received preliminary support. The causal relationship presumed

to exist between callous/unemotional traits, as measured by Factor 1 of the PCL: Youth Version

(Forth, Kosson, & Hare, 2003), and violent crime was found to be mediated by proactive criminal

thinking in a study by Walters and DeLisi (2015). This would appear to confirm the lower front

pathway in Figure 1 running from callous/unemotional traits to proactive criminal thinking to

delinquency. The connections between callous/unemotional traits and delinquency (Frick et al.,

2014) and between delinquency and delinquent peer associations (Kiesner, Poulin, & Nicotra,

2003) are well established, thus supporting the upper front pathways depicted in Figure 1. There

is also substantial support for the lower middle pathways running from delinquency to labeling to

delinquency and from delinquency to labeling to identity transformation (Bernburg & Krohn, 2003).

Less is known about the upper middle portion of the model, particularly the pathway running from

delinquent peer associations to proactive criminal thinking to identity transformation to delin-

quency, although even here there is evidence that neutralization techniques, which can be considered

a feature of proactive criminal thinking, are learned from friends (Maruna & Copes, 2005).

Research on the consequences of delinquent peer associations provides consistent support for the

peer socialization or influence effect that runs from deviant peer associations to delinquency

(Gifford-Smith, Dodge, Dishion, & McCord, 2005). It has traditionally been assumed that the peer

influence effect is mediated by adolescent involvement in unstructured and unsupervised routine

activities (Osgood, Wilson, O’Malley, Bachman, & Johnston, 1996), although several recent studies

have shown that proactive criminal thinking is also capable of mediating the peer influence effect

(Megens & Weerman, 2012; Walters, 2015b, 2016a). Walters (2016b), in fact, determined that

without cognitive mediation, the peer influence effect failed to materialize. An important research

question, then, is whether additional variables play a role in mediating the peer influence effect

given that multiple mediator models sometimes do a better job of modeling the real world than single

mediator models. Mediators in a multiple mediator model are aligned parallel to one another or

serially. Parallel multiple mediation is where two or more variables simultaneously mediate the

282 Youth Violence and Juvenile Justice 15(3)