Does the Residential Landscape Contextualize Friendships? Examining the Causes and Consequences of Affiliating with Older Friends

• Published date: 01 August 2020
• Author: Gregory M. Zimmerman,Alexis Yohros
• DOI: 10.1177/0022427819900644
• Date: 01 August 2020
• Subject Matter: Articles

Article

Does the Residential

Landscape

Contextualize

Friendships?

Examining the Causes

and Consequences

of Affiliating with

Older Friends

Alexis Yohros

1

and Gregory M. Zimmerman

1

Abstract

Objectives: Examine the relationships among structural disadvantage,

friendship network age composition, and violent offending by investigating

the contextual and individual etiology of affiliating with older friends and

exploring the mechanisms that link friendship network age composition to

violent offending. Method: Hierarchical linear models analyze 8,481 respon-

dents distributed across 1,485 census tracts from the National Longitudinal

Study of Adolescent to Adult Health. Social network data are used to

construct a measure of the proportion of a respondent’s friendship net-

work that is at least one grade older than the respondent. Results: Consis-

tent with hypotheses, structural disadvantage increases affiliations with

1

School of Criminology and Criminal Justice, Northeastern University, Boston, MA, USA

Corresponding Author:

Gregory M. Zimmerman, School of Criminology and Criminal Justice, Northeastern Univer-

sity, 431 Churchill Hall, 360 Huntington Avenue, Boston, MA 02115, USA.

Email: g.zimmerman@northeastern.edu

Journal of Research in Crime and

Delinquency

2020, Vol. 57(5) 571-611

ªThe Author(s) 2020

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/0022427819900644

journals.sagepub.com/home/jrc

older friends, older friendshi p networks report hi gher levels of viole nce,

and affiliating with older friends increases violence among respondents.

Contrary to expectations, the influence of affiliating with older friends

on respondent violence decreases, rather than increases, as levels of

violence in the friendship network increase. Conclusions: The results shed

light on the inextricable linkages among social context, friendship network

composition, and sociobehavioral outcomes among youth. The findings

inform peer mentoring program evaluations observing iatrogenic effects via

peer deviancy.

Keywords

peer delinquency, age–crime curve, concentrated disadvantage,

neighborhood effects

Introduction

Age and peer/friend delinquency

1

are two of the most widely documented

correlates of criminal behavior (see Farrington 1986; Hoeben et al. 2016;

Warr 1996). With respect to the relationship between age and crime, the

prevalence of criminal offending tends to increase from late childhood

through adolescence, reach an apex in the late teen years, and then sub-

sequently decline (Loeber, Farrington, and Petechuk 2013). There is no

shortage of research on the curvilinear relationship between age and

crime, commonly referred to as the age–crime curve (e.g., Cook and Laub

2002; Hirschi and Gottfredson 1983). Studies have uncovered important

nuances in the relationship between age and crime (Moffitt 1993), for

example, across crime type (Piquero, Hawkins, and Kazemian, 2012),

gender (Blokland and Palmen 2012), and socioeconomic context (Fabio

et al. 2011), prompting the conclusion that the relationship is complex

(Fagan and Western 2005). Nonetheless, the relative invariance of the

relationship (Hirschi and Gottfredson 1983) has led scholars to refer to

the patterned association as a “resilient empirical regularity” (Brame and

Piquero 2003:107).

Similarly, theory and empirical rese arch on the relationship between

personal delinquency and peer delinquency abounds. Social learning theory

(Akers 1998) suggests that associating with delinquent peer groups impacts

individual outcomes through t he transmission of definitions and be liefs

favorable to delinquency, the modeling and imitation of delinquency, and

the differential reinforcement of delinquency through an imbalance of

572 Journal of Research in Crime and Delinquency 57(5)

heightened social approval of delinq uent acts and infreq uent punishment .

The relevance of friendship formation dynamics for the link between peer

and personal delinquency has also been examined in the criminological

literature. For example, homophily, or the tendency for individuals to

associate with similar others, strengthens network ties among persons,

thus facilitating the socialization process whereby delinquent friends

influence network actors toward delinquency (McPherson, Smith-Lovin,

and Cook 2001).

Empirically, the relationship between personal delinquency and peer

delinquency is one of the most established statistical regularities in crim-

inology. Accordingly, social and behavioral scientists have produced thou-

sands of studies on the topic (Jaccard, Blanton, and Dodge 2005), and no

study has failed to demonstrate the peer effect (Warr 2002). There is evi-

dence that the magnitude of the peer effect varies across characteristics of

the target adolescent (see Brechwald and Prinstein 2011), features and types

of the peer relationship (e.g., Rees and Pogarsky 2011), network character-

istics (e.g., Haynie 2001), broader aspects of the social context, such as the

school (e.g., Cleveland and Wiebe 2003), and levels of peer delinquency

itself (Zimmerman and Messner 2011), suggestive of a nonlinear relation-

ship between peer and respondent delinquency. Nevertheless, social scien-

tists consider peer influence a sine qua non in the etiology of adolescent

delinquency, and scholars generally deem the inclusion of a measure of peer

delinquency essential in delinquency studies during the preadult years

(Haynie 2001).

While the criminological literature on age and peer delinquency is

dense and remarkably consistent, the intersection between these two areas

of inquiry is less well understood. In particular, we know relatively little

about (1) the factors that encourage youth to affiliate with older peers and

(2) whether and how affiliating with older peers facilitates youth delin-

quent behavior. Although one may assume that associating with older

peers during adolescence and young adulthood is criminogenic, empirical

research on the etiology and consequences of affiliating with older peers is

sparse. Harding’s (2009) seminal study represents one of the few sources

of criminological scholarship on the age composition of adolescents’

friendship networks and, accordingly, guides much of the scholarly dis-

cussiononthetopic.

Grounded in Harding’s (2009) insights, this study aims to shed addi-

tional light on the causes and consequences of affiliating with older peers

during adolescence and young adulthood. Accordingly, we assess (1)

whether sociostructural disadva ntage encourages adolescents and youn g

Yohros and Zimmerman 573