Examining the Effects of Individual-Level and Neighborhood-Level Characteristics on the Variability of Substance Use Rates and Changes

• Date: 01 July 2018
• Author: Adrian M. Jones,Richard E. Adams
• DOI: 10.1177/0022042618759322
• Published date: 01 July 2018
• Subject Matter: Articles

https://doi.org/10.1177/0022042618759322

Journal of Drug Issues

2018, Vol. 48(3) 337 –355

© The Author(s) 2018

Reprints and permissions:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/0022042618759322

journals.sagepub.com/home/jod

Article

Examining the Effects of Individual-

Level and Neighborhood-Level

Characteristics on the Variability of

Substance Use Rates and Changes

Adrian M. Jones1 and Richard E. Adams2

Abstract

We use social learning, self-control, and social disorganization theories to explain substance

use among urban adolescents. Using a sample of 2,048 adolescents and young adults from the

Project in Human Development in Chicago Neighborhoods, we develop and test longitudinal

growth models that examine how respondent and neighborhood characteristics relate to rates

and changes of tobacco, alcohol, marijuana, and hard drug use. We find empirical support for

social learning and self-control theories in relation to substance use rates, but no support for

social disorganization theory. In addition, we find evidence that deviant peers, self-control,

and concentrated disadvantage are associated with respondent changes in substance use. We

discuss these findings in relation to the three theories that informed our study.

Keywords

substance use, deviant peers, self-control, neighborhood effects, longitudinal growth models

Over the past 40 years, researchers have attempted to identify risk factors related to adolescent

substance use (Akers, 1973; Erickson, Harrison, Cook, Causineau, & Adlaf, 2012; Kandel, 1978;

Shedler & Block, 1990; Winfree & Bernat, 1998). Early efforts focused on psychological and

personality characteristics that differentiated substance users from nonusers (Shedler & Block,

1990). Researchers using more sociological perspectives have paid attention to how social cir-

cumstances and deviant peers are implicated in the likelihood of adolescent substance use (Akers,

1973, 2009). Few studies, however, have combined multiple perspectives that incorporated

neighborhood attributes and individual characteristics into models that predict rates and changes

in substance use (see Jackson, Denny, & Ameratunga, 2014). In this study, we attempt to address

this gap in the literature. Utilizing variables related to social learning theory (Akers, 1973), the

general theory of crime (Gottfredson & Hirschi, 1990), and social disorganization theory (Shaw

& McKay, 1929), we develop longitudinal models that examine how individual-level personality

and social characteristics, such as associations with deviant peers and self-control, and

1Kent State University, New Philadelphia, OH, USA

2Kent State University, Kent, OH, USA

Corresponding Author:

Adrian M. Jones, Department of Sociology, Kent State University, Tuscarawas Campus, 330 University Dr. N.E., New

Philadelphia, OH 44663, USA.

Email: amjones4@kent.edu

759322JODXXX10.1177/0022042618759322Journal of Drug IssuesJones and Adams

research-article2018

338 Journal of Drug Issues 48(3)

neighborhood contextual factors, such as collective efficacy and concentrated disadvantage, are

related to substance use and changes in substance use over three waves of data. We argue that

combining aspects of these three theories into a single analysis allows researchers to test utility

of each in explaining rates and changes in substance use among adolescents and to offer a more

complete explanation of substance use risk factors, while controlling for other individual, social,

and structural variables.

Theoretical Background

According to social learning theory, deviant behavior, including substance use, is not a product

of innate pathology, but rather, it is a social artifact that develops through socialization with pri-

mary group members (Sutherland & Cressey, 1970). During the socialization process, individu-

als learn values, norms, techniques, motivations, and favorable and unfavorable definitions

concerning behavior by observing, modeling, and imitating others (Sutherland & Cressey, 1970;

Winfree & Bernat, 1998). In addition, the socialization involves conditioning through rewards

and punishments, and individuals with many deviant associates are more likely to be rewarded,

not punished for committing deviant acts (Akers, 1985; Winfree & Bernat, 1998). Thus, indi-

viduals who have more deviant associations than nondeviant associations will be more likely to

learn values, norms, techniques, motivations, and favorable definitions about deviance, like sub-

stance use (Sutherland & Cressey, 1970).

Social learning theory has received significant empirical support for the proposition that prob-

lem behaviors are learned and reinforced through peer socialization (Akers, 1985; Akers, 2009).

Variables capturing deviant peer associations have been found to be the strongest predictor of

deviant behavior (Kornhauser, 1978) and substance use (Kandel, 1978; Winfree & Bernat, 1998).

In addition, previous studies have found significant positive relationships with deviant peer asso-

ciations and drinking (Akers, 2009; Jessor & Jessor, 1977), smoking (Akers, 2009; Kobus, 2003;

Krohn, Skinner, Massey, & Akers, 1985; Rooney & Wright, 1982), marijuana use (Akers, 2009;

Fagan, Wright, & Pinchevsky, 2013), and substance use in general (Desmond, Bruce, & Stacer,

2012; Fagan et al., 2013; Zimmerman & Farrell, 2016).

Other individual-level factors, such as self-control, have also been found to be associated with

delinquency and substance use, even after controlling for social learning variables such as devi-

ant peers (see Pratt & Cullen, 2000; Winfree & Bernat, 1998). According to Gottfredson and

Hirschi (1990), “people who lack self-control will tend to be impulsive, insensitive, physical (as

opposed to mental), risk-taking, short sighted, and nonverbal, and they will tend therefore to

engage in criminal and analogous acts” (p. 90). Thus, unlike individuals with high self-control,

individuals with low self-control are more likely to take risks, for example, use alcohol or drugs,

to fulfill immediate gratifications while overlooking potential consequences (i.e., long-term

health problems and legal trouble).

In this regard, Gottfredson and Hirschi’s (1990) general theory of crime has been important

for understanding the relationship between low self-control and a broad range of problem behav-

iors, including substance use, and the hypothesis that low self-control is associated with problem

behaviors has been generally supported (see Pratt & Cullen, 2000; Vazonyi, Mikuska, & Kelly,

2017, for meta studies). Relevant to the current study, Winfree and Bernat (1998) found that risk

taking, a dimension of self-control, was significantly related to substance use, although other

dimensions (e.g., impulsivity) were not. More recently, Desmond et al. (2012) found that low

self-control was positively related to tobacco and alcohol use and that delinquent peers’ use of

these substances interacted with the effect of self-control on rates of smoking and drinking.

In addition to examining how adolescent characteristics influence development and behavior,

social scientists from a variety of disciplines have also been concerned with the connection

between context and developmental and behavioral outcomes (e.g., Bronfenbrenner, 1994;