A Network Analysis of Factors Leading Adolescents to Befriend Substance-Using Peers

David R. Schaefer

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Objective: Our interest is in the systematic network selection processes that lead adolescents into friendships with substance-using peers. Theory suggests that adolescents with certain risk factors (i.e., weak attachments to conventional society and low self-control) are more likely to select substance-using friends. Our goal is to evaluate whether adolescents with particular risk factors have a greater risk for befriending substance-using peers, while controlling for common network selection processes that can produce the same friendship pattern. These selection processes are important as they help to set the stage for later peer influence on substance use. Methods: We use a Stochastic Actor-Oriented Model to examine network change among 1373 adolescents from the National Longitudinal Study of Adolescent Health. We test whether low self-control and indicators of weak attachments (to family, school, and religion) predict selecting friends engaged in alcohol, tobacco, and marijuana use. Results: We find widespread evidence of the hypothesized friendship pattern within adolescent friendship networks. In most cases this pattern is a product of selection based on the risk factor and substance use, and not attributable to other selection mechanisms. Conclusions: We highlight the need to broaden the study of delinquency to account for how adolescents come to acquire friends who may be negative sources of peer influence. We offer theoretical and methodological insight to this question, ultimately finding that only in limited cases are adolescents with particular risk factors more likely to select friends involved in substance use. We discuss implications for theory and future investigations of peer influence.

Original languageEnglish (US)
Pages (from-to)1-38
Number of pages38
JournalJournal of Quantitative Criminology
DOIs
StateAccepted/In press - Dec 28 2016

Fingerprint

network analysis
Statistical Factor Analysis
adolescent
friendship
self-control
National Longitudinal Study of Adolescent Health
Religion
Tobacco Use
Cannabis
delinquency
nicotine
longitudinal study
alcohol
Alcohols
health
school
evidence
Peer Influence

Keywords

  • Adolescents
  • Self-control
  • Social network
  • Stochastic Actor-Oriented Model
  • Substance use
  • Weak attachments

ASJC Scopus subject areas

  • Pathology and Forensic Medicine
  • Law

Cite this

A Network Analysis of Factors Leading Adolescents to Befriend Substance-Using Peers. / Schaefer, David R.

In: Journal of Quantitative Criminology, 28.12.2016, p. 1-38.

Research output: Contribution to journalArticle

@article{a575932ebca14ffab2aae858bfc6e531,
title = "A Network Analysis of Factors Leading Adolescents to Befriend Substance-Using Peers",
abstract = "Objective: Our interest is in the systematic network selection processes that lead adolescents into friendships with substance-using peers. Theory suggests that adolescents with certain risk factors (i.e., weak attachments to conventional society and low self-control) are more likely to select substance-using friends. Our goal is to evaluate whether adolescents with particular risk factors have a greater risk for befriending substance-using peers, while controlling for common network selection processes that can produce the same friendship pattern. These selection processes are important as they help to set the stage for later peer influence on substance use. Methods: We use a Stochastic Actor-Oriented Model to examine network change among 1373 adolescents from the National Longitudinal Study of Adolescent Health. We test whether low self-control and indicators of weak attachments (to family, school, and religion) predict selecting friends engaged in alcohol, tobacco, and marijuana use. Results: We find widespread evidence of the hypothesized friendship pattern within adolescent friendship networks. In most cases this pattern is a product of selection based on the risk factor and substance use, and not attributable to other selection mechanisms. Conclusions: We highlight the need to broaden the study of delinquency to account for how adolescents come to acquire friends who may be negative sources of peer influence. We offer theoretical and methodological insight to this question, ultimately finding that only in limited cases are adolescents with particular risk factors more likely to select friends involved in substance use. We discuss implications for theory and future investigations of peer influence.",
keywords = "Adolescents, Self-control, Social network, Stochastic Actor-Oriented Model, Substance use, Weak attachments",
author = "Schaefer, {David R.}",
year = "2016",
month = "12",
day = "28",
doi = "10.1007/s10940-016-9335-4",
language = "English (US)",
pages = "1--38",
journal = "Journal of Quantitative Criminology",
issn = "0748-4518",
publisher = "Springer New York",

}

TY - JOUR

T1 - A Network Analysis of Factors Leading Adolescents to Befriend Substance-Using Peers

AU - Schaefer, David R.

PY - 2016/12/28

Y1 - 2016/12/28

N2 - Objective: Our interest is in the systematic network selection processes that lead adolescents into friendships with substance-using peers. Theory suggests that adolescents with certain risk factors (i.e., weak attachments to conventional society and low self-control) are more likely to select substance-using friends. Our goal is to evaluate whether adolescents with particular risk factors have a greater risk for befriending substance-using peers, while controlling for common network selection processes that can produce the same friendship pattern. These selection processes are important as they help to set the stage for later peer influence on substance use. Methods: We use a Stochastic Actor-Oriented Model to examine network change among 1373 adolescents from the National Longitudinal Study of Adolescent Health. We test whether low self-control and indicators of weak attachments (to family, school, and religion) predict selecting friends engaged in alcohol, tobacco, and marijuana use. Results: We find widespread evidence of the hypothesized friendship pattern within adolescent friendship networks. In most cases this pattern is a product of selection based on the risk factor and substance use, and not attributable to other selection mechanisms. Conclusions: We highlight the need to broaden the study of delinquency to account for how adolescents come to acquire friends who may be negative sources of peer influence. We offer theoretical and methodological insight to this question, ultimately finding that only in limited cases are adolescents with particular risk factors more likely to select friends involved in substance use. We discuss implications for theory and future investigations of peer influence.

AB - Objective: Our interest is in the systematic network selection processes that lead adolescents into friendships with substance-using peers. Theory suggests that adolescents with certain risk factors (i.e., weak attachments to conventional society and low self-control) are more likely to select substance-using friends. Our goal is to evaluate whether adolescents with particular risk factors have a greater risk for befriending substance-using peers, while controlling for common network selection processes that can produce the same friendship pattern. These selection processes are important as they help to set the stage for later peer influence on substance use. Methods: We use a Stochastic Actor-Oriented Model to examine network change among 1373 adolescents from the National Longitudinal Study of Adolescent Health. We test whether low self-control and indicators of weak attachments (to family, school, and religion) predict selecting friends engaged in alcohol, tobacco, and marijuana use. Results: We find widespread evidence of the hypothesized friendship pattern within adolescent friendship networks. In most cases this pattern is a product of selection based on the risk factor and substance use, and not attributable to other selection mechanisms. Conclusions: We highlight the need to broaden the study of delinquency to account for how adolescents come to acquire friends who may be negative sources of peer influence. We offer theoretical and methodological insight to this question, ultimately finding that only in limited cases are adolescents with particular risk factors more likely to select friends involved in substance use. We discuss implications for theory and future investigations of peer influence.

KW - Adolescents

KW - Self-control

KW - Social network

KW - Stochastic Actor-Oriented Model

KW - Substance use

KW - Weak attachments

UR - http://www.scopus.com/inward/record.url?scp=85007449361&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85007449361&partnerID=8YFLogxK

U2 - 10.1007/s10940-016-9335-4

DO - 10.1007/s10940-016-9335-4

M3 - Article

AN - SCOPUS:85007449361

SP - 1

EP - 38

JO - Journal of Quantitative Criminology

JF - Journal of Quantitative Criminology

SN - 0748-4518

ER -