Main Story Line - 1.
Abstract
This study examined ways in which normal adolescent peer processes might begin to explain the dramatic increases in delinquency that occur from early to mid-adolescence (even among teenagers who are functioning relatively well). Adolescents (N=148), along with several of their peers have been participating annually in the study beginning at age 13.
Three main sets of findings are reported:
I. Adolescents who associate with more popular peers at
age 13, show greater increases in delinquency over the
following year.
II. When two peers are observed interacting around one
peer’s requests for help, the more dominant teen tends to
be the more deviant teen. If this pattern generalizes, it
would suggest that even within normal, heterogeneous
peer groups, there will be a bias toward greater deviance,
because the deviant teens will be more dominant.
III. Teens who are most susceptible to peer influence at age
13 are more likely to increase in deviance over the
following year.
We conclude that increasing delinquency rates in early adolescence may be at least partly explained by presence of a broad peer culture supporting delinquency, and that the presence of “dominant deviants” may help drive this culture.
；
Main Story Line - 2.
Introduction
The normative increases in delinquency that occur during early- to mid-adolescence are widely recognized but poorly understood [Moffitt, 1993 #2323]. Understanding the peer processes that help account for these increases is central to learning to reduce these normative increases in deviance, thus leading to our focus in this study:
Central Issue
What peer processes can help explain the dramatic increase in delinquency rates in early- to mid-adolescence?
Hypothesis:
To explain such widespread and sustained increases in delinquency, normal adolescent peer culture seems very likely to be implicated. But, how and why?
Three Primary Questions:
I. Does Tight Integration into Normal Peer Culture Predict
Increasing Deviance in Adolescence?
II. What Processes May Underlie Such an Effect?
III. Which Adolescents Are Most Susceptible to this Effect?
；
Main Story Line - 3.
Results & Discussion
I. Association with Popular Peers at Age 13
Predicts Increasing Rates of Delinquency by
Age 14:
Teen Delinquency (Age 14)
22 Β Total R;R
Step I. Teen Delinquency (13) .26** .07** .07**
Step II.
Gender (1=M; 2=F) -.02
Age .06
Family Income .14 .02 .09* Step III. Popularity of Closest Friend .20* .02* .11**
Teen Delinquency (Age 14)
22 Β Total R;R
Step I. Teen Delinquency (13) .27** .07** .07**
Step II.
Gender (1=M; 2=F) -.03
Age .04
Family Income .13 .02 .09* Step III.
THPopularity of 4 Closest Friend .24** .04** .13***
β weights are from variable’s entry into models.
；
Main Story Line - 4.
Why does friend popularity predict increasing delinquency?
These findings raised the question of whether adolescent interactions with their peers might in some way foster deviance. Others [Dishion, 1996 #2483] have found that delinquent peers tend to reinforce one another’s behaviors, but these findings exist within community samples of adolescents with relatively low base-rates of deviance.
We proposed that these effects could be explained if, even in heterogeneous groups of largely normal youths, the more deviant youths within the groups tended to dominate the discourse.
We needed a way to examine discourse processes that:
a. let us observe peers interacting; and
b. was not simply focused on deviant behavior (because
we were proposing a more general group process).
For these purposes, we developed a “Supportive Behavior
Task” and coding system (see details below), which did not focus on delinquency, and we coded the extent to which our target teen was more dominant in the interaction.
When we examined dyadic interaction processes using this system, we arrived at a robust finding:
；
Main Story Line - 5.
II. Adolescents who are most dominant of their
peers are also more likely to be delinquent:
Correlation of Observed Dominance
In Teen-Peer Discussions &
Self-reported Delinquency : r = .30***
The effect remains (and becomes slightly
stronger) when covariates are removed:
Correlation after Partialling
Age, Gender, Minority Status,
And Family Income, & Age
Difference between Peers : r = .33***
Implications:
The “dominant deviant” may well be in a position to sway
larger groups of adolescents toward deviant behavior. But
who is most likely to be swayed by these teens?
III. Adolescents who are most easily influenced
by their peer (again in a non-deviance-
related task), are most likely to increase in
delinquency over the following year,
particularly non-violent delinquency:
Teen Susceptibility to Peer Influence was assessed via:
？ An experimental measure
；
Main Story Line - 6.
？ Coded from teens’ reactions to disagreement with a
friend (details presented below)
？ Yielding a measure of how easily the teen changes
positions when a peer disagrees with him/her.
Teen Non-violent
Delinquency (Age 14)
22 Β Total R;R
Step I. Teen Delinquency (13) .34*** .07*** .12***
Step II.
Gender (1=M; 2=F) -.03
Age .01
Family Income .09 .02 .13***
Step III.
Teen Susceptibility to Peer .18* .04* .16***
Influence
β weights are from variable’s entry into models.
Conclusions
？ Tight integration into normal adolescent peer culture –
as reflected by association with popular peers –
appears associated with increasing levels of deviant
behavior in early adolescence.
？ In dyadic peer interactions, the more dominant
adolescent tends to be the more deviant adolescent.
；
Main Story Line - 7.
？ Adolescents who are most easily influenced by their
peers are also most likely to increase in deviant
behavior over time.
Implications
There is significant evidence that the tendency of deviant adolescents to dominate in discussions with their less deviant friends may help foster a culture in which adolescents perceive. This study suggests several processes that may account for this effect.
These results suggest that concern about adolescent peer group effects may not just be limited to groups composed primarily of rejected/deviant peers. Even within normal, heterogeneous groupings, deviant adolescents may dominate, and may help shift overall adolescent peer culture in the direction of supporting delinquent activity. This suggests one way by which peer culture may account for the normative increases in delinquency observed from early to mid-adolescence.
Limitations
？ We studied low-level “normative” delinquency, not
violent or hard core criminal behavior.
？ Now need to address fundamental question:
Why are delinquent adolescents more likely to be
dominant in peer interactions?
？ Lower levels of impulse control/inhibition?
？ Implicit peer valuing of deviant behavior and deviant
teens?
；
Supporting Details - 8a
Acknowledgments
We would like to thank the National Institute of Mental Health for funding provided to Joseph Allen, Principal Investigator (Grant # R01 MH 58066) for support for the conduct and write-up of this study.
The authors can be reached at the Department of Psychology, Box 400400, University of Virginia, Charlottesville, VA 22904-4400 or allen@virginia.edu.
Copies of this and related papers are available from our web site:
http://faculty.virginia.edu/allen
；
Supporting Details - 9a
Methods
Sample
N = 148 Adolescents Along with their Parents
and Best FriendsAssessed Annually,
Beginning at Age 13
？ Equal numbers of Males and Females
？ Community-based Sample from a small urban
area.
？ 31% African American; 69% European
American
？ Highly Socio-economically Diverse (Median
Family Income= $38,000)
Measures
Sociometric Data
Open-ended nomination procedure. Youths and
their peers were asked to name 10 students with whom they would most like to spend time on a Saturday evening, and 10 with whom they would least like to spend time.
Sociometric data were obtained regarding:
？ Target Teen
？ Their Closest Friend
；
Supporting Details - 10a
？ A representative of their broader social circle
th Closest Friend) (their 4
Mean Ratings were standardized within students’ grade in school
# of “Like”
Nominations Person Rated s.d.
Target Teen 0.96 1.36
Closest Friend 1.31 1.47
th4 Closest Friend 1.41 1.60
Friends received more “liked” nominations than target teens, partly because by definition they are persons being selected as popular (at least to our target teens).
Delinquency Assessment
Total Delinquency was assessed via a slightly modified version of the self-report of delinquency (Elliott, Huizinga, & Menard, 1989), which provides a list of 16-possible types of delinquent acts and captures the frequency with which those acts were committed in the prior 6 months.
Non-violent delinquency can also be assessed by summing 10-items related to non-violent
delinquent acts from this measure. We used this
；