The Teen Court Peer Influence Scale (TCPIS): Determining an Effective Way to Measure and Model Positive Peer Influence

• Author: Scott Smith,Jill M. Chonody
• DOI: 10.1177/1541204009353159
• Published date: 01 April 2010
• Date: 01 April 2010
• Subject Matter: Articles

Youth Violence and Juvenile Justice
8(2) 148-159
The Teen Court Peer Influence
ª SAGE Publications 2010
Reprints and permission:
Scale (TCPIS): Determining an
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1541204009353159
http://yvjj.sagepub.com
Effective Way to Measure and
Model Positive Peer Influence
Scott Smith,1 and Jill M. Chonody2
Abstract
This article describes the development and validation of a scale designed to measure positive peer
influence. Participants from six teen court sites completed an anonymous survey (N ¼ 202), which
included the newly developed Teen Court Peer Influence Scale (TCPIS). A confirmatory factor
analysis (CFA) was conducted to test three measurement models to determine the best fit to the
data. The final measurement model included three reciprocal factors: positive cognitions, positive
identity
development,
and
modeling
positive
behaviors.
The
model
illustrated
strong
psychometric properties and provides a framework that may be used to augment practice.
Keywords
teen courts, positive peer influence, measurement development, theory, confirmatory factor
analysis
Peer influence has been conceptualized and evaluated in a multitude of ways to garner a better
understanding of the conditions under which youth are likely to be influenced (Deptula & Cohen,
2004). Gaps in the literature exist with regard to the role that positive peers play in developmental
outcomes, and scholars conclude that studies that include the measurement of peer relationships in
the juvenile justice system are one of our utmost necessities (Osgood & Briddell, 2004). Those
individuals who counter the values of a negative peer group may reinforce alternative beliefs
through positive peer influence.
For example, the relationship between negative peer influence and substance use has long been
established. If peer influence can promote drug and alcohol experimentation, then it should also be
able to discourage it (Swadi & Zeitlin, 1988). In addition to curbing substance use, research from one
study indicated that adolescents who spent more time with positive peers had higher parent and
teacher ratings of behavior change, greater improvements in social problem solving, and a reduction
1Department of Rehabilitation, Social Work, and Addiction, University of North Texas, Denton
2School of Social Work, Temple University, Philadelphia, Pennsylvania
Corresponding Author:
Scott Smith, Department of Rehabilitation, Social Work, and Addiction, University of North Texas, Denton, TX 76203.
Email: Kenneth.Smith@unt.edu
148

---

Smith and Chonody
149
in aggressive behaviors (Conduct Problems Prevention Research Group, 1999). Moreover, in a study
of 6th to 12th graders (n > 500,000), outcomes offered that youth may acquire up to 40 developmen-
tal assets that can assist them in becoming mature and socially responsible young adults (see Scales,
1999, for more details). Results showed that as the number of assets increased, risk behaviors (e.g.,
use of drugs and alcohol) decreased and positive outcomes, such as school success and physical
health, were reported more frequently.
Youth involved in the juvenile justice system typically report struggles associated with acquiring
these developmental assets. Moreover, research has demonstrated that individuals within the system
have a greater need for positive peer influence as they are more vulnerable to repeat offending
reinforced by previous peer relations (Hirschi, 2005). In turn, the need for research that sheds light
on the impact of positive peer influence with this population is warranted.
The lack of research on positive peer influence is attributed to the scarcity of available measures
and programs incorporating positive peer influence. To fill this conceptual and empirical gap, a mea-
surement strategy along with programming that incorporates positive peer influence is necessary.
First, instruments designed to assess the effects of positive peer influence are a must. Unfortunately,
available scales for this substantive area primarily measure factors related to negative peer influence.
Moreover, programs that use positive peer influence, such as teen courts, as an intervention approach
cannot be fully evaluated without appropriate avenues for measurement. Teen courts provide a
unique opportunity to test positive peer influence because they present ‘‘an alternative sentencing
option for first-time non-violent juvenile offenders wherein they are sentenced by a jury of their
peers’’ (Williamson County, 2006, p. 1; see Butts, 2002, for more details). Teen court advocates
describe positive peer influence as a major feature of the program (Dick, Pence, Jones, & Geersten,
2004), but little, if any, empirical evidence exists to support this claim. Lack of adequate measure-
ment strategies is a likely contributor to this absence in the literature. Indeed, a comprehensive
literature search did not yield any scales designed to measure positive peer influence nor offered any
appropriate measures for teen court practices.
One theoretical perspective that offers a foundation for measurement development and that has
been used throughout literature to evaluate peer influence and its impact on negative behavior is
social learning theory (SLT). According to SLT, the learning process for criminal behavior occurs
through four principle mechanisms:
(1) Differential association (direct and indirect interaction with others), (2) differential reinforcement
(instrumental learning through rewards and punishers), (3) imitation (observational learning), and (4)
cognitive definitions (attitudes) that are favorable or unfavorable, functioning as discriminative (cue)
stimuli, for the behavior. (Akers & Lee, 1996, p. 318)
Differential associations have been measured in various ways. A common approach is to inquire
about the behavior of the participants’ friends. For example, Akers and Lee (1999) used a 3-item
scale to solicit responses about the number of friends who had smoked marijuana. Similarly, another
study assessed the influence of peer associations on cigarette smoking with questions such as how
many of their best (male/female) friends smoked cigarettes, how many of the friends (male/female)
they have known the longest smoke cigarettes, and how many of the friends whom they are around
the most smoke (Krohn, Skinner, Massey, & Akers, 1985).
Differential reinforcement has been operationalized in one study as an individual’s perceptions
regarding the effects of smoking cigarettes (Akers & Lee, 1999). External sources of reinforcement,
from parents and peers, were also included. Measurement of imitation or the effects of observational
learning targets information about the behavior of significant people in the respondent’s past. For
example, in a study of partner violence, imitation was measured by determining the total number
of role models a person had that hit, slapped, punched, or kicked a partner in a disagreement (Sellers,
149

---

150
Youth Violence and Juvenile Justice 8(2)
Cochran, & Branch, 2005). Finally, for cognitive definitions, researchers use questions to determine
whether an individual or their peers have favorable or unfavorable definitions toward a specific
behavior. For example, an item from a study on digital pirating stated: ‘‘I think it is okay to use
copied software because the community at large is eventually benefited’’ (Higgins & Makin,
2006, p. 21).
Although this is an important contribution to the field, most of these measurement strategies are
solely focused on the assessment of negative peer influence. Scholars also state that the generality of
social learning constructs is problematic for testing and evaluative purposes. Specifically, the
challenge is how to pinpoint the sources of reinforcement that occur in a youth’s day-to-day life
(Warr, 2002).
The effects of positive peer influence cannot be assumed to be known by the absence of negative
peer influence. Negative influence and positive influence may not occur via the same process, thus
measurement strategies should address them separately to avoid undue error and potentially skewed
outcomes. In conclusion, research needs to begin to develop measurement options that can capture
positive peer influence variables, instead of measurement options focused on the mere absence of
negative influence.
The purpose of this article is to fill this gap by describing a research design that develops and tests
a scale of positive peer influence, the Teen Court Peer Influence Scale (TCPIS). The development
process was...