University of Southern California Dissertations and Theses

Social network influences on depressive symptoms among Chinese adolescents

(USC Thesis Other)

Social network influences on depressive symptoms among Chinese adolescents

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

SOCIAL NETWORK INFLUENCES ON DEPRESSIVE SYMPTOMS AMONG CHINESE ADOLESCENTS

by
Janet Okamoto

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PREVENTIVE MEDICINE (HEALTH BEHAVIOR))

August 2010

Copyright 2010 Janet Okamoto
ii
DEDICATION

To my family for their unconditional support and faith in me.

iii
ACKNOWLEDGEMENTS

I would like to thank my committee members, Drs. Adam Leventhal, Joel Milam, Mary Ann
Pentz, and David Schwartz for their time, support, and assistance. I would especially like to
thank my mentor and committee chair, Dr. Thomas Valente for his guidance, advice, the
opportunities he has given me and for being a wonderful teacher.
I am grateful to the wonderful faculty, past and present, at the Institute for Health
Promotion and Disease Prevention Research (IPR) who have helped me along the way, with
special thanks to Drs. C. Anderson Johnson, Paula Palmer, Ping Sun, and Jennifer Unger. I would
also like to acknowledge the administrative staff at IPR who is so good at helping out with the
small, everyday details that were necessary to move through this process.
I would also like to thank all of the many colleagues I have met on my journey through this
program, including those in China who made this dissertation possible with their efforts to make
the TTAURC intervention project happen and who collected all of the data that was used for
these studies. They not only taught me a lot about the Chinese culture, but about patience,
understanding, and the power of cooperative effort. I am also grateful to all of my fellow
doctoral students throughout these years, who helped to challenge me and also lend an ear to
listen or a shoulder to lean on.
This research was supported in part by the Transdisciplinary Tobacco and Alcohol Use
Research Center (5P50-CA 084735) and a Cancer Control and Epidemiology Research Training
Grant (5T32CA009492-24

iv
TABLE OF CONTENTS
Dedication ii
Acknowledgements iii
List of Tables v
List of Figures vi
Abstract vii
Chapter 1: Introduction 1
Specific Aims 3
Background & Significance 5
Peer Relationships in Adolescence 6
Depression in Adolescence 7
Peer Relationships and Depressive Symptoms 9
Theoretical Overview of the Influence of Peers on Depression 13
Parent Study 19
Conceptual Model Overview 20
Conclusion & Chapter Summary 23
Chapter 2: Social integration and depression among adolescents: Associations between
social network status and depressive symptoms in a Chinese sample
24

Introduction 25
Methods 28
Results 34
Discussion 40
Chapter 3: Peer Contagion and Adolescent Depression: The Effects of Peers’ Depressive
Symptoms and Social Network Status
47

Introduction 48
Methods 53
Results 59
Discussion 68
Chapter 4: The Classroom Social Network and Depressive Symptoms: Do Network
Composition and Characteristics Influence Adolescent Depression?
75

Introduction 76
Methods 79
Results 85
Discussion 89
Chapter 5: Discussion & Conclusions 95
Bibliography 109
v
LIST OF TABLES

Table 1: Study One Sample Characteristics 35
Table 2: Standardized coefficients (β) for Three Models of Social Network Status and
Depressive Symptoms by School Type & Gender
38

Table 3: Adjusted Odds Ratios for Three Models of Social Network Status and
Depression (CES-D ≥28) by School Type & Gender
39

Table 4: Study Two Sample Characteristics 59
Table 5: Standardized Coefficients for Multilevel Models of Friend Depressive
Symptom on Depressive Symptoms
63

Table 6: Odds Ratios for Multilevel Models of Friend Depressive Symptom on
Depressive Symptoms
64
Table 7: Standardized Coefficients for Multilevel Models of Friend Depression on
Depressive Symptoms
66

Table 8: Odds Ratios for Multilevel Models of Friend Depression on Depression 67
Table 9: Standardized Coefficients for Multilevel Models of Friend Social Network
Status on Depressive Symptoms
69

Table 10: Study Three Sample Characteristics 86
Table 11: Standardized coefficients (β) for multilevel models for effects of social
network measures on classroom depression
88

vi
LIST OF FIGURES

Figure 1: Adapted Main Effects Theoretical Model 21
Figure 2: Ecological Perspective: Levels of Social-Contextual Analysis 22
Figure 3: Basic Conceptual Model 23
Figure 4: Friendship Network Diagram 31
Figure 5: Most Liked Network Diagram 32
Figure 6: Examples of Centralized and Decentralized Networks 82
Figure 7: Distribution of Peer Status Categories (Friend Nominations) 83
Figure 8: Distribution of Peer Status Categories (Most Liked Nominations) 84

vii
ABSTRACT
The goal of this dissertation was to investigate the impact of social network influences on
depression in adolescents. Three studies assessed this goal by examining the social
determinants of depressive symptoms among adolescents at different levels of analysis: the
intrapersonal or individual level, the friendship group level, and the social network level. Study
One tested a main effects model of depression at the individual level, hypothesizing that social
integration would be associated with depressive symptoms. It investigated whether network
status was related to depression and depressive symptoms. Study Two moved to the next level
of analysis and looked at depressive contagion. This was assessed by testing whether the level of
friends’ depressive symptoms and friend social status was related to an adolescent’s own
depressive symptoms. Finally, Study Three explored the network level by looking at the
classroom social network structure and characteristics and examined how these factors work to
influence depression outcomes. Results from the statistical models in the three studies indicated
that social network influences are associated with depression-related outcomes. However, this
basic pattern varied in relation to gender, school type, and social integration at the classroom,
peer group, and individual levels. General results indicate that individuals who are better
integrated into their social networks, who have fewer depressed friends, and whose classroom
social networks contain fewer peripheral members are less likely to report depressive symptoms
and depression. The findings from this dissertation suggest boys and girls and those in different
school types may be oriented to different contexts and types of social networks (e.g. friend
versus most liked networks); therefore, specific conceptualizations of social network influences
should be carefully formulated and assessed in relation to the particular contexts that are
relevant and meaningful to an adolescent. This work represents early steps in the investigation
viii
of the complex social environment in which adolescents must navigate and which are so
meaningful for healthy development. The ultimate goal of such work is to glean useful
constructs upon which prevention interventions can operate.

1
CHAPTER 1: INTRODUCTION
The prevalence of depressive symptoms increases substantially during adolescence,
particularly between ages 15 through 18 (Birmaher, Ryan, Williamson, & Brent, 1996; Petersen
et al., 1993). The incidence of depression similarly increases from adolescence to early
adulthood (Pine, Cohen, Brook, Gurley, & Ma, 1998).These high rates of depression and
depressive symptoms in adolescence has led many investigators to view depression as the most
serious mental health problem of the adolescent years (Aseltine, Gore, & Colton, 1994). Thus, it
is particularly important to identify and study etiologic factors, such as social processes and
cognitions, which may contribute to the development of depression in adolescence.
Peer relationships are critical to the social development of an adolescent (e.g.
Bronfenbrenner, 1979; Furman & Burhmester, 1992; Hartup, 1995). They also play a prominent
role in identity development, through peer experiences such as social comparison, reflected
self-appraisal, and peer group affiliation (Harter, Stocker, &Robinson, 1996). Membership,
acceptance, and healthy functioning in larger peer groups are also acutely important to
adolescents’ psychological well-being (Brown & Lohr, 1987). Adolescents spend almost one-third
of their free time interacting with peers, more than twice the time they spend with parents
(Csikszentmihalyi & Larson, 1984; Larson & Richards, 1991; Larson, Richards, Moneta, Holmbeck,
& Duckett, 1996), resulting in increased feelings of closeness and intimacy with peers during
adolescence (Furman & Buhrmester, 1992). Social interactions also become more emotionally
rewarding for youth (Larson & Richards, 1991). Peers are the primary components of an
adolescent’s social network and are relied upon more heavily as sources of support and advice
during this developmental period (Brown, 1990; Buhrmester, 1996). Therefore, problems in the
domain of peer relationships and social environment are likely to be particularly distressing
during adolescence (Hammen, 2009; La Greca & Harrison, 2005; La Greca & Lopez, 1998).

2
Individuals, dyads, and networks make up the intricate layers of the social world
adolescents inhabit, with peer relations being weighted differently across development.
Cillessen (2007) aptly noted that the unique processes at each of these three levels of analysis
have been found to be “uniquely affected by and have unique effects on social development.”
Several studies have identified structural features of social networks in childhood and
adolescence and examined the effects of social network membership on various outcomes (e.g.
Cairns, Cairns, Neckerman, Gest, & Gariépy, 1988; Kindermann, 1996; Urberg, Degirmencioglu,
& Pilgrim, 1997; Kiesner, 2002; Ueno, 2005). Social networks have been found to be an
important environment in which socialization occurs, for both positive and negative
developmental outcomes. For example, social networks have been associated with antisocial
behavior and delinquency (Dishion, McCord, & Poulin, 1999), but also with positive academic
outcomes (Espelage, Holt, & Henkel, 2003).
The majority of research examining the relationship between peer relationships and
psychosocial functioning has focused on this relationship in middle childhood and pre-
adolescence. In comparison, only a few studies have explored this relationship in detail during
adolescence (e.g. Chan & Poulin, 2009; La Greca & Harrison, 2005; Stevens & Prinstein, 2005;
Ueno, 2005). However, the link between peer relationships and depressive symptoms is not
necessarily the same for all age groups (Petersen et al., 2003; Ueno, 2005), and the
developmental changes in the quality and importance of friendships during adolescence make
understanding this association for this time of life particularly important (Rudolph & Hammen,
1999).
The relationship between social networks and depression has been studied in elderly
(Holahan & Holahan, 1987; Seeman et al., 1987), immigrant (Vega, Kolody, & Weir, 1991), and
post-pregnancy (O’Hara, Rehm, & Campbell, 1983) populations, but there has been a lack of

3
research using these concepts with adolescent populations to study depression. This represents
a large gap in the literature, considering both the significant influence peer relationships have
on healthy functioning and the dynamic, flexible nature friendship ties and social network
composition can have during adolescence.
Understanding the etiology of depression is one of the most important goals of depression
research, with adolescence emerging as a key developmental period to study depression and
depressive symptoms. Interpersonal theories of depression assert that without attention to
interpersonal functioning, depression cannot be fully understood (Coyne, 1976; 1999;
Lewinsohn, 1974; Cole, 1990). These interpersonal theories fit well within an ecological
framework, which focuses attention on both individual and social environmental factors as
important determinants of health behavior (Bronfenbrenner, 1979). Similarly, a social network
perspective is also distinguished by a focus on context, and the understanding that an individual
is embedded within an environment that influences, and is influenced by, the person. Network
analysis makes it possible to understand an adolescent within their “real-world” social
environment and provides an opportunity to observe the social structure, social influence, and
socialization processes that define interpersonal functioning. The major goal of this work is to
investigate the impact of social network influences on depression in adolescents from
different levels of network analysis. The relationships between social network factors and
depressive symptoms will be tested in a large sample of Chinese adolescents. In order to assess
these relationships, a specific aim at each level of social network analysis will be examined:
Aim One: Individual level
• To examine whether different measures of peer status are associated with depression
outcomes.

4
Aim Two: Friendship/Popularity Group Level
• To determine if the depressive symptoms of an adolescent are influenced by the
depressive symptoms and peer status of their self-reported friends.
Aim Three: Classroom Level
• To discover if classroom network composition and characteristics are associated with
students’ depression outcomes.

5
BACKGROUND AND SIGNIFICANCE
Adolescent depression has emerged as a significant and challenging research topic in the
past two decades, with important and far-reaching implications (see Petersen et al., 1993;
Birmaher et al., 1996a; 1996b; Costello, Erkanli, & Angold, 2006 for reviews). For example, due
to the high prevalence and chronic nature of depression, the World Health Organization (2008)
found it to be the third most burdensome disease in the world, with projections of it becoming
the most burdensome by 2030. Recent findings regarding noticeable increases in the prevalence
of depression and depressive symptoms during adolescence as well as the important predictive
role of depressive cognitions in the onset and maintenance of adolescent and later adult
depression, have led to a corresponding increased interest in the possible antecedents of these
outcomes in order to inform more effective prevention and intervention efforts (Burke, Burke,
Rae, & Regier, 1991; Klerman, 1988; Garber & Flynn, 2001). Understanding the etiology of
depression is one of the most important goals of depression research (Kistner, 2006).
It has become clear that depression has an impact on adolescent adjustment and
psychological functioning (Petersen et al., 1993). Depressed mood and depressive symptoms
have been found to explain over two-thirds of the variance in adolescent well-being (Mahon &
Yarcheski, 2001). These and other findings (e.g. Harrington, 1998) indicate that even relatively
mild depressive symptoms can result in impaired functioning. Therefore, those with subclinical
depressive symptoms during adolescence also need to be considered at increased risk for
negative outcomes, both during this developmental period and into adulthood.
Adolescence is a critical period in social development, marked by an expansion of peer
networks, increased importance of close friendships, and the emergence of romantic
relationships. Social networks increase not only in size, but in complexity and peer group
affiliations become a more important aspect of relating to peers (La Greca & Prinstein, 1999). A

6
transition from parents to close peers of the primary social support role is another important
change that can have important contributions to an adolescent’s self-concept and identity
development (Furman & Buhrmester, 1992).
The number of studies investigating characteristics and experiences of childhood and
adolescence as possible causes and contributors to depression has increased noticeably (Kistner,
2006). Most of the studies have focused on the contribution family factors have on the origins of
depression. And, while familial relations are critical in the etiology of depression, relations
outside the family, particularly with peers, need greater attention and consideration, especially
during adolescence. The detrimental effect of poor interpersonal functioning is certainly not
specific to depression; it has been associated with numerous negative developmental outcomes.
Therefore, comprehension of the developmental processes associated with depression is
essential. Those viewing risk and protective factors from a developmental perspective have
been particularly interested in depressive disorders due to their complexity and the unique
synthesis of psychological, social, and biological determinants involved.
Theory and research suggest that the increases in depressive symptoms observed during the
transition to adolescence are probably related in some way to the significant changes that occur
in interpersonal functioning during this same time (Hankin & Abramson, 2001; Rudolph &
Hammen, 1999).

Peer Relationships in Adolescence
It is well established that peer relationships and the social environment have an important
role in adolescent psychological and behavioral development (see Crosnoe & McNeely, 2008;
Hoffman, Sussman, Unger, & Valente, 2006; Kobus, 2003 for reviews). As adolescents become
increasingly independent from their families and oriented more towards peers, susceptibility to

7
peer influences increases (Aseltine, 1995; Chassin, Presson, Sherman, Montello, & McGrew,
1986; Glynn, 1981). Another potential result of the more extensive and intimate connections to
peers is the influence friends’ psychological and behavioral characteristics can have an
adolescent’s own functioning. Peer contagion has been studied primarily in regards to
externalizing behaviors, such as aggression (Cohen & Prinstein, 2006), deviant and antisocial
behavior (Prinstein & Wang, 2005), and other health risk behaviors (Dishion, McCord, & Poulin,
1999; Kandel, 1978; Heilbron & Prinstein, 2008). Fewer studies have looked at peer contagion of
internalizing problems among adolescents (Steven & Prinstein, 2005; Prinstein, 2007), but
findings from these select studies support the idea of peer contagion of depressive symptoms.
Peer relationships among adolescents are often organized in a hierarchical manner. The
daily social world of youth often carries a specific social order with rewards and penalties
depending on the skill with which an adolescent navigates this environment. Differences in
health by social status have been found among adolescents, showing a similarity to the adult
research on this topic. For example, peer status has been found to have effects on both physical
health and general feelings of well-being (Hansell, 1985). The peer status of youth has also been
linked to mental health and negative health behaviors in young adulthood (Michell & Amos,
1997; Ostberg, 2002; Roff & Wirth, 1984). Both friendship ties and the overall structure of the
school classroom, in addition to individual social status, have also been found to influence
health (Ueno, 2005; Kiesner, 2002). This suggests that the organization, structure, and
distribution of social status among a peer group can influence adolescent development.

Depression in Adolescence
Much of the research on depression has been conducted with adult and aging populations.
Developmental psychopathologists have focused attention on childhood and early adolescent

8
predictors for later depression, but fewer studies have looked at the etiology and health
implications of depression and depressive symptoms during adolescence. Depression in youth
has been associated with poor academic performance, social dysfunction, substance abuse,
suicidal ideation, and attempted and completed suicide (Institute of Medicine (IOM), 1994;
National Research Council & Institute of Medicine (NRCIOM), 2009). It has been suggested that
psychological problems, such as depression, may form the root of health problems for
adolescents (NRCIOM, 2009). Unfortunately, research examining the effects of mental health
problems on other health issues in adolescence has not been a top priority in the past, so the
true costs of depression during this developmental period are hard to accurately estimate (IOM,
1994; NRCIOM, 2009).

Adolescent Depression and Negative Health Outcomes. Study of the relationship between
depression and health outcomes has been a major focus in depression research. This is primarily
due to the concentration on finding cost-effective ways to treat and prevent depressive
disorders, as well as the many co-morbid health conditions associated with them. The results of
this work suggest that depression during adolescence is related to a variety of health-related
problems, both concurrently and into adulthood. For example, one study alone found
depression to be associated with low energy and fatigue, poor sleep, loss of appetite, aches and
pains, headaches, stomachaches, and heaviness in the chest in an adolescent sample
(Mondimore, 1990). In addition, longitudinal studies have linked depression to other problem
behaviors, such as higher rates of smoking, alcohol use, substance abuse, unhealthy eating, risky
sexual behavior, and less physical activity (Franko et al., 2005; Haarasilta, Marttunen, Evans, &
Aro, 2004; Hallfors et al., 2004; Keenan-Miller, Hammen, & Brennan, 2007; Mahon & Yarcheski,

9
2001). Furthermore, depression is the major risk factor for suicide, which is one of the five
leading causes of death in adolescents worldwide (Wasserman, Cheng, & Jiang, 2005).
Determining the circumstances leading to depression in adolescence, and from depression
to negative health behavior outcomes, will have important theoretical implications regarding
etiology as well as clinical and public health implications. Depression and depressive symptoms
during adolescence has far-reaching consequences as well as the more immediate negative
outcomes. One recent longitudinal study found an association between adolescent depression
and later medical and social costs, including more visits to medical professionals and impaired
work functioning due to physical health in young adulthood, even after controlling for current
depression (Keenan-Miller, Hammen, & Brennan, 2007). These findings provide further support
for the importance of examining risk factors for adolescent depression.

Peer Relationships and Depressive Symptoms
Due to better understanding of the importance peer relationships have in adolescent
development, an increasing number of studies have examined the associations between peer
relations in adolescence and internalizing problems, such as depression (Aseltine, Gore, &
Colton, 1994; Chan & Poulin, 2009; Hogue & Steinberg, 1995; Mahon & Yarcheski, 2001). These
symptoms of depression are becoming more common among adolescents and may be a risk for
later impaired functioning in adulthood (Keenan-Miller, Hammen, & Brennan, 2007; Peirce, et
al., 2000). Research is also finding that depressive symptoms may be precursors to more severe
disorders, such as Major Depressive Disorder (MDD) and social anxiety disorder (Birmaher et al.,
1996a; 1996b).
Integration into a social network that provides social support has been found to have overall
beneficial effects for adults (Cohen & Wills, 1985). The association is assumed to be similar for

10
adolescents, but further details of the influence peer processes have on a variety of health
outcomes need to be examined. Some have found that friendships among depressed
adolescents include lower levels of positive reinforcement and social support than other
friendships (Barrera & Garrison-Jones, 1992; Hammen & Brennan, 2001; Stice, Ragan, & Randall,
2004). Depressed mood and depressive symptoms have been linked to peer rejection (Kiesner,
2002; Prinstein & Aikens, 2004) and peer victimization (Hawker & Boulton, 2000; Nolen-
Hoeksema, & Girgus, 1994; Prinstein, Cheah, & Guyer, 2005) in adolescents. Quantity and
quality of adolescents’ friendships (Nangle, Erdley, Newman, Mason, Carpenter, 2003) have also
been related to depression. Overall, substantial evidence now exists to establish that the social,
emotional, and psychological adjustment of an adolescent is associated with the psychosocial
adjustment of their peers, particularly close friends.

Social Networks and Depression. An individual is embedded within an environment that
influences, and is influenced by, the person. A wealth of research exists that emphasizes the
importance of the social context on health. In order to understand these social contextual
influences on health behavior, social networks have been identified as a potentially rich source
of information that can provide greater insight into the complex relationships between social
processes and health outcomes. Social networks are often defined simply as a set of actors or
people who are connected by a set of ties, or relations. An example of the network ties of a
person would be their familial, professional, friend, and community relationships. Healthy
integration into a social network has been related to physical and mental well-being (Ostberg,
2003; Ueno, 2005), while the type, quality, and quantity of support provided by one’s social
network have been linked to mental distress (Rosenberg & McCullough, 1981). Some network
researchers have suggested that the link between social network integration and mental health

11
symptoms, including depressive symptoms, could work through promoting positive attitudes
and feelings about one’s relationships with others and one’s ability and skill in fostering these
relationships (Hansell, 1985; Ueno, 2005). Given the importance of social integration and peer
influences, many researchers have turned to formal network analysis to understand peer
network structures.
There are a number of ways that a network can be described. The size of one’s network is
one of the most common measures examined. It would seem that those with larger networks
have a greater potential for social support, and to some extent this is supported by the
literature. Larger egocentric networks have been associated with better mental health (Coates,
1985; Field, Diego, & Sanders, 2001; Gest, Graham-Bermann, & Hartup, 2001). For example,
larger networks are associated with fewer depressive symptoms (Cohen & Wills, 1985).
However, some have also found that larger networks may create the potential for more
negative interactions (Stokes, 1983) and that beyond a certain network size, larger networks
may not be helpful for an individual (Polister, 1980). A few studies have looked at other
measures of network structure and found that density is also important to measures of well-
being, and may be more important than network size (Kadushin, 1983; Berkman, Glass,
Brissette, & Seeman, 2000).
Social network research is usually interested in the structure or the network and the effects
of relationships between people and groups, however, more recently, the effect of individual
psychological characteristics on social networks and network characteristics on individual’s
psychological functioning have been examined (Kalish and Robins, 2006; Mehra, Kilduff, & Brass,
2001). For example, one study examining the multiple types of networks found lower
neuroticism, characterized by traits such as depression and anxiety, was associated with higher

12
degree centrality (Klein, Lim, Saltz, & Mayer, 2004). This has been supported by other work that
found those high in neuroticism have fewer people who provide social support (Stokes, 1983).
Research examining social networks of adolescents has naturally focused on school and
classroom networks. Youth who are rejected or neglected have been found to have higher levels
of depression than those who are more accepted by their peers (Hecht, Inderbitzen, &
Bukowski, 1998; Lansford et al., 2007). Popular individuals are less likely to become depressed
than those who are unpopular (Prinstein & La Greca, 2002). Findings from a study looking at
social network integration among adolescents found an association with good mental health
that resulted from positive social cognitions (Cauce, 1986) Others have provided additional
evidence that point to adolescent’s peer relationships having an association with depressive
symptoms and with social cognitions (Jacobson, 2007), suggesting that how an adolescent
interprets their social world and what they think of their ability to navigate within that
environment plays a significant role in vulnerability to depression and could possibly mediate
the relationship between social status and depressive symptoms.
Most of these studies with adolescents use peer nominations of “friends” or, more
commonly, those who are “well liked.” Some have pointed out that these are not two measures
of the same construct, however, as the influences of an adolescent’s friends and the popular
crowd may not be the same. For certain outcomes friends may be more influential, but for other
health behaviors the behavior of the popular crowd may hold sway. It is reasonable, then, that
nomination measures of popularity and friendship are generally only moderately correlated and
have been found to have independent effects on such things as emotional development (Gest,
Graham-Bermann, & Hartup, 2001). Other studies have looked at larger adolescent networks
and found few effects of school-level network measures on depressive symptoms, indicating

13
that the more immediate classroom and friendship networks have more impact than the larger,
less intimate social environment (Fischer, 1982).
One limitation in the health behavior research using social network approaches is that much
of it examines only a few measures of social networks, such as network size and sociometric
status. Therefore, it is difficult to fully understand social network influences on depression
because of the lack of diversity in measures used by current studies. Network analysis has an
array of different measures that could potentially provide important information on the
relationship between social processes and depression in adolescents.
Establishment of healthy social networks is a critical and developmentally appropriate task
of adolescence. However, for those individuals with underdeveloped or dysfunctional networks,
the many stressors that accompany the changes of adolescence may potentially cause great
harm. Disrupted or poorly functioning social networks can fail adolescents in times when they
seek and perhaps need the most support. They can reinforce thoughts of worthlessness,
rejection, failure, and hopelessness, increasing the risk of developing depressive symptoms
(Davidson, Rieckmann, & Lesperance, 2004; Wade & Kendler, 2000). Further study of the peer
processes of adolescents through social network measures is needed to understand how
dysfunction occurs and the means by which it influences depression.

Theoretical Overview for the Influence of Peers on Depression
Of the various theoretical explanations developed to describe and predict depression, main
effects models of social relationships and depression may be the most relevant for the
developmentally-driven influences on adolescent mental health. These models also fit well
within both ecological and social network perspectives, which add depth to existing models by
highlighting contextual and structural factors in the social environment.

14
Main Effects Model of Social Relationships and Depression.
The link between social isolation and poor psychological well-being is well established; with
poor quality relationships, small social networks, and few close relationships all linked to
depressive symptoms. Cohen and Wills (1985) proposed two models by which social
relationships influence health outcomes, such as depression. The first model is the stress-
buffering model, which proposes that social relationships influence well-being only for people
under stress. The main effects model, however, states that social relationships have a beneficial
(or deleterious) effect regardless of whether a person is under stress. Much of the work
examining Cohen and Will’s models regarding depression-related outcomes has focused on the
stress-buffering model. However, more recent work has suggested that the structural aspects of
social ties, such as social network factors, operates through the main effect model rather than
the stress-buffering model. Kawachi & Berkman (2001) stated, for example, that “the perceived
availability of functional support is thought to buffer the effects of stress by enhancing an
individual’s coping abilities. In contrast, the degree of integration in a social network is
postulated to have a direct effect on well-being regardless of the presence of stressful
circumstances.”
Others have pointed to the inability of stress-buffering models to fully explain the
magnitude of the relationship between social ties and health outcomes (for example, see
Hammer, 1983). They have posited that this is an indication that social network factors are more
directly involved in psychological and physical well-being than stress-buffering hypotheses
assume. In Hammer’s (1983) study, it was found that “people in very poor health do tend to
have smaller social networks than people in better health; but in every health category, people
with higher social network scores have lower mortality rates over the subsequent 9 years,”
which was suggested as further support for the relevance of a main effect model.

15
Social Contagion Theories.
Contagion theories seek to understand social networks as pathways for the spread of
“infectious” attitudes, norms, beliefs, and behaviors. A core assumption of these theories is that
contact between members of a social network serves as the mechanism that exposes
individuals, groups, and networks to information, attitudes, norms, beliefs, and behaviors of
others (Burt, 1980; Carley, 1991). Network factors such as frequency of interaction, strength of
ties, asymmetry, and stability can impact the extent to which others within a network can
influence an individual and each other. Contagion theories are related to a number of other
theories, such as Symbolic Interaction, Gatekeeping Theory, and Structurational Theory, which
all focus on the influence of social networks and social structures on behavior. These theories do
not require intent or even awareness to influence to be present, only the exposure itself.
Support for these theories can be found in the robust findings in the peer relations literature
regarding similarities between adolescents and their friends’ beliefs, behaviors, and
psychological characteristics (for examples, see Keenan, Loeber, Zhang, Stouthamer-Loeber, &
Van Kammen, 1995; Prinstein, 2007; Prinstein, Boergers, & Spirito, 2001).
Social Networks.
Analysis of social networks has become an increasingly popular way to examine social
relationships. However, a clear definition of social network theory is difficult to determine.
There are many explanations of what constitutes social network theory, but no real consensus.
Many of these explanations describe social network analysis and the concepts used in this
analysis. A general description of the theory often given is that it focuses on the relations among
actors, where actors can consist of individuals, organizations, or other components of a social
environment. Perhaps this theoretical ambiguity stems from the multidisciplinary background of
this body of research and the varied way these different disciplines view theory. Major

16
contributions to social network analysis have come from almost every major scientific field of
study; including, sociology, anthropology, mathematics, physics, communication, business, and
public health. It may be more appropriate to refer to this as a theoretical perspective or
orientation than to represent it as a single theory.
No matter the background or terminology, two of the main theoretical precursors of
network theory are mathematic graph theory and social theory. Both are very broadly defined
and suffer similar conceptualization problems as social network theory, but they provide the
foundation from which much of this work has developed. Graph theory is the formal
mathematic study of graphs as a representation of relations, either symmetric or asymmetric,
between discrete objects. Social theory has been described as the use of theoretical frameworks
to “study and interpret social structures and phenomenon within a particular school of thought”
(Harrington, 2005). If viewed as a social network theoretical perspective then there are
numerous theories that could be categorized under this designation, including Social
Comparison Theory, Social Identity Theory, Contingency theory, Social Exchange Theory, and
Critical Mass Theory.
According to the network perspective, individuals are “embedded within networks of
interconnected relationships that provide opportunities for and constraints on behavior” (Brass,
Galaskiewicz, Greve, & Tsai, 2004). This perspective takes a different view than much of the past
work examining adolescent health behavior, which typically examines individual actors in
isolation. It also has multiple levels of analysis, from pairs of close friends to peer groups to
classrooms within schools to schools within a district and beyond. Each level has properties that
are unique, but the properties and structure of one level can have an influence on other levels.
One of the basic tenets of the social network perspective is that social structure matters and can
be operationalized as patterns of interactions and relationships within social groups (Hinde,

17
1976). This network structure and an individual’s position in the social network can have
significant effects on how the group functions, social dynamics, and the individual’s skill and
efficacy regarding their social competence (Sih, Hanser, & McHugh, 2009). The metrics of social
network analysis provide the means by which to quantify this structure.
Social network analysis has a unique terminology that can prove challenging. Just a few of
the common terms from network analysis include actors, relations, attributes, and egocentric
and sociocentric. Actors, also called nodes, are the members of the network and can be either
individuals or collective units, such neighborhoods that make up a city or classrooms that make
up a school. Relations are the second necessary component of a social network and serve as the
ties by which actors are linked. Attribute data is information about the properties, qualities, or
characteristics of the actors. Egocentric networks are also called personal networks and are
defined from a focal actor’s (or ego’s) point of view. The ties leading to and from the focal actor
in the network are the primary interest when examining this type of network. Sociocentric
networks, also called complete networks, are the ties among members of a bounded group. The
information about the relations of all members of the group to each other is the focus and
usually the network’s structure along with social position and roles are of key interest.
Ecological Frameworks.
A theoretical framework that is complimentary to, and often overlapping with, the social
network perspective is Bronfenbrenner’s (1979; 1999) ecological systems theory. This is one of
the early, and still one of the most widely used, theoretical frameworks for understanding how
environment influences the behavior and experiences of an individual. The original models
viewed the individual within the center of concentric circles representing the ever-widening
levels of the environment. There were four original levels, with the microsystem representing an
individual’s immediate environment, such as the family, peers, or workplace. The mesosystem

18
connects the different microsystem components, for example the connection between an
adolescent’s teacher and parents. The exosystem is explained as the larger social system which
can indirectly influence an individual. The final level is the macrosystem and is defined as the
cultural context that is made up of norms, values, rules and laws, and customs.
There are many ecological theories in the social sciences literature. Cassel’s work (1976)
outlines another seminal ecological perspective. He delineated the importance of the social
environment, which consists of psychosocial factors that are generated by social interactions. In
his original work (1976), the central hypothesis states that the social environment can alter
susceptibility to disease by directly influencing neuroendocrine functioning. He went on to list
some psychosocial factors that make up the social environment, including isolation, the
hierarchical organization of social networks, and loss and grief. Much of the recent work on
social capital has been grounded in Cassel’s work.
Another, more recent theoretical model developed from an ecological perspective that is
relevant to describing the relationship between social interactions and depression is Ho and
Chiu’s (1998) model called Methodological Relationism. This theoretical view focuses not on the
individual or the context separately, but rather proposes that the unit of analysis is “person-in-
relations.” This means a focus on individuals interacting within a relational context. It maintains
that the social network of an individual must be included in any analysis of the individual’s social
behavior.
The delineation of different levels of analysis is similar in both ecological and social network
perspectives. The different levels of network analysis can be fit into the levels put forth by
ecological systems theory. The ecological systems can be translated into intrapersonal factors,
interpersonal processes, institutional factors, and community and cultural factors. When broken
down in this manner, using an ecological framework as a guide for examining the influence of

19
interpersonal relationships on adolescent health behavior by looking at social network
characteristics appears to be a natural collaboration.

Parent Study
The reform of social and economic systems in China over the past few decades has resulted
in significant challenges to traditional concepts, values, and lifestyles. Previous work has shown
that the psychological stress originating from these rapid changes in the form and function of
daily living has had increasing effects on both the physical and mental health of the Chinese
people (Li et al., 2006; Miller, 2006; Phillips, Liu, & Zhang, 1999; Smith, 1998). Adolescence is
already a time of rapid change, so it may be that adolescents are particularly vulnerable to the
health behavior effects of additional life stressors. To that end, the University of Southern
California Pacific Rim Transdisciplinary Tobacco Use Research Center (TTURC) was created to
investigate how social circumstances, cultural setting, dispositional phenotype, and genes might
influence the course of tobacco and alcohol use trajectories in youth, and how they moderate
the effectiveness of programs for tobacco and alcohol abuse prevention.
The studies presented in the following chapters examined data from Project 2 of the second
generation TTURC at USC that was implemented with a large sample of Chinese adolescents in
the city of Chengdu. Project 2, Effect Moderation and Mediation in Prevention Programs, sought
to fill current research gaps by enhancing understanding of the extrinsic and intrinsic factors
that modify the effects of prevention programs. A large intervention trial was conducted in
Chengdu high schools using a curriculum developed specifically for the trial that focused on
enhancing students ’ social decision-making and assessing the presence of interactive effects
between adolescents’ personal and social characteristics and program characteristics in the
prediction of prevention outcomes.

20
Chengdu is a densely populated and prosperous city in southwest China, with a population
of over 10 million people (United Nations Population Division). It is the capital of Sichuan
Province and is considered a progressive city and, due to this openness, has experienced some
of the most rapid changes in recent decades in China; moving from a primarily agricultural
provincial capital to an urban, industrial city. The city was selected as the site of the intervention
trial due to a strong collaborative relationship with the Chengdu Center for Disease Control and
Prevention (CDC), who provided unequivocal support for the project and also had the capacity
to carry out such a large research trial. The study was approved by the Institutional Review
Boards of the University of Southern California and the Chengdu Center for Disease Control.

Conceptual Model Overview
Peer relations constructs, such as low peer acceptance and social support, have been
associated with dysphoria and depressive symptoms among adolescents and have also been
found to predict later depression (Boivin, Hymel, & Bukowski, 1995; Coie, Terry, Lenox,
Lochman, & Hyman, 1995; Kupersmidt & Patterson, 1991; Panak & Garber, 1992). According to
main effects models of depression (Hammer, 1983; Kawachi & Berkman, 2000), interpersonal
functioning contributes significantly to the onset and maintenance of depressive symptoms.
Research utilizing these models has found promising support; for example, reduced use of a
social network has been associated with increasing depressive symptoms (Possel et al., 2005).
Further studies among adolescents can help to clarify how these models operate and suggest
factors that can be targeted in future prevention work.

21

This dissertation examined data from adolescents attending high schools enrolled in a large
smoking prevention trial in Chengdu, China to investigate interpersonal influences on depressive
symptoms. The main theoretical model guiding the proposed studies suggests a main effects
model of social relationships and depression, in which social network factors directly influence
adolescent depressive symptoms and depression (see Figure 1). The model presented in Figure 1
is an adaptation of the main effects model proposed by Cohen and Wills (1985) and further
refined by Kawachi and Berkman (2000). The model shows a portion of the original main effects
model and focuses on the influence of social network factors on negative affective states
(depressive symptoms) and internalizing symptoms (depressive symptoms and depression).
These two constructs, along with social influences lead to health behaviors (either positive or
negative) or directly to mental health.
The three studies are also conceptualized from an ecological perspective and each will test
components of this overall model from different levels of analysis (see Figure 2). Study one
examines the individual level of influence and focuses on social integration of the individual
adolescent and depressive symptoms. Study two extends out to the level of the friendship- or

22
peer-group level. It examines peer contagion effects on adolescent depression. The final study
was focuses on network-level influences on depression outcomes.
The final conceptual model for the study (Figure 3) combines elements from all of the
theories presented, with the strongest influences from ecological theoretical perspectives and
the main effects model. It differentiates three levels of social analysis: the individual level, the
friendship- or peer-group level, and the network level and proposes that social network factors
at each level of analysis influence adolescent depressive symptoms and depression. The
individual level factors should have the strongest effects and have been tested the most in
previous research, but the group and network levels will also contribute significantly to
psychological well-being. The peer-group level of analysis addresses peer contagion effects,
while the network level focuses more the influence of network structure and composition on
the outcomes.

23
Conclusion and Chapter Summary
Social networks can influence health on many levels, through different pathways. The
present work examines the influence of social network factors on depression among
adolescents. It will test these associations at three different levels of social analysis: the
individual, the peer group, and the network level. Social network influences are expected to be
associated with depression at all three levels through a modified main effects model of
depression that is heavily influenced by ecological theoretical perspectives.
Chapter two describes the first study, which examined social network factors, peer status in
particular, and depressive symptoms at the individual level of analysis. Chapter three, presents
the second study, which also examined the effects of interpersonal factors and depressive
symptoms, though it more closely investigated the effects interpersonal factors have on
intrapersonal outcomes at the level of the friendship and peer group. The influence of friends’
social position and depressive symptoms on an individual’s own depressive symptoms was the
main focus of study two. Finally, chapter four describes study three, which was analyzed at the
classroom level of analysis and examined the influence of network structure on depression.

24
CHAPTER 2:
Social integration and depression among adolescents: Associations between social network
status and depressive symptoms in a Chinese sample

CHAPTER TWO ABSTRACT
The developmentally important tasks related to peer functioning and competence become
criteria for judging successful adjustment and aptitude by an individual, parents, teachers, and
the larger society for youth. Despite the importance of peer experiences during adolescence,
there has been relatively little research examining the importance of peer context and social
environment in understanding the increase in the prevalence of depression-related outcomes
that accompanies the transition into adolescence. Methods: Multilevel regression models were
used to analyze a sample of 5,563 10
th
grade students in China. Social network nominations
were used to calculate peer status categories and the CES-D was used to measure depressive
symptoms. Results: Having peers nominate one as a friend is more important than peers
considering one as most liked in relation to depressive symptoms. In-degree is associated with
depression (AOR=0.88, p<0.01), with the association being protective and driven by girls
attending professional schools (AOR=0.83, p<0.05). The risk of depression for those who are
marginal members of professional school networks is substantial, with marginalized girls
showing the greatest risk (AOR=4.71, p<0.01) and boys reporting fairly high risk as well
(AOR=2.51, p<0.01). For professional school girls, popular status network members are also at
greater risk for depression (AOR=1.68, p<0.05). Conclusions: Lower social network status is
associated with increased risk, so increasing social integration within a classroom, particularly
for those on the fringes of the network may help to decrease depression outcomes. These
findings suggest that a social network perspective could help to increase the effectiveness of
programs aimed at preventing adolescent depression.

25
INTRODUCTION
Adolescence is a developmental period marked by significant growth and change,
particularly as it relates to the social environment. It is a time of self-concept and identity
development, when peers and peer groups often become a way to explore these new roles and
identities. Combine this with research demonstrating that individuals who are less socially
integrated have more internalizing problems, and social environmental and competence factors
become increasingly important (Ueno, 2005). Social integration is sometimes measured with
social network instruments which allow for examination of the structural elements of peer
groups and also permits a more comprehensive view of the patterns and influence of social
relationships.
The link between social isolation and decreased psychological well-being has been well
established (for reviews see, Berkman, 1995; House, Landis, & Umberton, 1988), historically
dating back to Durkheim’s work in sociological theory (see Gibbs, & Martin, 1958; Inkeles, 1959).
Small social networks, low social support, and few or poor quality close relationships have all
been linked to depressive symptoms (see Seeman, 1996). For children and adolescents in
particular, depression and depressive symptoms have been associated with lower peer
acceptance (Rose & Rudolph, 2006) and being less liked by classmates (Caldwell, Rudolph,
Troop-Gordon, & Kim, 2004).
Many developmental theories include peer relations as important concepts in both
positive and negative adaptation during adolescence, which is not surprising given the
prominent place they occupy in an adolescent’s life. The developmentally important tasks
related to peer functioning and competence become criteria for judging successful adjustment
and aptitude by an individual, parents, teachers, and the larger society in general. Despite the
importance of peer experiences during adolescence, there has been relatively little research

26
examining how peer context and social environment might be important in understanding the
increase in the prevalence of depression-related outcomes that accompanies the transition into
adolescence.

Social Network Status and Depression among Adolescents
Theorists have long proposed that the environment, the social environment in particular,
plays a critical role in the development of psychological distress and illness (for example, see
Bowlby, 1951; 1980; 1988; Bronfenbrenner, 1979; 1999), including depression and depressive
symptoms. Bowlby (1951) even stated a universal human need to form close affective bonds.
Over the past several decades of research, these ecological models and methodologies have
been increasingly emphasized in efforts to understand health behaviors.
Luke and Harris (2007) described social network analysis as “uniquely suited to describing,
exploring, and understanding these types of structural and relational aspects of health.” As
mentioned previously, social integration has received the bulk of the attention, but more recent
efforts have begun examining other structural elements of social networks and their influence
on health and well-being (Falci & McNeely, 2009; Ostberg, 2003; Ueno, 2005). For example,
measures of centrality in networks look at the extent to which one node connects to other
nodes, and the tendency of that node to be positioned near the center of its network. Centrality
is often considered an indicator of importance or influence, with the simplest centrality measure
being a count of the number of network nominations received from others (the term for this is
degree centrality). Those with more connections or ties tend to be more central to the network.
Network centrality has been associated with various health behavior outcomes, such as sexual
activity (Ellen et al., 2001), aggression (Xie, Cairns, & Cairns, 2002), and smoking (Valente, Unger,
& Johnson, 2005). Other research has suggested associations with more positive outcomes, such

27
as being more athletic, popular, and better students (Farmer & Rodkin, 1996). Fowler and
Christakis (2008) found centrality to be associated with greater happiness, with centrality
leading to happiness rather than the opposite.
Peer status normally has a positively skewed distribution, with a few individuals that are
quite popular but the majority liked by some and many not chosen by anyone (Ostberg, 2003).
Social status in a classroom is assigned by others and is not a measure of how an individual
perceives their social influence or competence. Peer status, much like similar measures of social
isolation and integration, has been linked to psychological health and well-being (Berkman,
Glass, Brissette, & Seeman, 2000; Ostberg, 2003). As adolescents seek out connections to others
in the course of developmental processes, peer status becomes an important indicator of
growth and adjustment. If efforts to connect are unsuccessful, result in failure, or dysfunctional
relationships and social networks, youth can feel socially isolated and excluded which can
develop further into depressive symptoms and other maladaptive health outcomes.

Study Hypotheses
This study will use social network measures to investigate the influence of network status
on adolescent depressive symptoms. Structural aspects of peer networks during adolescence—a
period when friendships play an increasingly powerful role in development—merits examination
as a potential cause of depressive symptomatology.
The primary aim of this study was to examine whether social network status in the
classroom is associated with adolescent’s depression outcomes. To assess this aim, the study
tested whether adolescents with more limited social networks had higher self-reported
depressive symptoms. Six hypotheses were developed and tested for this study:

28
H1: In-degree is negatively associated with adolescent depressive symptoms.
H2: High social network status based on friend nominations is negatively associated with
depressive symptoms.
H3: Low social network status based on friend nominations is positively associated with
depressive symptoms.
H4: High social network status based on most liked nominations is negatively associated
with depressive symptoms.
H5: Low social network status based on most liked nominations is positively associated
with depressive symptoms.
These models were tested using the continuous measure of depressive symptoms, as well as
using cut-points identified in past work with the Center for Epidemiologic Studies Depression
Scale (CES-D; Radloff, 1977) as a measure of depression. The hypothesis tested for these
analyses was:
H6: The associations between social network measures and depression are stronger than
those for depressive symptoms in the same proposed directions.

METHODS
Sampling. Unlike in the United States, China has two primary high school types, academic
(regular) and professional (vocational). Students are tested in middle school and scores from
those tests determine which school type they will attend. Generally, those attending academic
schools have a rigorous schedule of courses for three years of high school (10
th
through 12
th

grades), designed to increase the chances students will pass college entrance exams. Those
attending professionals schools are enrolled in a career track, such as computers/IT or public
relations, and will complete an internship during their final year(s) in high school with the goal of

29
entering the workforce once they have completed high school. Both school types were included
in the study, with an equal number of schools and students from each school type. There are
more professional school classes in the study, however, due to a trend for these classes to be
smaller.
The study included a total sample of 5,563 10
th
grade students from a convenience sample
of 24 schools, 12 of each school type, from 124 classes.
Data Collection. The data for this study was collected as pre-intervention, baseline
measurements of a larger parent study examining the efficacy of a school-based tobacco
prevention program focusing on social influences. All assessments were paper-based surveys
and all participants had active parental consent and student assent. Pre-intervention
assessments were collected prior to any intervention delivery, with the pre-test occurring
approximately one week prior to intervention implementation for those assigned to the
Program condition. The baseline survey was collected approximately one month prior to the
pre-test. There were two reasons for the two waves of pre-intervention data collection: 1) the
major holiday, Chinese New Year, occurred during that time and would have disrupted data
collection efforts and 2) the assessments were each fairly substantial and splitting the measures
between two surveys prevented students from becoming fatigued and improved the quality of
the data.

Measures
Depressive Symptoms (CES-D; Radloff, 1977). The CES-D is a self-administered scale containing
20 items. The answers to each item, which indicate the frequency of the symptoms, are
structured on a 4-point likert scale ranging from 1 (Less than 1 day) to 4 (5-7 days). The answers

30
are summed across all 20 items. The total scores can range from 0 to 60. Radloff (1977)
suggested and it is commonly accepted that persons who score 16 or above are likely to be
clinically depressed; however, some studies suggest that the cut-point needs to be raised to 24
or higher if the tool is to achieve appropriate diagnostic characteristics (Gotlib, Lewinsohn, &
Seeley, 1995; Roberts, Lewinsohn, & Seeley, 1991). A Mandarin-Chinese version of the CES-D
(MC-CES-D) has been validated and used in Hong Kong (Cheung & Bagley, 1998; Lam et al.,
2004) and among community samples of adolescents in Taiwan a cut-point of scores ≥ 28 has
been determined, with those above this score more likely to have major depressive disorder
(Yang, Soong, Kuo, Chang, & Chen, 2004; Lin et al., 2008). The Cronbach’s alpha for the scale for
the entire sample was good at α = 0.89.
Social Network Measures. Sociometric procedures were used to obtain measures of
adolescents’ social network status. Peer nominations were used to collect information from
each student regarding peer status. Adolescents were asked to list up to five people in their
class who they considered their best friends. Students wrote the first and last names of their
friends. Similarly, students nominated the five people they considered to be the most liked by
people in their class. Peer nominations were later matched to study id’s and names were
dropped from the data to protect confidentiality. These nominations were then used to
calculate network metrics for the study.
Measures of centrality in social networks capture the extent to which a node connects
other nodes and its tendency to be positioned near the center of a network. Centrality is often
used as a marker of importance. The simplest centrality measure is a count of the number of
nominations (this is called “degree” centrality) a person gives (“out-degree”) or receives (“in-
degree). People with more friends will tend to be more central. Degree centrality, more
specifically in-degree, was calculated and used in the models for this study.

31
In order to define peer or social network status in a school classroom, researchers often
divide individuals into different status categories based on their social network nominations.

The status categories are created to represent a more normal distribution. Five social
network status categories were created from the friend nomination data for this study, similar
to those created in past research (Ostberg, 2003). The first category included those who did not
receive any nominations. This group, the “marginalized” category, is usually considered at
highest risk for many negative health behavior outcomes. The second group consists of the
“peripheral” members of the network, who received one or two friend nominations. The
“accepted” category is the largest and includes those receiving three to five nominations. The
“popular” category is those with six or seven nominations and the “favorite” those who received
eight or more (see Figure 4). The same five status categories were created for the most liked
nominations; however, because the distribution for this type of nomination differs from the
friend data, the construction of the categories was different. Most liked nominations, often
called “sociometric popularity,” usually have a larger range than friend nomination data with
more network members receiving zero or only one nomination. A few network members usually

32

receive the bulk of the most liked nominations in a classroom (see Figure 5). Therefore, the
marginal category remains the same, but the others are slightly different (peripheral = 1
nomination; accepted = 2-4; popular = 5-8; and favorite = 9 or more nominations).

Demographic covariates and control variables. Variables controlled for in the models
include: gender (female), school type (academic or professional), socioeconomic status, and age.
Socioeconomic status was calculated using both mother’s and father’s education level and a
measure of an adolescent’s disposable income.
Perceived social self-efficacy was also controlled for in the models and was measured with
the Perceived Social Self-Efficacy Scale (Smith & Betz, 2000). This is a twenty-five item scale that
asks about confidence in your ability to perform social tasks, with items structured on a 5-point
likert scale ranging from 1 (No confidence at all) to 5 (Complete confidence). Answers are
summed with total scores ranging from 1 to 125.

33
Data Analysis
Both multilevel linear and logistic regression analysis procedures were used to analyze the
data. Multilevel analysis was needed to explain individual-level outcomes with school and class
level characteristics. All analyses were done using STATA version 11 (StataCorp, 2009), with
social network status variables calculated using UCINet (Borgatti, Everett, & Freeman, 1999).
There is a large body of research indicating cognitions regarding social relationships and
social self-efficacy are strongly associated with depression-related outcomes (for review, see
Jacobs, Reinecke, Gollan, & Kane, 2008); therefore, perceived social self-efficacy was controlled
for in all models in order to get a true measure of the associations between social network
status and depressive symptoms.
To examine the first hypothesis, that in-degree centrality is negatively associated with
depressive symptoms, multi-level linear regression analyses were conducted, simultaneously
entering in-degree and the demographic and control variables in the level-one (fixed effects)
model and school type in the level-three (school-level random effects) model for depressive
symptoms. There were no level two (class level) variables, though level of analysis was included
in the random effects model.
Two other models were tested in the same fashion to examine hypotheses two through
five, that social network status is associated with depressive symptoms. Dummy variables for
each status category were included in these models, Model 2 for status based on friend
nominations and Model 3 for status based on most liked nominations. The accepted category
served as the reference group, as it represented the largest groups and was not addressed in the
hypotheses. High status as mentioned in hypotheses two and four was considered the two
categories higher than “accepted,” the “popular” and “favorite” categorizations. Low status as

34
mentioned in hypotheses three and five was considered the two categories below “accepted,”
the “peripheral” and “marginalized” categorizations.
The final hypothesis was analyzed by multilevel logistic regression and examined the same
independent variables as the first three models. These final three models used an outcome that
was a dichotomized version of the CES-D score, with scores of ≥28 equal to one and all other
scores equal to zero (see Measures for further details of cut-scores).

RESULTS
Sample Characteristics
Distribution of Demographic Characteristics and Main Study Variables.
The participation rate for the study was high, with a 93.3% participation rate. Table 1
presents descriptive data for the sample. The average age of participants was 15.8 and the
mean SES, ranging from 1-10 on a composite measure including mother and father education
and allowance per week, was 4.4. The mean depressive symptoms score (CES-D; range: 1-60) for
the sample is 17.9. Using a cut-point score of 28 or greater on the CES-D resulted in 16.8% of the
sample reporting high depressive symptoms. The mean score on the perceived social self-
efficacy scale was 80.4 with a range of 1-125. The average number of friends a student
nominated was 3.7, and 3.5 was the average for most liked nominations. The most friend
nominations received by a single student was 17 and the most liked nominations received by a
single student was 40. About half of the sample (49.3%) had a friend network status of
“accepted,” whereas 33% had an “accepted” most liked network status.
Differences in Study Variables by School and Gender.
A series of t-tests were conducted to determine equivalence on the main study variables
for adolescents in the different school types and for males and females. Age and perceived

35
social self-efficacy (PSSE) did not differ between school types or males and females. The mean
age was about 16 for all groups and the mean PSSE score was about 80. Academic and
professional school students were significantly different in depressive symptoms (t=5.29,
p<0.01) with those attending academic schools having significantly higher CES-D scores (a mean
score of 18.7 versus 17.1 for those in professional schools).
Table 1. Sample Characteristics
Total Academic Professional Girls Boys Range
Schools 24 12 12
Classes 124 56 68
Mean class size (SD) 51.8 (12.2) 55.6 (13.8) 48.3 (9.3)
Gender
Male (%) 50.3% 49.4% 50.9%
Female (%) 49.7% 50.6% 49.1%
Mean Age (SD) 15.8 (0.72) 15.7 (0.64) 16.0 (0.77) 15.7 (0.70) 15.9 (0.73) 13-19
Mean SES
a
(SD) 4.4 (1.5) 5.0 (1.5) 3.8 (1.2) 4.4 (1.5) 4.4 (1.5) 1-10
Mean CES-D Score (SD) 17.9 (11.0) 18.7 (11.4) 17.1 (10.5) 18.6 (11.4) 17.1 (10.5) 0-60
CES-D Score ≥ 28
b
(%) 16.8% 21.7% 16.2% 21.5% 16.2%
Mean PSSE (SD) 80.4 (19.9) 80.8 (20.4) 80.0 (19.5) 80.3 (18.6) 80.6 (21.3) 1-125
Social network measures
Mean friend nominations (SD) 3.7 (2.4) 4.0 (2.3) 3.8 (2.3) 4.1 (2.2) 3.7 (2.3) 0-17
Mean friends nominated (SD) 3.7 (2.0) 3.9 (1.9) 3.6 (2.1) 4.0 (1.8) 3.5 (2.1) 0-5
Mean most liked nominations (SD) 3.2 (4.3) 3.5 (4.5) 3.5 (4.3) 3.2 (4.2) 3.8 (4.6) 0-40
Mean most liked nominated (SD) 3.5 (1.9) 3.6 (1.8) 3.5 (1.9) 3.8 (1.8) 3.3 (1.9) 0-5
Friend social network status (%)
Marginalized (0 friend noms.) 3.0% 2.4% 3.2% 2.0% 3.7%
Peripheral (1-2 noms.) 11.7% 25.7% 28.3% 23.4% 30.8%
Accepted (3-5 noms.) 49.3% 47.1% 46.8% 49.3% 44.3%
Popular (6-7 noms.) 28.8% 17.2% 15.0% 17.5% 14.7%
Favorite (8+ noms.) 7.2% 7.6% 6.7% 7.8% 6.5%
Most liked network status (%)
Marginalized (0 most liked noms) 21.0% 22.3% 19.7% 24.6% 17.3%
Peripheral (1 nom.) 20.1% 20.1% 20.0% 20.3% 19.8%
Accepted (2-4 noms.) 33.0% 31.5% 34.5% 31.2% 34.9%
Popular (5-8 noms.) 15.7% 14.7% 16.7% 14.9% 16.7%
Favorite (9+ noms.) 10.2% 11.4% 9.1% 9.0% 11.3%
a
SES was measured by calculating the mean of the answers on three questions for each student. Two
questions asked about father’s and mother’s highest level of education (1-7) and the third question asked
about allowance per week (1-10).

Students in the different school types also differed significantly on the SES measure (t=33.7,
p<0.01), which is not that surprising considering those in academic schools are probably more
likely to have parents that attended academic schools themselves which also makes it more

36
likely they attended college than parents of professional school students. There were also
significant differences on the social network measures for the different school types, with
academic school students nominating more friends (t=5.41, p<0.01) and most liked (t=3.38,
p<0.01) as well as being nominated more as friends (3.51, p<0.01). The only network measure
with no significant differences by school type was most liked nominations received. Due to these
significant baseline differences by school type, all models were run separately for each school
type.
Differences also existed between males and females on certain study variables. As
expected, depressive symptoms were significantly different (t=5.13, p<0.01), with females
reporting higher mean CES-D scores (18.6 versus 17.1 for males). Females nominated more
friends (t=8.87, p<0.01) and more people as most liked (t=9.14, p<0.01). Girls also received more
friend nominations (t=6.77, p<0.01) than boys, but boys received more nominations for most
liked (t=5.33, p<0.01) than girls. Males and females did not differ on socioeconomic status.
Again, due to these significant differences in the main variables, the models were run separately
for girls and boys in order to more fully examine the study hypotheses.

Main Results
Multilevel Linear Regression Analysis of Social Network Status and Depressive Symptoms.
The first model tested hypothesis one, that in-degree is negatively associated with
depressive symptoms (see Model 1 in Table 2) indicating that those who receive more
nominations are less likely to report depressive symptoms. Perceptions of social self-efficacy
was the strongest predictor of adolescents’ depressive symptoms across all subgroups. In-
degree centrality was significantly associated with CES-D scores for the total sample (β= -0.04,
p<0.01) and for the professional school sample (β= -0.05, p<0.05). The association, however, is

37
driven primarily by the significant association found for professional school girls (β= -0.08,
p<0.01). Hypothesis one was not confirmed for academic school students. Socioeconomic status
was also significantly associated with depressive symptoms, though mainly for professional
school students (β= 0.17 for girls and β= 0.07 for boys). The relationship between SES and
depressive symptoms was positive, indicating that as SES increases so do depressive symptoms.
Model 2 examined hypotheses two and three, which state that social network status
based on friend nominations is associated with depressive symptoms, and resulted in mixed
findings (see Table 2). Hypothesis two, high social status is negatively associated with depressive
symptoms, was not confirmed overall. The one exception was a trend towards significance for
academic school boys (β= -0.05, p<0.10). Hypothesis three appeared more likely, with both
marginal and peripheral status being positively associated with CES-D scores (β=0.04, p<0.05 for
both). Again, when subgroups were analyzed separately the results were different. Marginal
status was significant for professional students (β= 0.07, p<0.01), though is stronger for girls (β=
0.11, p<0.01) with boys only showing a trend towards significance (β= 0.05, p<0.10). For
academic school students, the marginal categorization was not significantly associated with
depressive symptoms. Peripheral status was only significant for academic school girls (β= 0.07,
p<0.05).
Model 3 examined hypotheses two and five, which state that social network status based
on most liked nominations is associated with depressive symptoms, and was not well supported
(see Table 2). Only marginal status was significantly associated with depressive symptoms for
academic girls (β= 0.07, p<0.05). This indicates, that for this sample, not being nominated as a
friend is more important than having peers consider one as most liked in relation to depressive
symptoms.

Table 2: Standardized coefficients (β) for Three Models of Social Network Status and Depressive Symptoms by School Type & Gender
Depressive Symptoms
Total Sample Academic Professional
All Girls Boys All Girls Boys
Model 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
Fixed Effects Estimates (Odds Ratios)
In-Degree (Friend)
-0.04
**
-0.03 -0.03 -0.03 -0.05
*
-0.08
**
-0.01
Perceived Social Self
Efficacy
-0.19
**
-0.23
**
-0.25
**
-0.20
**
-0.16
**
-0.23
**
-0.10
**

Age
0.02 0.02 0.03 0.02 0.03 0.05
+
0.02
SES
0.08
**
0.05
*
0.06
+
0.05
+
0.12
**
0.17
**
0.07
*

Female
0.07
**
0.17
**
0.12
**

Network Status
(Friend)

Marginal
0.04
**
0.01 0.01 <0.01 0.07
**
0.11
**
0.05
+

Peripheral
0.04
*
0.04
+
0.07
*
0.02 0.04
+
0.05 0.02
Accepted ref. ref. ref. ref. ref. ref. ref.
Popular
-0.01 -0.01 0.02 -0.05 -0.01 -0.02 0.01
Favorite
0.01 -0.01 0.02 -0.05
+
0.02 0.02 0.03
Network Status (Liked)

Marginal
0.02 0.04 0.07
*
-0.01 0.01 <0.01 0.02
Peripheral
0.03 0.03 0.05 0.02 0.02 0.01 0.03
Accepted ref. ref. ref. ref. ref. ref. ref.
Popular
0.02 0.01 0.03 -0.02 0.02 0.05 -0.01
Favorite
0.01 <0.01 0.05 -0.04 0.01 <0.01 0.03
Random Effects Estimates
Level 3 (School)
<0.01 <0.01 <0.01
School Type
0.06
**
0.06
**
0.05
**

Level 2 (Class)
0.07
**
0.08
**
0.10
**
0.09
**
0.10
**
0.13
**
0.11
*
0.12
*
0.17
**
0.14
**
0.16
**
0.19
**
0.10
**
0.08
*
0.09
*
0.10
*
0.09
*
0.12
*
0.20
**
0.17
**
0.12
*

Observations 4626 4442 3935 2267 2162 1929 1202 1164 1048 1065 998 881 2359 2280 2006 1262 1229 1095 1097 1051 911
+
p<0.10,
*
p<0.05,
**
p<0.01; 95% confidence intervals in brackets
38

Table 3: Adjusted Odds Ratios for Three Models of Social Network Status and Depression (CES-D ≥28) by School Type & Gender
Depression (CES-D ≥28)
Total Sample Academic Professional
All Girls Boys All Girls Boys
Model 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
Fixed Effects Estimates (Odds Ratios)
In-Degree (Friend) 0.88
**
0.91
+
0.91 0.89 0.84
*
0.83
*
0.85
PSSE 0.69
**
0.58
**
0.53
**
0.64
**
0.84
*
0.66
**
1.07
Age 1.06 1.04 1.07 0.93 1.07 1.10 1.04
SES 1.17
**
1.06 1.08 1.05 1.36
**
1.56
**
1.14
Female 1.48
**
1.52
**
1.52
**

Network Status
(Friend)
Marginal 1.92
**
0.95 0.69 1.22 3.30
**
4.71
**
2.51
*

Peripheral 1.29
**
1.26
+
1.25 1.28 1.38
*
1.39
+
1.35
Accepted ref. ref. ref. ref. ref. ref. ref.
Popular 0.85 0.76
+
0.86 0.61
+
0.96 0.98 0.97
Favorite 0.92 0.72 0.79 0.63 1.2 1.36 0.97
Network Status (Liked)
Marginal 1.14 1.24 1.33 1.17 1.04 0.81 1.44
Peripheral 1.16 1.15 1.12 1.19 1.19 1.26 1.1
Accepted ref. ref. ref. ref. ref. ref. ref.
Popular 1.17 1.05 1.18 0.88 1.3 1.68
*
0.82
Favorite 1.06 0.98 1.3 0.7 1.12 1.1 1.27
Random Effects Estimates
Level 3 (School) <0.01 <0.01 <0.01
School Type 0.14
**
0.13
**
0.12
**

Level 2 (Class) 0.11 0.11 0.16
*
0.15 0.10 0.15 0.20
+
0.19 0.22 0.40
**
0.32
*
0.36
*
0.23
*
0.16 0.20
+
0.05 0.02 <0.01 0.60
**
0.48
**
0.49
**

Observations 4629 4444 3937 2268 2162 1929 1202 1164 1048 1066 998 881 2361 2282 2008 1262 1229 1095 1099 1053 913
+
p<0.10,
*
p<0.05,
**
p<0.01; 95% confidence intervals in brackets
39

40
Multilevel Logistic Regression Analysis of Social Network Status and Depression.
Table 3 present the results for logistic regression analyses testing hypothesis six, that the
associations will be stronger between social network status and depression than depressive
symptoms. While direct comparisons between coefficients cannot be made, the magnitude of
the associations can be considered. The associations between in-degree and depression show a
similar pattern in this analysis as with the linear regression. Higher in-degree is protective
against depression (AOR=0.88, p<0.01), with the association being primarily driven by
professional school girls (AOR=0.83, p<0.05). The main differences between the logistic and
linear regression results appear for Model 2. While the pattern of significant associations remain
mostly the same, with professional school students showing significant associations between
lower social status and depression (AOR=3.30, p<0.01) and academic school students not
showing this, the additional information comes in the form of the magnitude of the associations.
The risk of depression for those who are marginal members of professional school social
networks is substantial, with marginal girls at greatest risk (AOR=4.71, p<0.01) and with boys
reporting fairly high risk as well (AOR=2.51, p<0.01). Peripheral status showed a trend towards
significance for professional girls (AOR=1.39, p<0.10), but not for boys.
Model 3 in Table 3 does not support an association between network status based on
most liked nominations, similar to the linear regression results, with one notable exception. For
professional school girls, those with popular status are at greater risk for depression (AOR=1.68,
p<0.05).

DISCUSSION
This study examined the association between social network status and depressive
symptoms among Chinese adolescents. Consistent with previous research on social integration

41
and depressive symptoms, the findings support a negative association between the more
general measure of social network status, in-degree centrality for friend nominations, and
depressive symptoms. These findings indicate that the higher the in-degree, the lower the
depressive symptoms. The specific hypothesis regarding low friend social network status and
depressive symptoms was also supported, which is consistent with previous research that has
shown that a lack of integration is associated with depression outcomes. People with fewer
friends report higher depressive symptoms (Brendgen, Vitaro, & Bukowski, 2000; Falci &
McNeely, 2009; Ueno, 2005). This finding was primarily driven by professional school students,
with significant associations for both boys and girls, yet a stronger association for girls. The
hypotheses regarding high social status being negatively associated with depressive symptoms
were not confirmed, indicating that while low social network status appears to be a risk factor
for depression outcomes, it does not necessarily follow that high status is protective. Contrary
to one study hypothesis, professional school girls with popular status based on “most liked”
nominations actually had an increased risk for depression. There is some research supporting
this finding, Falci & McNeely (2009) reported adolescents with social networks that were either
too small or too large reported higher levels of depressive symptoms. They put forth the idea
that for those with high social status, higher depressive symptoms stem from a sense of duty or
obligation. They stated that more friends require more time and energy to maintain and “could
result in role strain.” Findings from this study indicate that this is the case for girls but not boys
and professional school students and not academic school students. Also, this was not the case
for status based on friend nominations.
Differences in study findings between social status based on friend nominations and
status based on most liked nominations indicates that these are distinct constructs that measure
unique aspects of an adolescent’s social status. Status based on friend nominations was more

42
significantly associated with depressive symptoms, indicating that close friendships are perhaps
a more important indicator for depression-related outcomes for Chinese adolescents. However,
there was one instance where “most liked” status was important, for professional school girls,
as mentioned previously. This finding suggests that while lower friend status is a risk factor as
would be expected, it may be higher “most liked” status that is a risk depression.
There are clear differences between the two school types in the relationship between
social network status and depressive symptoms. The associations are stronger for professional
school students. The Chinese school system is set up differently than Western school systems,
with academic school students required to spend many more hours per week studying for
college entrance exams. They also have more pressure put on them to do well academically by
their parents and teachers and more competition from their classmates. They have little free
time to socialize, which may be one reason for the differences noted in the study. Professional
school students have more free time and fewer pressures – though for some, they may already
consider themselves “failures” for not getting into an academic school which could present a
risk factor for depressive symptoms. While academic school students, particularly the girls,
report the highest levels of depressive symptoms, they do not appear to be that closely
associated with what occurs in their social environment. Dealing with school and family stress as
well as gender role expectations could be a larger factor in their depression outcomes; however,
further study would be required to examine this.

Study Limitations
The study presented several limitations that should be considered when interpreting the
findings. First, this study did not include measures of the quality or strength of a reported
relationship. It may be that the quality of social network ties can weight the relationship

43
between these measures and depressive symptoms. For example, it may not matter if an
individual has few friends if the quality of those friendships is very high, where friends provide
maximum support, feelings of closeness, and understanding. It may also be the case that
reciprocated nominations are more important than the overall number of nominations to
depressive symptoms.
Second, while the associations reported were significant, the effects were small. This
could potentially limit the meaningfulness of the findings; however some have suggested that
small effect size can still have practical applicability (Rosenthal, Rosnow, & Rubin, 2000). Due to
the cross-sectional nature of the data, causal direction cannot be determined. The results can
suggest avenues for future research, but cannot explain causal mechanism leading from social
network status to depressive outcomes. Further study with longitudinal data is necessary.
Also, the CES-D is a self report assessment of depressive symptoms, rather than a clinical
assessment of syndrome or disorder, therefore the findings of this study do not generalize to
populations who have diagnosable disorders and it is not possible to determine whether the
suggested relationships between social network status and depression-related outcomes are
important mechanisms operating in more clinical populations. Also, though the design and
measurement of the study enables examination of data from more than one source, which is a
considerable strength of the study, the social network nomination format was limited to school-
based peers which may not be the most influential peers for all youth.
Despite these limitations, the results confirm, although modestly, that social network
status is significantly associated with depressive symptoms for Chinese adolescents.

44
Implications and Future Directions
It has become clear that depression has an impact on adolescent adjustment and
psychological functioning (Petersen et al., 1993). Depressed mood and depressive symptoms
have been found to explain over two-thirds of the variance in adolescent well-being (Mahon &
Yarcheski, 2001). These and other findings (e.g. Harrington, 1998) indicate that even relatively
mild depressive symptoms can result in impaired functioning. Therefore, those with sub-clinical
depressive symptoms during adolescence need to be considered at increased risk for negative
outcomes.
The developmental importance of peer relations places a premium on understanding their
influence on the etiology and psychopathology of health behavior. The study of physical and
mental well-being must include consideration of the importance of context, and during
adolescence the context is inherently social. Peer relationships among adolescents are
commonly organized hierarchically. The daily social world of youth often carries a specific social
order with rewards and penalties depending on the skill with which an adolescent navigates this
environment. Differences in health by social status have been found among adolescents,
showing a similarity to the adult research on this topic. The peer status of youth has been linked
to mental health and negative health behaviors into young adulthood (Michell & Amos, 1997;
Ostberg, 2002; Roff & Wirth, 1984). Both friendship ties and the overall structure of the social
network, in addition to individual social status, have been found to influence adolescent health
(Ueno, 2005; Kiesner, 2002). This suggests that the organization, structure, and distribution of
social status among peer groups can influence adolescent development.
Prevention work with youth often takes a “social influences” approach. The findings from
this study suggest that a social network perspective could help to increase the effectiveness of
programs aimed at preventing adolescent depression-related outcomes. Lower social network

45
status is associated with increased risk, so increasing social integration within a classroom,
particularly for those on the fringes of the network may help to decrease depression. Creating a
social environment that values acceptance and inclusion could go a long way in decreasing the
depressive symptoms of an adolescent at risk. The findings can also help to identify those at the
highest risk, who could benefit from more intensive or targeted intervention.
Significant biological, cognitive, and social changes occur during adolescence that can have
an impact on developmental competencies related to health behavior decision making. The high
potential for both risk and opportunity makes this an important point at which to intervene to
promote healthy lifestyle choices as well as prevent dysfunction and risky behavior decisions
that can persist throughout adulthood. Many factors have been found to increase vulnerability
to depression for adolescents and many of the problem behaviors that prevention research has
targeted have been linked to depression as well. While depression is generally conceptualized as
an individual disorder with a strong focus on problematic thoughts and perceptions,
maladaptive interpersonal functioning also contributes to its development and progression.
Preventing distress and dysfunction in the social networks of youth at risk for depression, as well
as improving interpersonal competencies, should be a key aim of future prevention work
focusing on this developmental period.
The results of this study contribute to a better understanding of how peer relations are
related to adolescent depressive outcomes. This work helps to further understand the
complicated social environments of adolescents. It further clarifies the association between
social network influences and depression during adolescence. It adds to current research
examining social integration and depression among adolescents by including two measures of
peer status, friend nominations and most liked nominations. These are two related, yet distinct
measures of social status that uniquely contribute to depression and depressive symptoms in

46
their own way. This study also examined these associations in a sample of Chinese adolescents,
when the majority of this research has been done in Western cultures. This examination of a
collectivistic culture that is undergoing rapid social and cultural changes provides some insight
into the social environments of Chinese youth and how their peer interactions influence their
psychological well-being.

47
CHAPTER 3:
Peer Contagion and Adolescent Depression: The Effects of Peers’ Depressive Symptoms and
Social Network Status

CHAPTER THREE ABSTRACT
Peer contagion effects have been studied across various disciplines and represent an
intriguing phenomenon in the study of social influence. Contagion theories seek to understand
social networks as pathways for the spread of “infectious” attitudes, norms, and behaviors.
Adolescent peer contagion research has primarily focused on externalizing behaviors, while few
studies have looked at internalizing symptoms. This study was designed to examine whether
depressive symptoms and social network status of friends are significantly associated with an
adolescent’s own self-reported depression in a sample of Chinese adolescents. Methods:
Measures of depressive symptoms and social network nominations were examined using
multilevel linear and logistic regression models for an overall sample of 5,563 10
th
grade
students in China. Results: Mean friends’ CES-D scores were significantly associated with
depressive symptoms (β=0.10, p<0.01), even after controlling for social cognitions. The
proportion of depressed friends was strongly associated with depression (AOR=2.85, p<0.01 for
the total sample) with a stronger association for those in professional than academic schools
(AOR=3.34 versus AOR=2.44 for academic students, p<0.01 for both). Peers’ social network
status was not significantly associated with depressive symptoms with notable exceptions for
academic school girls (β=0.07, p<0.05) and professional school boys (β=-0.06, p<0.05).
Conclusions: Adolescents are embedded within social networks that have the potential to
influence health behavior in positive and negative ways. From the network perspective, the
psychological well-being of one individual has implications for the health and well-being of their
peers.

48

INTRODUCTION
Research has effectively established that psychological well-being is connected to people’s
social relationships. Wide ranging work has shown both beneficial consequences of positive
social relations (for reviews, see Berkman, Glass, Brissette, & Seeman, 2000; Cohen, 2004), and
the detrimental effects negative social interactions can have on psychological and physical
health (Okun & Keith, 1998; for review, see Rook, 1992). Research on social network
participation, under the term social integration and measured in terms of the variety of
relationships one participates and engages in, has found that those who are socially integrated
are in better mental health than those who are more socially isolated (for review, see Cohen &
Wills, 1985). This robust finding is based on the premise that those who participate in social
networks are influenced to normative health behaviors due to social controls and peer
pressures (Cassel, 1976; Hammer, 1981). Social integration is also hypothesized to engender
general positive affect due to an increased sense of belonging, stability, and self-worth (Cassel,
1976; Thoits, 1983; Wills, 1985). For example, Fowler and Christakis (2008) found that a person
is more likely to be happy if directly connected to someone who is happy. They also noted that
those at the core of their network are more likely to be happy, while those on the periphery are
more likely to be unhappy. The magnitude of the health risks associated with social isolation and
low levels of social integration have been compared to the risks associated with smoking, high
blood pressure, and obesity (Cohen, Gottlieb, & Underwood, 2000; House et al., 1988).
While the bulk of the research regarding social ties and mental health has been conducted
with adult populations, the link between social relationships and depressive symptoms has also
been clearly established during adolescence. This finding is not surprising given peer
relationships are critical to the social development of an adolescent (e.g. Bronfenbrenner, 1979;

49
Furman & Burhmester, 1992; Hartup, 1995). They play a prominent role in identity
development, through peer experiences such as social comparison, reflected self-appraisal, and
peer group affiliation (Harter, Stocker, & Robinson, 1996). Membership, acceptance, and
healthy functioning in larger peer groups are also critical to adolescents’ psychological well-
being (Brown & Lohr, 1987). Adolescents spend more of their free time interacting with peers
than in childhood and early adolescence (Csikszentmihalyi & Larson, 1984; Larson & Richards,
1991), which results in increased feelings of closeness and intimacy with peers during
adolescence (Furman & Buhrmester, 1992). Social interactions become more emotionally
rewarding (Larson & Richards, 1991) for youth. Peers are the primary components of an
adolescent’s social network and are relied upon more heavily as sources of support and advice
during this developmental period (Brown, 1990; Buhrmester, 1996). Therefore, problems in the
domain of peer relationships and social environment are likely to be particularly distressing
during adolescence (Hammen, 2009; La Greca & Harrison, 2005; La Greca & Lopez, 1998).
Several studies have identified structural features of social networks in childhood and
adolescence and examined the effects of social network membership on various outcomes (e.g.
Cairns, Cairns, Neckerman, Gest, & Gariépy, 1988; Kindermann, 1996; Urberg, Degirmencioglu,
& Pilgrim, 1997; Kiesner, 2002; Ueno, 2005). Social networks have been found to be an
important environment in which socialization occurs, for both positive and negative
developmental outcomes. For example, social networks have been associated with antisocial
behavior and delinquency (Dishion, McCord, & Poulin, 1999), but also with positive academic
outcomes (Espelage, Holt, & Henkel, 2003). However, only a few studies have examined how
social networks are associated with depressive symptoms among adolescents (Hansell 1985;
Ueno 2005).

50

Peer Contagion and Depressive Outcomes
Peer contagion effects have been studied across various disciplines with a multitude of
outcomes. Contagion theories seek to understand social networks as pathways for the spread of
“infectious” attitudes, norms, beliefs, and behaviors. Contagion theories are related to a number
of other theories, such as Symbolic Interaction, Gatekeeping Theory, and Structurational Theory,
which all focus on the influence of social networks and social structures on behavior. Social
contagion suggests that individuals adopt the attitudes or behaviors of others in their social
network with whom they communicate with or are exposed to. The theory, however, does not
require intent or even awareness to influence, only exposure.
Peer contagion effects, have been demonstrated among adolescents for a host of negative
behaviors, such as substance use (Kobus, 2003; Li, Barrera, Hops, & Fisher, 2002; Oetting &
Beauvais, 1986; Valente, Gallaher, & Mouttapa, 2004), sexual risk behavior (Prinstein, Meade, &
Cohen, 2003), and suicidal ideation and behavior (Cerel, Roberts, & Nilsen, 2005; Mercy et al,
2001). The research focusing on adolescent peer contagion has primarily examined the diffusion
of externalizing and health risk behaviors and has typically found strong relationships between
an adolescent’s own externalizing behavior and that of their friends’ (Dishion, McCord, & Poulin,
1999; Kandel, 1996). Only a few studies have looked at internalizing symptoms and peer
contagion, however. Stevens and Prinstein (2005) for example, found that best friends’ reported
level of depressive symptoms was prospectively associated with adolescents’ own depressive
symptoms and with adolescents’ negative social cognitions. These results suggest that
adolescents’ depressive symptoms and cognitions are directly influenced by the presence of
similar thoughts and feelings among their close friends. There are more studies among adults
that have demonstrated that depressed individuals are likely to select other depressed

51
individuals as friends and respond more favorably to social interactions that involve a depressed
partner (Rosenblatt & Greenberg, 1991).

Gender, Social Relationships, and Depression
Consideration of gender is critical in the study of internalizing symptoms, particularly
during adolescence. Female adults and adolescents are twice as likely to experience internalizing
symptoms as males (Nolen-Hoeksema & Girgus, 1994). In terms of the developmental period,
adolescents report more internalizing problems than children (Zahn-Waxler, Klimes-Dougan, &
Slattery, 2000). Research examining interpersonal relationships has found that females are more
likely to seek emotional support from their social network (Nolen-Hoeksema, 1991) and to value
social ties more (Swap & Rubin, 1983) than males. Males do turn to their social networks for
support, but the interactions tend to focus more on shared activities or “distractions” (Bolton &
Oakley, 1987; Nolen-Hoeksema, 1990).
As for peer and social network influences on internalizing problems, such as depression
and depressive symptoms, gender effects are not as clearly delineated as the gender differences
in depression outcomes. Evidence has shown that boys are more susceptible to peer influences
for certain externalizing problems, antisocial and delinquent behaviors in particular (Berndt,
1979; Brown, Clasen, & Eicher, 1986; Liu & Kaplan, 1999), but findings of gender differences
have been mixed for other externalizing and internalizing problems (Cohen, 1977; Kandel, 1978).
However, more recent work examining interpersonal relationships and depression has shown a
stronger relationship between social support and depression for females than males
(Schraedley, Gotlib, & Hayward, 1999). A study, by Hogue and Steinberg (1995) found
homophily of internalized distress for adolescent boys and girls equally, but noted some gender
differences for those at both the high and low ends of the distress spectrum. More specifically,

52
males who initially had low internalized distress were found to have an increase in negative
affect if they associated with more distressed peers and those who started relatively high in
distress tended to have decreased distress over time if members of their social networks were
less distressed.

Depression and Social Relationships in the Chinese Culture
Depression rates are increasing everywhere, with depressive disorders affecting all ethnic
groups and becoming one of the most prevalent diseases globally (Daly, 2009). For many in
China, depression is expressed in physical rather than psychological terms. Often, those
experiencing depression in Chinese society report psychosomatic symptoms such as boredom
and discomfort, feelings of inner pressure and more physical symptoms of pain, dizziness, and
fatigue (Kleinman, 2004). Recent research has shown increased prevalence of depressive
symptoms and suicidal ideation in China. Family conflict, work or school stress, peer relations,
and financial pressure have been identified as important factors in depression in China (Pearson,
Phillips, He, & Ji, 2002; Yip, 2001).
Different cultures may share a similar etiology of depression among adolescents, but
patterns in the expression of depression can vary greatly by the socialization that occurs due to
cultural traditions, beliefs, norms, and values. An example of this is individualistic and
collectivistic cultural orientations, which could predispose adolescents from a particular culture
to a greater responsiveness or sensitivity to interpersonal relationships, which in turn may
influence the contribution these relationships have on depression-related outcomes. It has been
suggested that adolescents in more collectivist cultures could be more prone to distress caused
by problematic social interactions, due to greater cultural emphasis placed on social acceptance
and approval (Sastry & Ross, 1998; Tafarodi & Smith, 2001).

53
Study Hypotheses
A closer examination of the relationships between friends’ depressive symptoms, friends’
social status, and an adolescent’s own depressive symptoms could provide greater
understanding of the cognitive, interpersonal, and social environmental factors that place
adolescents at such increased risk for depressive outcomes. The overarching goal of this study
was to examine peer contagion effects on adolescent depressive symptoms. To that end, this
study used social network methods and a social contagion theoretical perspective to examine
how peer social behavior and internalizing problems relate to adolescent depressive symptoms.
The main purpose of the study was to examine whether depressive symptoms and social
network status of self-reported friends are significantly associated with an adolescent’s own
self-reported depression in a sample of Chinese adolescents. To investigate this, three primary
hypotheses were tested:
H1: Friends’ depressive symptoms are positively associated with depressive symptoms.
H2: The proportion of nominated friends’ who are depressed is positively associated
with depressive symptoms.
H3: Friends’ social network status based on friend nominations and friends’ status
based on most liked nominations are negatively associated with depressive
symptoms.
METHODS
Sampling. Unlike in the United States, China has two primary high school types, academic
(regular) and professional (vocational). Students are tested in middle school and scores from
those tests determine which school type they will attend. Generally, those attending academic
schools have a rigorous schedule of courses for three years of high school (10
th
through 12
th

54
grades), designed to increase the chances students will pass college entrance exams. Those
attending professionals schools are enrolled in a career track, such as computers/IT or public
relations, and will complete an internship during their final year(s) in high school with the goal of
entering the workforce once they have completed high school. Both school types were included
in the study, with an equal number of schools and students from each school type. There are
more professional school classes in the study, however, due to a trend for these classes to be
smaller.
The study included a total sample size of 5,563 10
th
grade students from a convenience
sample of 24 schools, 12 of each school type, from 124 classes.
Data Collection. The data for this study were collected as pre-intervention, baseline
measurements of a larger parent study examining the efficacy of a school-based tobacco
prevention program focusing on social influences. All assessments were paper-based surveys
and all participants had active parental consent and student assent. Pre-intervention
assessments were collected prior to any intervention delivery, with the pre-test occurring
approximately one week prior to intervention implementation for those assigned to the
program condition. The baseline survey was collected approximately one month prior to the
pre-test. There were two reasons for the two waves of pre-intervention data collection: 1) the
major holiday, Chinese New Year, occurred during that time and would have disrupted data
collection efforts and 2) the assessments were each fairly substantial and splitting the measures
between two surveys prevented students from becoming fatigued and improved the quality of
the data.

55

Measures
Depressive Symptoms (CES-D; Radloff, 1977). The Center for Epidemiologic Studies Depression
Scale (CES-D) is a self-administered scale containing 20 items. The answers to each item, which
indicate the frequency of the symptoms, are structured on a 4-point likert scale ranging from 1
(Less than 1 day) to 4 (5-7 days). The answers are summed across all 20 items. The total scores
can range from 0 to 60. Radloff (1977) suggested, and it is commonly accepted, that persons
who score 16 or above are likely to be clinically depressed; however, some studies suggest that
the cut-point needs to be raised to 24 or higher if the tool is to achieve appropriate diagnostic
characteristics (Gotlib, Lewinsohn, & Seeley, 1995; Roberts, Lewinsohn, & Seeley, 1991). A
Mandarin-Chinese version of the CES-D (MC-CES-D) has been validated and used in Hong Kong
(Cheung & Bagley, 1998; Lam et al., 2004) and among community samples of adolescents in
Taiwan, a cut-point of score ≥ 28 has been determined for Chinese youth, with those above this
score more likely to have major depressive disorder (Yang, Soong, Kuo, Chang, & Chen, 2004; Lin
et al., 2008). The Cronbach’s alpha for the scale for the study sample was good at α = 0.89.

Social Network Measures. Sociometric procedures were used to obtain measures of adolescents’
social networks. Peer nominations were used to collect information from each student
regarding peer status. Adolescents were asked to list up to five people in their class who they
considered their best friends. Students wrote the first and last names of their friends. Similarly,
students nominated the five people they considered to be the most liked by people in their
class. Peer nominations were later matched to study id’s and names were dropped from the
data to protect confidentiality. These nominations were then used to calculate network metrics
for the study.

56
In order to define peer or social status in a school classroom, researchers often divide
individuals into different status categories based on their social network nominations. The
status categories are created to represent a more normal distribution. Five social network status
categories were created from the friend nomination data for this study, similar to those created
in past research (Ostberg, 2003; Stutz, 1985). The first category included those who did not
receive any nominations. This group, the “marginalized” category, is usually considered at
highest risk for many negative health behavior outcomes. The second group consists of the
“peripheral” members of the network, who received one or two friend nominations. The
“accepted” category is the largest and includes those receiving three to five nominations. The
“popular” category is those with six or seven nominations and the “favorite” those who received
eight or more. The same five status categories were created for the most liked nominations;
however, because the distribution for this type of nomination differs from the friend data, the
construction of the categories was different. Most liked nominations, often called “sociometric
popularity,” usually have a larger range than friend nomination data with more network
members receiving zero or only one nomination. A few network members usually receive the
bulk of the most liked nominations in a classroom. Therefore, the marginal category remains the
same, but the others are slightly different (peripheral = 1 nomination; accepted = 2-4; popular =
5-8; and favorite = 9 or more nominations).
Friend Depressive Symptoms and Social Network Status. Because nominated friends were
themselves respondents in the study sample, the level of depressive symptoms reported by
each nominated peer was available. Therefore, it was possible to construct a measure of peer
group depressive symptoms by calculating the group mean of the individual depressive
symptoms reported by each nominated peer. These peer group variables do not include the
score of the target adolescent. A “proportion of depressed friends” variable was created using

57
the dichotomized CES-D variable, where those with a CES-D score of ≥28 are assigned a “1” and
all others assigned a “0.” A sum of this variable was calculated for all friends nominated, the
sum was then divided by the total number of friends nominated in order to get a proportion.
A mean of the social network status variables for friend and most liked nominations was
also calculated as a measure of friends’ social network status.
Demographic covariates and control variables. Variables controlled for in the models
include: gender (female), school type (academic or professional), socioeconomic status, and age.
Socioeconomic status was calculated using both mother’s and father’s education level and a
measure of an adolescent’s disposable income. Due to the differences in depression among
adolescents, friends’ gender was also controlled for by including a variable that was the
proportion of friends nominated who were female.
Perceived social self-efficacy was also controlled for in the models and was measured with
the Perceived Social Self-Efficacy Scale (PSSE; Smith & Betz, 2000). This is a twenty-five item
scale that asks about confidence in your ability to perform social tasks, with items structured on
a 5-point likert scale ranging from 1 (No confidence at all) to 5 (Complete confidence). Answers
are summed with total scores ranging from 1 to 125.

Data Analysis
Both multilevel linear and logistic regression analysis procedures were used to analyze the
data. Multilevel analysis was needed to explain individual-level outcomes with school and class
level characteristics. All analyses were done using STATA version 11 (StataCorp, 2009), with
social network status variables calculated using UCINet (Borgatti, Everett, & Freeman, 1999).
There is a large body of research indicating cognitions regarding social relationships and
self-efficacy are strongly associated with depression-related outcomes (for review, see Jacobs,

58
Reinecke, Gollan, & Kane, 2008); therefore, perceived social self-efficacy was controlled for in all
models. An adolescent’s own social network status was also controlled for in all models - as
measured by in-degree, a measure of network centrality which is essentially a standardized
measure of the number of friend nominations a student received from classmates.
To examine the first hypothesis, that friend depressive symptoms are positively associated
with depressive symptoms, multilevel linear regression analyses were conducted,
simultaneously entering friends’ depressive symptoms and the demographic and control
variables in the level-one (fixed effects) model and school type in the level-three (school-level
random effects) model for depressive symptoms. There were no level two (class level) variables
examined, though this level of analysis was included in the random effects model. The second
model tested included both mean friend depressive symptoms and mean most liked depressive
symptoms. Two additional models with these same independent variables were run using
logistic regression with a dichotomized version of the CES-D score as the outcome, with scores
of ≥28 being equal to one and all other scores equal to zero. Hypothesis two was tested in a
similar manner using the same demographic covariates and control variables, with the
proportion of depressed friends serving as the main independent variable in place of mean
friend depression scores. The third hypothesis again included the covariates, this time also
controlling for mean friends’ depressive symptoms. The main independent variables of interest
in these models were friend social network status variables. Finally, gender was examined in
each model in order to investigate the fourth study hypothesis and all models were run
separately for each gender.

59
RESULTS
Distribution of Demographic Characteristics and Main Study Variables. The participation rate for
the study was high, with a 93.3% participation rate. Table 1 presents descriptive data for the
sample. The average age of participants was 15.8 and the mean SES, ranging from 1-10 on a
composite measure including mother and father education and allowance per week, was 4.4.
The mean depressive symptoms score (CES-D; range: 1-60) for the sample is 17.9. Using a cut-
point score of 28 or greater on the CES-D resulted in 16.8% of the sample reporting high
depressive symptoms.

The mean score on the perceived social self-efficacy scale was 80.4 with a range of 1-125.
The average number of friends a student nominated was 3.7, and 3.5 was the average for most
Table 4. Sample Characteristics
Total Academic Professional Girls Boys Range
Classes 124 56 68
Mean class size (SD) 51.8 (12.2) 55.6 (13.8) 48.3 (9.3)
Gender
Male (%) 50.3% 49.4% 50.9%
Female (%) 49.7% 50.6% 49.1%
Mean Age (SD) 15.8 (0.72) 15.7 (0.64) 16.0 (0.77) 15.7 (0.70) 15.9 (0.73) 13-19
Mean SES (SD) 4.4 (1.5) 5.0 (1.5) 3.8 (1.2) 4.4 (1.5) 4.4 (1.5) 1-10
Mean CES-D Score (SD) 17.9 (11.0) 18.7 (11.4) 17.1 (10.5) 18.6 (11.4) 17.1 (10.5) 0-60
CES-D Score ≥ 28 (%) 16.8% 21.7% 16.2% 21.5% 16.2%
Mean PSSE (SD) 80.4 (19.9) 80.8 (20.4) 80.0 (19.5) 80.3 (18.6) 80.6 (21.3) 1-125
Social network measures
Mean friend nominations (SD) 3.7 (2.4) 4.0 (2.3) 3.8 (2.3) 4.1 (2.2) 3.7 (2.3) 0-17
Mean friends nominated (SD) 3.7 (2.0) 3.9 (1.9) 3.6 (2.1) 4.0 (1.8) 3.5 (2.1) 0-5
Mean most liked nominations (SD) 3.2 (4.3) 3.5 (4.5) 3.5 (4.3) 3.2 (4.2) 3.8 (4.6) 0-40
Mean most liked nominated (SD) 3.5 (1.9) 3.6 (1.8) 3.5 (1.9) 3.8 (1.8) 3.3 (1.9) 0-5
Friend social network status (%)
Marginalized (0 friend noms.) 3.0% 2.4% 3.2% 2.0% 3.7%
Peripheral (1-2 noms.) 11.7% 25.7% 28.3% 23.4% 30.8%
Accepted (3-5 noms.) 49.3% 47.1% 46.8% 49.3% 44.3%
Popular (6-7 noms.) 28.8% 17.2% 15.0% 17.5% 14.7%
Favorite (8+ noms.) 7.2% 7.6% 6.7% 7.8% 6.5%
Most liked network status (%)
Marginalized (0 most liked noms.) 21.0% 22.3% 19.7% 24.6% 17.3%
Peripheral (1 nom.) 20.1% 20.1% 20.0% 20.3% 19.8%
Accepted (2-4 noms.) 33.0% 31.5% 34.5% 31.2% 34.9%
Popular (5-8 noms.) 15.7% 14.7% 16.7% 14.9% 16.7%
Favorite (9+ noms.) 10.2% 11.4% 9.1% 9.0% 11.3%

60
liked nominations. The most friend nominations received by a single student was 17 and the
most liked nominations received by a single student was 40. About half of the sample (49.3%)
had a friend network status of “accepted,” whereas 33% had an “accepted” most liked network
status.
Differences in Study Variables by School and Gender. A series of t-tests were conducted to
determine equivalence on the main study variables for adolescents in the different school types
and for males and females. Age and perceived social self-efficacy (PSSE) did not differ between
school types or males and females. The mean age was about 16 for all groups and the mean
PSSE score was about 80. Academic and professional school students were significantly different
on depressive symptoms (t=5.29, p<0.01) with those attending academic schools having
significantly higher CES-D scores (a mean score of 18.7 versus 17.1 for those in professional
schools). Students in the different school types also differed significantly on the SES measure
(t=33.7, p<0.01), which is not that surprising considering those in academic schools are probably
more likely to have parents that attended academic schools themselves which also makes it
more likely they attended college than parents of professional school students. There were also
significant differences on the social network measures for the different school types, with
academic school students nominating more friends (t=5.41, p<0.01) and most liked (t=3.38,
p<0.01) as well as being nominated more as friends (3.51, p<0.01). Due to these significant
differences by school type, all models will be run separately for each school type.
Differences also existed between males and females on certain study variables. As
expected, depressive symptoms were significantly different (t=5.13, p<0.01), with females
reporting higher mean CES-D scores (18.6 versus 17.1 for males). Females nominated more
friends (t=8.87, p<0.01) and more people as most liked (t=9.14, p<0.01). Girls also received more
friend nominations (t=6.77, p<0.01) than boys, but boys received more nominations for most

61
liked (t=5.33, p<0.01) than girls. Males and females did not differ on socioeconomic status.
Again, due to these significant differences in the main variables, the models will be run
separately for girls and boys in order to more fully examine the study hypotheses.

Main Results
Multilevel Analysis of Friend Depression and Adolescent Depressive Symptoms. The first
model tested for the study examined the effects of friends’ depressive symptoms on an
adolescent’s own depressive symptoms. Perceptions of social self-efficacy were the strongest
predictors of depressive symptoms included in the model, confirming past research on cognitive
models of depression. As predicted, mean friends’ CES-D scores was significantly associated with
depressive symptoms (β=0.10, p<0.01), even after controlling for social cognitions (see Table 2).
For the most part, this was consistent across both school type and gender, with the magnitude
of the association remaining fairly consistent. Boys in academic schools, however, did not show
the association predicted in hypothesis one.
The second model shown in Table 2 includes the mean depressive symptoms for those
nominated as most liked. This is significantly associated with depressive symptoms for the total
sample, but when looking at the models by school type and gender, it becomes clear that this is
mainly due to professional school students (β=0.09, p<0.01). The relationship is strongest for
professional school boys (β=0.14, p<0.01), where the association between friends’ depressive
symptoms (from model 1) actually becomes non-significant when the most liked CES-D variable
is added to the model (model 2), indicating that for boys in professional schools the influences
on depressive symptoms from those whom they think are sociometrically popular are greater
than the influences from friends.

62
Table 3 reports logistic regression models with the same independent variables as Table 2,
but further examining how these variables are related to depression outcomes for those
reporting levels of depressive symptoms that indicate greater impairment. For the most part,
the results are the same as those reported in Table 2. However, the magnitude of the
associations between most liked peers’ depressive symptoms and depression appears more
clearly in this analysis for professional school students. In the previous models, both friend and
most liked depressive symptoms were significantly associated with depressive symptoms for
girls which is still the case in the logistic regression for depression; however, the adjusted odds
ratio is greater for most liked than friends’ CES-D (AOR = 1.23 versus AOR = 1.03).
Table 4 presents the results for the analysis for hypothesis two, which examined the
proportion of depressed friends rather than just the average of CES-D scores. As would be
expected, since mean friends’ depressive symptoms was associated with depressive symptoms,
the proportion of depressed friends is significantly associated with individual CES-D scores
(β=0.09, p<0.01), again with the exception of academic school boys. The standardized
coefficients in Table 4 are quite similar to those in Table 2, which could indicate this measure of
peers’ depressive symptoms is not that different from the mean CES-D score.
However, the logistic regression analysis (Table 5) shows a different picture. The
proportion of depressed friends is associated with depression (AOR=2.85, p<0.01 for the total
sample), still with academic school boys remaining the exception. The proportion of depressed
friends is more highly associated with depression for professional school students (AOR=3.34
versus AOR=2.44 for academic students, p<0.01 for both). Another difference from the models
testing the first hypothesis is the association between proportion of depressed most liked is
significantly associated with depression for academic school girls (AOR=3.35, p<0.01), which was
not the case for mean most liked CES-D. Again, that magnitude of the association between

Table 5: Standardized Coefficients for Multilevel Models of Friend Depressive Symptom on Depressive Symptoms
Depressive Symptoms (CES-D)

Total Sample
Academic School Type Professional School Type
All Girls Boys All Girls Boys
Model 1 2 1 2 1 2 1 2 1 2 1 2 1 2
Fixed Effects Estimates
Mean Friend CES-D 0.10
**
0.09
**
0.10
**
0.09
**
0.11
**
0.09
**
0.05 0.06 0.11
**
0.08
**
0.12
**
0.12
**
0.07
*
0.01
Mean Liked CES-D 0.06
**
0.04 0.05 0.03 0.09
**
0.06
+
0.14
**

Prop. of Female Friends -0.02 -0.03 -0.11 0.06 0.02 -0.05 0.09
In-Degree (Friend) -0.04
**
-0.04 -0.05 -0.03 -0.04
*
-0.07
*
-0.01
PSSE -0.19
**
-0.23
**
-0.25
**
-0.20
**
-0.16
**
-0.23
**
-0.10
**

Age 0.03
+
0.02 0.04 0.02 0.04
+
0.05
*
0.02
SES 0.08
**
0.06
**
0.06
+
0.07
*
0.11
**
0.14
**
0.09
*

Female 0.06
**
0.08
**
0.05
*

Random Effects Estimates
Level 3 (School) 0.08
+
0.07
**

School Type 0.02 <0.01
Level 2 (Class) <0.01 <0.01 0.06 0.06 <0.01 0.05 0.17
**
0.15
**
<0.01 <0.01 <0.01 <0.01 0.08 <0.01
Observations (N) 4026 3543 1976 1748 1097 980 879 768 2050 1795 1127 1013 923 782
+
p<0.10,
*
p<0.05,
**
p<0.01

63

Table 6: Odds Ratios for Multilevel Models of Friend Depressive Symptom on Depressive Symptoms
Depression (CES-D≥28 = 1; CES-D<28 = 0)
Total Sample Academic School Type Professional School Type
All Girls Boys All Girls Boys
Model 1 2 1 2 1 2 1 2 1 2 1 2 1 2
Fixed Effects Estimates (Odds Ratios)
Mean Friend CES-D 1.04
**
1.03
**
1.03
**
1.03
**
1.04
**
1.04
**
1.01 1.01 1.05
**
1.03
+
1.05
**
1.03
*
1.04
*
1.01
Mean Most Liked CES-D 1.14
**
1.02 1.02 1.01 1.34
**
1.23
*
1.63
**

Prop. of Female Friends 0.94 0.98 0.77 1.42 0.94 0.72 1.13
In-Degree (Friend) 0.96
**
0.96
*
0.96
+
0.97 0.96 0.96
*
0.94
*
0.97 0.97
Perceived Social Self Efficacy 0.69
**
0.60
**
0.57
**
0.65
**
0.87
+
0.68
**
1.17
Age 1.09 1.04 1.15 0.9 1.14 1.21
*
1.17 1.27
*
1.12
SES 1.17
**
1.06 1.14
+
1.04 1.29
**
1.49
**
1.11
Female 1.43
**
1.43
**
1.45
**

Random Effects Estimates
Level 3 (School) <0.01 <0.01
School Type 0.13
**
0.10
*

Level 2 (Class) <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.33
*
0.32
*
<0.01 <0.01 <0.01 <0.01 0.37
*
<0.01
Observations (N) 4027 3544 1976 1748 1097 980 879 768 2051 1796 1127 1013 924 783
+
p<0.10,
*
p<0.05,
**
p<0.01

64

65
proportion of depressed most liked and depression is greater than that between the
proportion of depressed friends and depression for both professional school girls (AOR=7.68
versus AOR=3.28 for friends) and boys (AOR=7.24 versus AOR=2.64 for friends).
Multilevel Analysis of Friend Social Network Status and Adolescent Depressive Symptoms. The
final analysis consisted of three models examining the association between peers’ social
network status and depressive symptoms. As Table 6 indicates, peers’ social network status is
not significantly associated with depressive symptoms. However, there are two notable
exceptions that are interesting enough to warrant the inclusion of this analysis in the study. First
and most surprisingly, for academic girls, the social network status of those they nominate as
most liked is significantly positively associated with depressive symptoms (β=0.07, p<0.05). This
indicates that the higher the status of those they nominate as most liked, the higher their
depressive symptoms. Second, for professional school boys, the social status of those they
nominate as friends is negatively associated with depressive symptoms (β=-0.06, p<0.05). This
indicates that the higher the status of professional school boys’ nominated friends, the lower
their depressive symptoms.
Interaction Terms. Due to potential interdependence issues regarding gender and
depression, it was necessary to examine the gender of an adolescent’s friends and determine if
this was the reason for the significant association between friends’ depression and an
individual’s depressive outcomes. The proportion of friends’ nominated who were female was
included in all models tested as a control variable. Interaction terms including female friends’
and mean friends’ CES-D and proportion of depressed friends were also included in each of the
models tested. These terms are not included in the tables provided as most of the interactions
were non-significant. There were two significant interactions to note, both for professional
school boys. A significant interaction term was found for mean friends CES-D and proportion of

Table 7: Standardized Coefficients for Multilevel Models of Friend Depression on Depressive Symptoms
Depressive Symptoms (CES-D)

Total Sample
Academic School Type Professional School Type
All Girls Boys All Girls Boys
Model 1 2 1 2 1 2 1 2 1 2 1 2 1 2
Fixed Effects Estimates
Prop Depressed Friends 0.09
**
0.07
**
0.09
**
0.03 0.10
**
0.11
**
0.09
**

Prop. Depressed Most Liked 0.08
**
0.04 0.09
*
-0.03 0.12
**
0.09
*
0.12
**

Prop. of Female Friends -0.01 -0.02 -0.11 0.07 -0.03 -0.04 0.06
In-Degree (Friend) -0.04
**
-0.04 -0.05 -0.03 -0.04
*
-0.07
*
-0.01
PSSE
-0.19
**

-
0.23
**

-0.25
**

-
0.20
**

-0.16
**
-0.23
**
-0.10
**

Age 0.03
+
0.02 0.04 0.02 0.04
+
0.05
*
0.02
SES 0.08
**
0.06
**
0.06
+
0.07
*
0.11
**
0.14
**
0.09
*

Female 0.06
**
0.08
**
0.05
*

Random Effects Estimates
Level 3 (School) 0.10
**
0.08
**

School Type <0.01 <0.01
Level 2 (Class) <0.01 0.04 0.09
*
0.12
**
0.06 0.14
*
0.18
**
0.17
**
<0.01 <0.01 <0.01 0.10 0.07 0.12
+
Observations (N) 4028 2659 1977 1329 1098 753 879 576 2051 1330 1127 775 924 555
+
p<0.10,
*
p<0.05,
**
p<0.01

66

Table 8: Odds Ratios for Multilevel Models of Friend Depression on Depression
Depression (CES-D≥28 = 1; CES-D<28 = 0)

Total Sample
Academic School Type Professional School Type
All Girls Boys All Girls Boys
Model 1 2 1 2 1 2 1 2 1 2 1 2 1 2
Fixed Effects Estimates (Odds Ratios)
Prop. of Depressed Friends 2.84
**
2.44
**
2.75
**
1.59 3.34
**
3.28
**
2.64
+

Prop. of Depressed Most Liked 3.39
**
1.68 3.35
**
0.28
+
8.43
**
7.68
**
7.24
**

Prop. of Female Friends 0.94 0.99 0.78 1.52 0.90 0.72 1.1
In-Degree (Friend) 0.96
**
0.96
*
0.96
+
0.96 0.96
*
0.94
*
0.97
Perceived Social Self Efficacy 0.69
**
0.60
**
0.64
**
0.57
**
0.65
**
0.87
+
0.68
**
1.17
Age 1.09 1.04 1.15 0.9 0.69
+
1.14 1.17 1.12
SES 1.17
**
1.06 1.14
+
1.14
+
1.25
*
1.04 1.29
**
1.43
**
1.49
**
1.60
**
1.11
Female 1.43
**
1.57
**
1.43
**
1.59
**
1.45
**
1.60
**

Random Effects Estimates
Level 3 (School) <0.01 0.02
School Type 0.13
**
0.09

Level 2 (Class) <0.01 <0.01 <0.01 0.13 <0.01 0.05 0.31
*
0.50
*
<0.01 <0.01 <0.01 <0.01 0.35
*
<0.01
Observations (N) 4029 2660 1977 1329 1098 753 879 576 2052 1331 1127 775 925 556
+
p<0.10,
*
p<0.05,
**
p<0.01

67

68
female friends for professional school boys (β = -0.37 for depressive symptoms; AOR=1.37 for
depression, p<0.05 for both). The proportion of depressed friends and proportion of female
friends’ interaction was also significant (AOR=1.77, p<0.01), though for depression and not
depressive symptoms in this case. This indicates that the proportion of friends’ who are female
does influence the depression scores of friends for professional school boys. For these boys,
having more female friends results in their friends’ having higher depression scores, which in
turn influences that individual’s own depression outcomes.
Other interaction terms including friends’ demographic characteristics (age and SES) and
mean friends’ depression were also tested. None of these other interaction terms were
significant, so the parsimonious models without the interaction terms are presented.
DISCUSSION
A robust body of work exists among the peer relationship research citing meaningful links
between adolescents’ and their friends’ psychological characteristics. Research supports the
proposition that various social processes occur concurrently at multiple levels of the social
environment, though the bulk of the work examining interpersonal relationships and influences
focuses exclusively on individual characteristics often to the exclusion of group and system level
factors. There are many determinants of depressive symptoms (for review, see Petersen et al.,
1993), but it is becoming clear that an individual’s experience of depressive outcomes also
depends on whether others in the individual’s social network report depressive symptoms.
Depression and depressive symptoms, therefore, are not solely a result of individual
experiences, cognitions, and choices but can be an attribute of groups and networks.

Table 9: Standardized Coefficients for Multilevel Models of Friend Social Network Status on Depressive Symptoms
Depressive Symptoms (CES-D)
Total Sample
Academic Professional
All Girls Boys All Girls Boys
Model 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
Fixed Effects Estimates
Mean Friend Status
(Friend) -0.01 0.01 -0.01 0.03 -0.03 0.01 -0.06
*

Mean Liked Status
(Friend) 0.02 0.02 0.02 0.01 0.02 0.05
+
-0.03
Mean Liked Status
(Liked) 0.03
+
0.04
+
0.07
*
0.01 0.02 0.02 <0.01
Mean Friend CES-D 0.10
**
0.10
**
0.11
**
0.05 0.11
**
0.13
**
0.07
*

Prop. of Female Friends -0.01 -0.02 -0.11 0.07 0.02 -0.04 0.08
In-Degree (Friend) -0.04
*
-0.04
+
-0.05
+
-0.03 -0.04
+
-0.07
*
-0.08
**
-0.05
+
0.01
PSSE -0.19
**
-0.23
**
-0.25
**
-0.21
**
-0.16
**
-0.23
**
-0.08
**

Age 0.03
+
0.03 0.04 0.01 0.04
+
0.04 0.06
*
0.03
SES 0.08
**
0.06
**
0.06
+
0.06
*
0.10
**
0.11
**
0.09
*

Female 0.06
**
0.08
**
0.05
**

Random Effects Estimates
Level 3 (School)
0.08
+
0.09
+
0.08
+

School Type
0.02 0.02 0.02

Level 2 (Class)
<0.01 <0.01 <0.01 0.06 0.06 0.09
*
<0.01 <0.01 0.08 0.17
**
0.17
**
0.16
**
<0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.07 0.06 0.07
Observations (N) 4626 4442 3935 2267 2162 1929 1202 1164 1048 1065 998 881 2359 2280 2006 1262 1229 1095 1097 1051 911
+
p<0.10,
*
p<0.05,
**
p<0.01
69

70
The results of this study support previous research examining the peer contagion effects
on internalizing symptoms, that adolescent depressive symptoms are associated with those of
their peers. After controlling for individual-level social cognitions and social network status, self-
reported friends’ depressive symptoms were positively associated with depressive symptoms
for Chinese adolescents, with the exception of academic school boys. Those an adolescent
nominated as “most liked,” often considered sociometric popularity, also had some effect on
depressive symptoms for professional school students, particularly boys. Both friend
nominations and most liked nominations were significantly correlated for all subgroups, as are
friends’ and most liked depressive symptoms, though the correlation was moderate-to-small
(r<0.50 for all groups, including professional school boys). This points to these two social
network measures as being related yet separate distinctions that (as indicated by the model 2
findings in both Table 2 and Table 3) contribute meaningfully in their own ways to depressive
outcomes.
The second portion of the analysis examined peer social status and depressive symptoms
and the results were more modest. The social network status of those nominated as “most
liked” was positively related to depressive symptoms for academic school girls. Similar findings
were reported by Prinstein (2007) in a study of peer contagion, though among boys rather than
girls. Prinstein suggested that adolescents, in this case academic school girls, could be more
likely to “emulate the behaviors of friends who are especially popular among peers.” The
present findings could indicate an imbalance of social “power” or “influence” for girls in
academic schools, or at least a perceived discrepancy between their own popularity and that of
those they consider to be particularly well-liked among their peers. It could also be, as indicated
by the higher average “most liked” nominations received for boys than girls, that boys are the
ones with all the social “influence” and this imbalance or discrepancy may be more noticeable or

71
meaningful in academic rather than professional schools. This does make some sense when
considering the culturally supported gender inequality that still occurs in China, which could
make girls feel as if they have to “work twice as hard for half as much” in terms of social
influence.
Friends’ social status was negatively associated with depressive symptoms for professional
school boys. This result seems the more straightforward of the two significant associations
between peers’ social status and depressive symptoms, since it seems logical that the more
friends an individual’s friends have the fewer depressive symptoms they report. It indicates that
those who are accepted within their classrooms and who have friends who are accepted
experience less internalized distress. Those who are less accepted and are part of a peer group
that is less accepted overall report more depressive symptoms. These are often the individuals
and peer groups that are considered “at risk” in much of the adolescent health behavior
research for outcomes such as substance use, deviant and antisocial behavior, aggression, and
suicidal ideation.
The results for friends’ and most liked depressive symptoms relationship to an
adolescent’s own depressive outcomes was significant for the total sample of Chinese
adolescents, but it was clear when examining the models by the different subgroups that there
are noticeable differences in magnitude and meaningfulness for these groups. Chinese youth
experience unique social environments that, in some regards, are quite different from Western
youth. As much of the previous work on peer contagion effects of depression-related outcomes
has been conducted with Western cultures, this study adds to the work by examining these
hypotheses in a collectivist, Eastern culture. The meaning and influence social interactions have
for Chinese adolescents may affect their association with depressive symptoms. This is a
promising avenue for future work, as the impact of rapid social and cultural changes in China is

72
still unclear and rates of psychological distress and outcomes, such as depression and suicidal
behavior is on the rise (Daly, 2009).

Limitations
The findings from this study should be considered within the limitations of the research.
First, past research examining the influence of social network members (Fowler & Christakis,
2008) has found that the strength of an effect depends on the strength of the tie (e.g. the
strength or quality of a friendship). This study did not include measures of the quality or
strength of a reported relationship. It may also be the case that the strength of any association
is stronger for relationships that are reciprocated (e.g. a reciprocal friendship occurs when both
people nominate the other as a friend). Another consideration is the fact that unreciprocated
friendships themselves may cause or contribute to depressive symptoms.
Second, while the associations reported were significant, the effects were small. This
could potentially limit the meaningfulness of the findings; however some have suggested that
small effect size can still have practical applicability (Rosenthal, Rosnow, & Rubin, 2000). This, in
conjunction with a fairly conservative analytic strategy suggests that the results do represent
meaningful processes for adolescents across the social environments sampled for this study.
The CES-D is a self report assessment of depressive symptoms, rather than a clinical
assessment of syndrome or disorder, therefore the findings of this study do not generalize to
populations who have diagnosable disorders and it is not possible to determine whether the
suggested peer contagion effects are important mechanisms operating in more clinical
populations. Also, though the design and measurement of the study enables examination of
data from more than one source, which is a considerable strength of the study, the social

73
network nomination format was limited to school-based peers which may not be the most
influential peers for all youth.
Despite these limitations, the results confirm that peers’ depressive symptoms have a
significant association with depressive symptoms for Chinese adolescents.

Implications and Future Directions
The data analyzed for this study do not allow for identification of causal mechanisms of
peer contagion effects on depression-related outcomes, but the results can suggest and aid in
understanding potential causal pathways for future testing with longitudinal data. Existing
longitudinal results already indicate that friends’ behaviors are significant predictors of
adolescents’ own behavior over time (Hogue & Steinberg, 1995; Jaccard, Dodge, & Blanton,
2005; Prinstein, 2007). It may be that those experiencing depressive symptoms share their
sadness or dissatisfaction with their friends or commiserate with their peers about their own
depressive feelings, thoughts, and attitudes. Some have suggested that emotion is genuinely
contagious, and it may be that over time this contagious affect can result in depression. It may
also be that social behavior changes towards friends and other peers when one experiences
depressive symptoms, which can change and quality and tone of social interactions causing
others to experience depressive symptoms as a result.
Adolescents are embedded within social networks that have the potential to influence
health behavior in both positive and negative ways. From this network perspective, the health
and well-being of one individual has implications for the health and well-being of their peers.
The role social relationships play in mental and physical health is significant and by examining
theories formulated with considerations of the multiple levels the social environment
encompasses, such as peer contagion theories, a better understanding of the complex social

74
processes that occur to influence health behavior can be gained. This is crucial for the
development, design, and implementation of effective interventions. Past prevention research
efforts have spent a great deal of time, effort, and resources attempting social interventions
focused on adolescents and their peers for a multitude of health outcomes with mixed success.
Even with the large body of work that exists on social influences interventions among
adolescents, the understanding of the ever-changing social forces and environments youth
inhabit is still surprisingly sparse. Attention to social network influences on depressive
symptoms and other depression-related outcomes among adolescents could provide a greater
understanding of peer contagion and other social processes impacting the health of young
people.

75
CHAPTER 4:
The Classroom Social Network and Depressive Symptoms: Do Network Composition and
Characteristics Influence Adolescent Depression?

CHAPTER FOUR ABSTRACT
A school class can be seen as a combination of work group and peer group. The social
interplay that takes place within the school class can be considered an influential life situation
for the individual adolescent’s social, emotional, and behavioral development. Methods:
Measures of depressive symptoms and social network nominations were examined at the
classroom social network level using multilevel linear regression models for an overall sample of
124 Chinese 10
th
grade classrooms in 24 schools. Results: For the total sample, proportion of
peripheral network members based on friend nominations was associated with depressed
students (β=0.26, p<0.05). A higher proportion of high status students in a classroom was
positively associated with the proportion of depressed in a class for academic school classes
(β=0.39, p<0.01). Findings suggest that classes with higher proportions of peripheral members
have more depressed students and those with higher proportions of favorite status members
have fewer depressed students in professional schools. Measures of social network density,
centralization, and reciprocity were not significantly associated with depression in this sample.
Conclusions: An individual adolescent experiences their social environment often as more than
a mere sum of its parts, so examination of each element individually or even additively may not
fully describe the complex social processes youth encounter. Social network influences work
together interdependently to collectively impact health and well-being.

76

INTRODUCTION
Several decades of research make a clear link between social relationships and ties and
depressive symptoms during adolescence. In general, these studies have examined an individual
adolescent’s perceptions of their social ties and social acceptance and have rarely considered
social relationships from a holistic perspective – one that considers social interaction as a
composite of the perceptions and actions of others. It is therefore important to also view the
relationship between social relationships and depression-related outcomes from a broader,
more systemic perspective to better understand the processes involved with this association.
Group characteristics, such as social network structure, role expectations, and hierarchical social
organization, can potentially have a large impact on psychological well-being (Boivin, Coie, &
Dodge, 1995).
There have been relatively few studies that have investigated the structure of social
networks as it relates to depressive symptoms among adolescents (for exceptions see Hansell
1985; Ueno 2005). Social network measures and methodologies have at least one distinct
advantage over non-network studies in that they do not rely on self-report alone. It has been
proposed that participation in social networks subject an individual to “social controls and peer
pressures that influences normative health behaviors” (Cohen, Gottlieb, & Underwood; 2000). It
is also thought that shared norms regarding physical and psychological well-being can be
powerful sources of social influence that have direct effects on the health and behavior of social
network members (Berkman, Glass, Brissette, & Seeman, 2000).
The study of context and health outcomes spans various disciplines. Much of the
sociological literature, for example, has examined the neighborhood context as an important

77
contributor to health. For adolescents, the school is a more relevant environment. It is the main
social environment in which they interact with their peers and serves as the very core of youth
culture.

School Context during Adolescence: The Classroom Social Network
A school class can be seen as a combination of work group and peer group. It is constant
over a relatively sustained period time and is a setting in which adolescents spend a great deal
of time at an age when peer relations are important. At the same time, group membership can
be considered imposed, rather than a matter of selection. In some ways, group members do not
have the ability to exclude or change network membership on their own. They do not control
the entry or egress of members as occurs in other types of social networks. These factors all
combine to create an environment in which social hierarchies, both simple and complex,
commonly occur; unique social norms and cultures are developed; and positive and negative
behaviors modeled. Therefore, the social interplay that takes place within the school class can
be considered an influential life situation for the individual adolescents’ social, emotional, and
behavioral development.
Of the different levels of network analysis (individual, dyadic, and group), the majority of the
work has been done at the first two levels, with far fewer studies examining the influence the
structure and attributes of the social network may have on the overall health or health behavior
of the group (Cillessen, 2007). Much of the work that has examined the influence of classroom
network composition and structure has been conducted with children. Findings from studies
examining network characteristics of classrooms have found support for the idea that classroom
social structure can influence health behaviors in children and adolescents. For example,
Ostberg (2003) found that children in classes with no isolated or extremely popular students had

78
fewer psychosomatic symptoms overall than those in classes that had a more disproportionate
distribution of influence (e.g. a few very popular people maintaining the majority of the social
“power”). Another study with Chinese children found that in classes characterized as having a
high level of social withdrawal, the behavior was more acceptable than in classes with low levels
of withdrawal (Chang, 2004). Studies examining aggression have found similar outcomes, with
aggressive individuals being more accepted in aggressive playgroups where they are typically
rejected in less aggressive playgroups (Boivin, Dodge, & Coie, 1995; Stormshak, et al., 1998;
Wright, Giammarino, & Parad, 1986).
Psychological functioning and adaptation are influenced by numerous factors. From an
ecological perspective, the classroom is one social context where adolescents spend the
majority of their time with peers and should therefore prove influential to mental health
outcomes, such as depressive symptoms and depression. The structure and organization of the
classroom social network is a facet of the adolescent social environment in need of further
elucidation.

Study Hypotheses
The present study examined how the structure and characteristics of the classroom social
network influences group-level adolescent depression outcomes. More specifically, it examined
the social structural characteristics and organization of school classrooms to determine whether
these factors influence overall depressive tendencies within a group. The level of analysis, as
identified by social network methodologies, was the network level. The primary purpose of the
study was to discover if classroom network composition and characteristics were associated
with students’ depressive symptoms. To examine this aim, three hypotheses were tested:

79
H1: Classrooms with higher proportions of marginalized and peripheral network members
(those on the fringes of a network) have a greater proportion of depressed students.
H2: Network density and reciprocity are associated with lower proportions of depressed
students.
H3: Centralization (network level centrality) is positively associated with the proportion of
depressed students.

METHODS
Sampling. Unlike in the United States, China has two primary high school types, academic
(regular) and professional (vocational). Students are tested in middle school and scores from
those tests determine which school type they will attend. Generally, those attending academic
schools have a rigorous schedule of courses for three years of high school (10
th
through 12
th

grades), designed to increase the chances students will pass college entrance exams. Those
attending professionals schools are enrolled in a career track, such as computers/IT or public
relations, and will complete an internship during their final year(s) in high school with the goal of
entering the workforce once they have completed high school. Both school types were included
in the study, with an equal number of schools and students from each school type. There are
more professional school classes in the study, however, due to a trend for these classes to be
smaller.
The study included a total sample of 124 10
th
grade classroom social networks from a
convenience sample of 24 schools, 12 of each school type, with 5,563 students. Aggregate data
was calculated for each class.

80
Data Collection. The data for this study was collected as pre-intervention, baseline
measurements of a larger parent study examining the efficacy of a school-based tobacco
prevention program focusing on social influences. All assessments were paper-based surveys
and all participants had active parental consent and student assent. Pre-intervention
assessments were collected prior to any intervention delivery, with the pre-test occurring
approximately one week prior to intervention implementation for those assigned to the
Program condition. The baseline survey was collected approximately one month prior to the
pre-test. There were two reasons for the two waves of pre-intervention data collection: 1) the
major holiday, Chinese New Year, occurred during that time and would have disrupted data
collection efforts and 2) the assessments were each fairly substantial and splitting the measures
between two surveys prevented students from becoming fatigued and improved the quality of
the data.

Measures
Depressive Symptoms (CES-D; Radloff, 1977). The CES-D is a self-administered scale
containing 20 items. The answers to each item, which indicate the frequency of the symptoms,
are structured on a 4-point likert scale ranging from 1 (Less than 1 day) to 4 (5-7 days). The
answers are summed across all 20 items. The total scores can range from 0 to 60. Radloff (1977)
suggested and it is commonly accepted that persons who score 16 or above are likely to be
clinically depressed; however, some studies suggest that the cut-point needs to be raised to 24
or higher if the tool is to achieve appropriate diagnostic characteristics (Gotlib, Lewinsohn, &
Seeley, 1995; Roberts, Lewinsohn, & Seeley, 1991). A Mandarin-Chinese version of the CES-D
(MC-CES-D) has been validated and used in Hong Kong (Cheung & Bagley, 1998; Lam et al.,
2004) and among community samples of adolescents in Taiwan where cut-point scores ≥ 28

81
have been determined, with those above this score more likely to have major depressive
disorder (Yang, Soong, Kuo, Chang, & Chen, 2004; Lin et al., 2008). The proportion of depressed
students, as measured by those reporting scores greater than or equal to 28 on the CES-D, was
the primary outcome of interest and was calculated for each class.

Social Network Measures. Sociometric procedures were used to obtain measures of adolescents’
social network status. Peer nominations were used to collect information from each student
regarding peer status. Adolescents were asked to list up to five people in their class who they
considered their best friends. Students wrote the first and last names of their friends. Similarly,
students nominated the five people they considered to be the most liked by people in their
class. Peer nominations were later matched to study id’s and names were dropped from the
data to protect confidentiality. These nominations were then used to calculate network metrics
for the study.
Network level measures of density, reciprocity, and centralization were calculated for each
class from these nominations. Network-level density describes the general level of linkage
among the members in a network. It is the proportion of nominations in a network relative to
the total number possible. A dense network is one in which all the network members are
connected directly to every other member, a rare to impossible occurrence in real-world social
networks, particularly among adolescents. A sparse network is one where there are few ties
between members and a network where communication and social influence would be slow to
diffuse. Reciprocity is the extent to which ties are reciprocated between network members in
the network as a whole. Assessing network reciprocity can identify imbalances, indicating
dysfunction in the network. Centralization is the difference between the number of nominations
for each network member divided by maximum possible sum of differences. A centralized

82
network will have many nominations focused around one or a few network members, while a
decentralized network is one in which there is little variation between the number of
nominations each member receives (see Figure 6).

In order to define peer or social network status in a school classroom, researchers often
divide individuals into different status categories based on their social network nominations.
The status categories are created to represent a more normal distribution. Five social network
status categories were created from the friend nomination data for this study, similar to those
created in past research (Ostberg, 2003). The first category included those who did not receive
any nominations. This group, the “marginalized” category, is usually considered at highest risk
for many negative health behavior outcomes. The second group consists of the “peripheral”
members of the network, who received one or two friend nominations. The “accepted” category
is the largest and includes those receiving three to five nominations. The “popular” category is
those with six or seven nominations and the “favorite” those who received eight or more (see
Figure 7).

83

The same five status categories were created for the most liked nominations; however,
because the distribution for this type of nomination differs from the friend data, the
construction of the categories was different. Most liked nominations, often called “sociometric
popularity,” usually have a larger range than friend nomination data with more network
members receiving zero or only one nomination. A few network members usually receive the
bulk of the most liked nominations in a classroom. Therefore, the marginal category remains the
same, but the others are slightly different (peripheral = 1 nomination; accepted = 2-4; popular =
5-8; and favorite = 9 or more nominations). The proportion of students in each class in each
status category was calculated as an additional measure of network structure (see Figure 8).

Demographic covariates and control variables. Variables controlled for in the models included
aggregated measures of gender (ratio of female students in a class), school type, class size,
socioeconomic status (means), and age (mean). Socioeconomic status was calculated using both
mother’s and father’s education level and a measure of an adolescent’s disposable income.

84

Data Analysis
Multilevel linear regression analysis procedures were used to analyze the data. Multilevel
analysis was needed to explain classroom-level outcomes with school-level characteristics. All
analyses were done using STATA version 11 (StataCorp, 2009), with social network variables
calculated using UCINet (Borgatti, Everett, & Freeman, 1999) and then imported into STATA.
The analyses for all study hypotheses were treated similarly, with a two-level model
including a random effects model for school-level effects and a fixed effects model for
classroom-level effects. The control variables were also consistent across the different models,
including age, socioeconomic status, class size, and the proportion of female students (gender).
In the models examining peer status, the accepted group was used as the reference to avoid
over-specification of the models, because it was the largest group, and also because it is
considered the “normative” group. All coefficients reported are standardized in order to better
compare across models and variables.

85
RESULTS
Sample Characteristics
Distribution of Demographic Characteristics and Main Study Variables. Table 1 presents
descriptive data for the sample. Of the 124 classroom that participated in the study, 56 were
from academic high schools and 68 from professional schools. The greater number of classes for
professional schools was due to the smaller class size in those school (an average class size of
46.3 versus 52.6 for academic schools).The average age of participants was 15.8 and the mean
SES, ranging from 1-10 on a composite measure including mother and father education and
allowance per week, was 4.4. The mean proportion of female students in classes was 0.49,
indicating equal numbers of female and male participants.
The mean depressive symptoms score (CES-D; range: 1-60) for the sample is 17.9. Using a
cut-point score of 28 or greater on the CES-D resulted in an average proportion of depressed
students in a class of 17%.
The average number of friend nominations received (friend in-degree) was 3.7, and 2.3 was
the average for most liked nominations received (liked in-degree). The most friend nominations
received by a single student was 17 and the most liked nominations received by a single student
was 40. On average, 44% of students in classrooms were categorized as “accepted” using
friendships nominations, and 36% using most liked nominations. The mean proportion of
marginalized and peripheral network members showed the biggest differences between the
different nomination types, with more marginal members for most liked (12%) than friend
networks (only 2%).
Differences in Study Variables by School Type. A series of t-tests were conducted to
determine equivalence on the main study variables for adolescents in the different school types.
Gender was not different between school types.

86
Table 10: School & Classroom Characteristics
Frequency Range
Participating 10
th
grade classes 124
Academic 56
Professional 68
Mean class size (SD) 51.8 (12.3) 24-87
Academic 55.6 (13.8) 33-87
Professional 48.3 (9.3) 24-67
Mean age (SD) 15.8 (0.72) 13-19
Mean proportion of female students 0.49 (0.19)
Mean socioeconomic status (SD) 4.4 (1.5) 1-10
Depression Outcomes
Mean CES-D 17.9 (11.0) 0-60
Mean proportion depressed (CES-D≥28) 0.17 (0.07)
Social network measures
Mean Density (SD) 0.07 (0.02)
Mean Centrality (SD) 0.07 (0.03)
Mean Reciprocity (SD) 0.35 (0.07)
Mean proportion of status categories
Friend nomination categories
Marginal 0.02 (0.03)
Peripheral 0.26 (0.09)
Accepted 0.44 (0.09)
Popular 0.15 (0.06)
Favorite 0.06 (0.04
Most liked nomination categories
Marginal 0.12 (0.07)
Peripheral 0.11 (0.05)
Accepted 0.36 (0.08)
Popular 0.14 (0.06)
Favorite
0.09 (0.05)
Mean In-Degree Measures (SD)
Friend in-degree 3.7 (2.4) 0-17
Liked in-degree 2.3 (4.3) 0-40

However, there was a greater mean proportion of popular adolescents based on friend
nominations in academic schools versus professional schools (t = 2.1, p<0.05; 16% versus 14% in
professional schools). The same is seen for the proportion of favorite students based on most
liked nominations (t = 2.7, p<.0.01). Network level reciprocity was also different by school type
(t = 4.1, p<0.01) with reciprocity lower in professional schools. For demographic variables,
professional school students had a higher mean age (t = 12.2, p<0.01), lower average SES (t =
14.7, p<0.01), and smaller class sizes (t = 3.1, p<0.01). The most meaningful differences between

87
school types for the study were in depressive symptoms, with academic classrooms having a
greater proportion of depressed students (t = 5.3, p<0.01; mean proportion of 20% versus 14%
for professional schools). Therefore, due to these significant differences by school type, all
models were run separately for each school type.

Main Results
Multilevel Linear Regression Analysis of Social Network Status Measures and Depression.
Table 2 presents the results of models examining social network status (models 1 & 2) and
structure (model 3) on depression, the proportion of depressed students in the class in this case.
The first model tested the associations between status variables based on friend nominations
and depression. Hypothesis one, stating that greater proportions of lower status network
members would be associated with greater classroom-level depression was partially supported
in the analysis. For the total sample, proportion of peripheral network members based on friend
nominations is associated with depressed students (β=0.26, p<0.05), but this association is not
significant for either school type individually. A higher proportion of favorite status students in a
classroom is positively associated with the proportion of depressed in a class for academic
school students (β=0.39, p<0.01). This finding seems somewhat contrary to hypothesis one. It is
the only finding of significance for academic school students. For professional school students,
proportion of peripheral (β=0.30, p<0.01)and favorite (β=-0.23, p<0.05) status individuals in a
class based on most liked nominations are associated with the proportion of depressed in a
class, though in opposite directions. These findings suggest that classes with higher proportions
of peripheral members based on most liked nominations have more depressed students and
those with higher proportions of favorite members have fewer depressed students, but only in
professional schools.

Table 11. Standardized coefficients (β) for multilevel models for effects of social network measures on classroom depression
Proportion of Depressed (CES-D ≥ 28) in Class (β)
MODEL 1 MODEL 2 MODEL 3
Total Academic Professional Total Academic Professional Total Academic Professional
Fixed Effects Estimates
Friend Nomination Network Status
Marginal 0.11 0.12 0.19
Peripheral 0.26
*
0.28
+
0.18
Accepted ref. ref. ref.
Popular 0.03 0.02 0.01
Favorite 0.12 0.39
**
-0.05
Liked Nomination Network Status
Marginal 0.17
+
0.12 0.18
Peripheral 0.24
*
0.14 0.30
**

Accepted ref. ref. ref.
Popular 0.15 0.31
+
0.08
Favorite -0.05 0.15 -0.23
*

Network Density 0.15 0.36 0.05
Network Centralization -0.12 -0.17 -0.12
Network Reciprocity -0.01 -0.13 -0.08
Mean Age 0.02 -0.25 0.2 0.03 -0.05 0.17 0.53
**
0.02 0.67
**

Mean SES 0.54
**
-0.04 0.67
**
0.53
**
0.067 0.65
**
0.38
**
0.47
+
0.43
**

Female (proportion) 0.35
**
0.4 0.37
**
0.36
**
0.28 0.38
**
-0.49
**
-0.28 -0.75
**

Class Size -0.42
**
-0.31
*
-0.51
**
-0.41
**
-0.28
+
-0.52
**
0.53
**
0.02 0.67
**

Random Effects Estimates
School 0.11 0.18 0.19 0.10 0.21 0.21 0.08 <0.01 0.10
School Type 0.12 0.11 0.09
Observations 124 56 68 124 56 68 124 56 68
+
p<0.10,
*
p<0.05,
**
p<0.01
88

89
Socioeconomic status and gender were associated with the outcome for professional but
not academic schools. In the case of socioeconomic status, it was the greatest predictor of the
outcome for professional students, with the magnitude of the association quite high (β>0.70 for
all three models). Interestingly, higher SES was associated with greater depressive symptoms.
Class size was negatively associated with depressive symptoms for professional schools as well.
The third model presented in Table 2 examined three measures of social network
structure: density, centralization, and reciprocity. This model tested hypotheses two and three
for the study. None of these network measures was significantly associated to the depression
outcome, so results did not support the hypotheses.

DISCUSSION
During childhood and adolescence, an individual spends a major portion of their time in
school and the classroom provides the background social environment that includes social
interactions between classmates. This environment, and its attendant social processes and
pressures, is an inevitable circumstance of daily life for young people. Naturally, the classroom
social network should have an influence on healthy adaptation and adjustment. The social
network methodology used in the present study allowed for the examination of adolescent
social relationships in their natural and complex state, which can better inform future social
theoretical models of risk and protective factors for depression.
Group characteristics may influence the shape of the peer status distribution. For example,
certain characteristics such as the level of psychological distress, could determine the existence
of the marginal or favorite peer status categories in a classroom. A large spread in the peer
status distribution in a class, or where the social status differences between students is great,
could be an indication of an imbalance in social influence and “power.” In such situations, we

90
expect that issues of peer status may be more important and more closely related to
depression. The existence of a large proportion of marginal network members is most likely
detrimental, not only to those adolescents, but to all students in a classroom, who may fear
ending up in that position themselves or could indicate other issues such as bullying or low
tolerance for those who do not conform.
The results indicate that social network structure and characteristics are associated with
depression outcomes more for professional school students in China than for academic school
students. This could have several explanations that require additional investigation to further
describe. One explanation could be that depression and depressive symptoms include a more
social causal mechanism for professional school students – that is, social factors are more
important for the mental well-being of those in professional schools versus those in academic
schools. One possible explanation is that because of the immense stress and pressure often put
on academic school students by parents, teachers, peers, and society in general to study hard
and learn more in order to gain entrance to college and set up a more secure and profitable
future, a larger factor influencing depression outcomes would be stress. It may also be that
academic school students don’t have the time to socialize in the same manner as professional
school students, or that the nature of their social interactions are different. A concern with a
“social life” or peer status, not related to academic success, could just be a “luxury” these
students cannot afford to indulge in. This generalization does not preclude stress and academic
strain from influencing depression among professional students, however. Social interactions
and competence may also be skills that are developed and valued more in professional schools;
for example, these students are enrolled in majors such as public relations and hospitality where
such skills would be seen as an asset. It may also be that adolescents for whom social
interactions are more influential to their well-being are drawn to professional schools or certain

91
majors within those schools. This suggests further study on the differences between these
school types, both for individuals and at the classroom social network level, is warranted. This is
also a unique cultural phenomenon in China that could benefit from comparison to other
cultures that have different school types and systems.
Social status distribution within a classroom was related to classroom level depression for
professional school students. The proportion of lower status network members serves as a risk
factor for depression, or more specifically, the proportion of depressed adolescents in a
classroom. These results indicate that social status in the classroom is important for adolescent
mental well-being and that it is important to consider the entire social network composition
within the classroom. Ostberg (2003) pointed out that the existence of a wide status distribution
in a class can indicate “that the social interplay to a higher degree circulates around issues of
popularity and status.”
The existence of the lower status positions could indicate a less healthy social network or
at least less than optimal social conditions within a class. Higher proportions of network
members in the lower status categories could mean there is less tolerance within a group or
that there is more pressure placed on network members to conform. Why classrooms have
different distribution bears further study. Social network information such as that used in this
study could examine this question by investigating the characteristics of classrooms with
different types of status distributions and also identify if patterns of status distribution exist.
The teacher and other school administrators and staff could also have an important impact
on these issues. Teachers have a lot influence on the classroom environment and can set the
tone and norms for a social network. They can also lend support to certain groups, structures,
and characteristics of the network that can affect social interactions and their association with
well-being in the classroom.

92
Study Limitations
The findings from this study should be considered within the limitations of the research.
Studies that have examined ecological influences on behaviors considered to be individual in
nature, such as mental well-being via depression-related outcomes, have often been accused of
an inferential error referred to as the ecological fallacy. This type of error has been described as
misleading due to explaining micro-level events in terms of macro-level properties. Based on
this description, much of the work examining social network influences on health behavior
outcomes would fall prey to this mistake. Although this limitation warrants mention, the use of
multilevel models, covariates to control for contextual effects (e.g. SES, the proportion of female
students, etc.), and the use of outcome measures constructed from the proportion of depressed
students in each classroom, rather than just taking a simple mean of CES-D scores, allows for
confidence that an error in inference based on the ecological fallacy is less likely in the present
study.
Generalizability is limited to Chinese adolescents attending either professional or academic
high school in a large urban environment. The data collection process also limited nominations
within classrooms and could affect the generalizability of the study. The CES-D is a well validated
measure of depressive symptoms and has been linked to more clinical outcomes; however, the
use of this self report measure does not extrapolate to more clinical populations of adolescents.
The primary limitation of the analysis and the study overall is the reliance on cross
sectional data, which limits the ability to draw causal inferences. This may be even more of an
issue in studies examining multiple levels of analysis as there are often concerns regarding
school and classroom level effects resulting from some type of selection bias or other
circumstances that could lead to systematic biases.

93
Regardless of the limitations, these findings suggest that classroom social network
structure and attributes contribute to adolescent psychological well-being through associations
with depression outcomes.

Implications and Future Directions
This study adds to the relatively limited research on the association between social
network structure and characteristics and mental well-being. Social network research allows for
closer examination of the influence social structure and group-level relationships have on health
behavior, and can identify how these network characteristics can impact different groups, such
as academic and professional school classes, in unique ways that promote different outcomes.
Longitudinal research should be done to provide better understanding of the social
processes involved in the relationship between social network influences and adolescent
depression. Longitudinal data can further reveal socialization versus selection effects, which
data that is cross sectional in nature cannot. Therefore, future work on the longitudinal data
collected from this sample, once it becomes available, is critical in order to more closely
examine the associations presented in the current findings. Also, most of the work examining
social network influences on the mental well-being of adolescents has been done with Western
samples. It would be of interest to have a comparable sample of adolescents from a Western
culture with which to compare these findings in order to further examine any cultural and
societal differences and influences.
An individual adolescent experiences their social environment often as more than a mere
sum of its parts, so examination of each element individually or even additively may not fully
describe the complex social processes youth encounter. Social network influences work
together interdependently to collectively impact health and well-being.

94
Certain network characteristics could be encouraged that promote positive health behavior
outcomes, while others receive subtle deterrents or dissuasion in order to provide protective
and preventive effects. Modification of existing social networks could prove a challenge, but for
some outcomes might be necessary. However, the best results might most easily come from
providing support, guidance, and opportunities for healthy growth of existing classroom social
networks for adolescents. A greater understanding of how social network influences affect
depression outcomes can suggest the most efficient and effective ways in which to intervene
with youth. A developmental mandate for adolescents’ is the need to feel connected to others,
so their health and well-being is also connected to that of their peers. Interventions focusing on
the social environment, the classroom in particular, may result in greater overall impact. This
type of intervention, primarily preventive in nature, has the added benefit of cost effectiveness
over more intensive and targeted intervention and treatment later. They also have the potential
due to their ecological nature to spread to others, thus setting up potential secondary effects.

95
CHAPTER 5: DISCUSSION AND CONCLUSIONS

The goal of this dissertation was to investigate the impact of social network influences on
depression in adolescents. The three studies assessed this goal by examining the social
determinants of depressive symptoms among adolescents at different levels of analysis: the
intrapersonal or individual level, the friendship group level, and the classroom social network
level. Study One tested a main effects model of depression at the individual level, hypothesizing
that social integration would be associated with depressive symptoms. It investigated whether
network status was related to depression and depressive symptoms. Study Two moved to the
next level of analysis and looked at depressive contagion. This was assessed by testing whether
the level of friends’ depressive symptoms and friend social status was related to an adolescent’s
own depressive symptoms. Finally, Study Three explored the network level by looking at the
classroom social network structure and characteristics and examining how the structure of the
classroom social network works to influence depression outcomes. This was accomplished by
testing whether social network measures and proportions of high and low status members were
related to the proportion of depressed students in the class.

Summary of Main Findings
The analyses in this dissertation examined the classroom social network, the friendship
group, and individual data from a large sample of Chinese high school students. Results from the
statistical models in the three study chapters indicate that, overall, social network influences are
associated with depression-related outcomes. However, this basic pattern varies in relation to
gender, school type, and social integration at the classroom, peer group, and individual levels.

96
At the individual level, consistent with previous social integration research, Study One found
support for a negative association between in-degree centrality and depressive symptoms.
Those with low peer status in the classroom social network also reported higher depressive
symptoms, which is again consistent with past research examining under-integration (Brendgen,
Vitaro, & Bukowski, 2000; Falci & McNeely, 2009). High social status, on the other hand, was not
associated with depressive symptoms, either as a protective factor as hypothesized, or as a risk
factor. This finding at the individual level indicates that while being on the periphery of a
classroom social network and only marginally connected to others within a school class puts
adolescents at an increased risk for depression outcomes, the opposite does not hold true.
More specifically, being accepted and well liked by the classroom network does not, generally
speaking, protect an adolescent from developing depressive symptoms or depression.
Therefore, being on the low end of the social spectrum appears to be more important for
psychological well-being (in a negative manner) than being on the high end.
Another contribution this work makes regarding the peer relationships of adolescents was
to examine peer nominations for both friends and those who are most liked in the class. The
first study, focusing on individual-level factors, clearly noted that these two measures of peer
relationships and status, while related to each other, are distinct constructs and behave in
unique ways regarding the outcomes. Status based on friend nominations was more significantly
associated with depressive symptoms, indicating that close friendships are perhaps a more
important indicator for depression-related outcomes for Chinese adolescents. However, the
popular categorization of “most liked” status was important in regards to professional school
girls’ depression. This finding suggests that while lower friend status is a risk factor, as expected,
it may be that higher “most liked” status is also a risk factor for depression for these girls.

97
Peer groups also contribute to depression and depressive symptoms, which are not solely a
result of individual experiences, cognitions, and choices but can be an attribute of groups and
networks. The results of Study Two supports previous work examining peer contagion effects on
internalizing symptoms. Peers’ symptoms were associated with depressive symptoms among
adolescents. After controlling for individual-level social cognitions and social network status,
self-reported friends’ depressive symptoms were positively associated with depressive
symptoms for Chinese adolescents, with the exception of academic school boys. Those an
adolescent nominated as “most liked,” often considered sociometric popularity, also had some
effect on depressive symptoms for professional school students, particularly boys. Friends’
social status was negatively associated with depressive symptoms for professional school boys.
It indicates that those who are accepted within their classrooms and who have friends who are
accepted experience less internalized distress. Those who are less accepted and are part of a
peer group that is less accepted overall report more depressive symptoms. These are often the
individuals and peer groups that are considered “at risk” in much of the adolescent health
behavior research for outcomes such as smoking, antisocial behavior, aggressive behaviors.
Chinese youth experience unique social environments that, in some regards, are quite different
from Western youth. However, as Chapter Three reports, peer contagion effects on depression-
related outcomes are present for Chinese adolescents much as they are for Western samples of
adolescents.
Finally, the classroom level social network results presented in Chapter Four indicate that
social network structure and characteristics are associated with depression outcomes for
professional school students in China, but are not present for academic school students. One
explanation could be that depression and depressive symptoms include a more social causal
mechanism for professional school students – that is, social factors are more important for the

98
mental well-being of those in professional schools versus those in academic schools. Social
status distribution within a classroom is related to classroom level depressive symptoms for
professional school students. It appears that the proportion of lower status network members is
more important and serves as a risk factor for depression. These results indicate that social
status in the classroom is important for adolescent mental well-being and that it is important to
consider the entire social network composition within the classroom.
The analyses indicated both expected and unexpected differences regarding gender
differences in the influence of social relationships on depression. Based on the robust evidence
regarding gender differences in adolescent depression, the findings that girls reported higher
overall depressive symptoms in the studies is not surprising. One might also suspect, based on
evidence that girls develop more quickly emotionally than boys and value their social
relationships and interactions more highly (Rose & Rudolph, 2006; Stapley & Haviland, 1989),
that girls would report a greater association between social network measures and depression.
Results support the existence of gender-specific conceptions of peer status and social
integration since aspects of status are important to the girls in the sample, particularly in
professional schools, whereas the boys appear less concerned about popularity and friendships
and the associations between these social network measures and depression are either absent
or are weaker in magnitude.
School type differences were also found in each of the three studies. While interaction
effects by school type were not found in the studies, when the analyses were done separately
for each school type, it is clear there are differences. It may be that the interaction effects are
more complex than can be determined in these analyses, such as three-way interactions
including gender, school type, and social network measures. Nevertheless, there are clear
differences between the two school types in the relationship between social network status and

99
depressive symptoms. The associations are stronger for professional school students in all
studies at all three levels of contextual analysis.

Limitations
Although this dissertation benefits from the use of advanced statistical methodology, such
as multilevel models and social network analysis, limitations specific to each type of analysis
must be addressed. Specific limitations for each statistical model are discussed in their
respective chapters; however, a common issue for all of the analyses concerns the relevance of
the social network measures for the adolescents involved in the study. First, past research
examining the influence of social network members (Fowler & Christakis, 2008) has found that
the strength of an effect depends on the strength of the tie (e.g. the strength or quality of a
friendship). These studies did not include measures of the quality or strength of a reported
relationship; therefore, an assumption (and potential limitation) that must be made is that all
friends and most liked nominations are equally influential or that the group-level influence is
more important than one individual’s influence on another. It may also be the case that the
strength of any association is stronger for relationships that are reciprocated (e.g. a reciprocal
friendship occurs when both people nominate the other as a friend). It may be that the quality
of social network ties can weight the relationship between these measures and depressive
symptoms. For example, it may not matter if an individual has few friends if the quality of those
friendships is very high, where friends provide maximum support, feelings of closeness, and
understanding.
A second limitation regarding the relevance of the social network nomination measures is a
lack of clarity concerning whether the nominated peers are really the peers that exert the most
influence on the health and psychological well-being of these adolescents. Adolescents were

100
asked to nominate their best friends in the class and who they thought was most liked in the
class. The studies make assumptions that the networks created from these types of nominations
are influential for youth. It seems natural that an individual’s close friends would exert the
strongest social influences on health behavior, but that may not be the case. It may be some
other type of designation, such as if they had been asked to nominate those they found to be
the most influential in the class. Further, since this study was conducted in China with a Chinese
sample of adolescents using translated measures, these nomination types may not be culturally
relevant designations. Despite these concerns regarding assumptions, these measures have
been used extensively in past research and correlate well with other measures of influence. The
inclusion of two nomination measures also helps to ease these concerns as they allow some
measure of comparison.
There were also contextual constraints these network analyses needed to work within,
namely the network nomination data had to be bounded due to time and cost concerns.
Therefore, adolescents were not able to nominate as many people as they may have wanted
and were limited to five. They also were asked to only nominate from within their classrooms,
so some of relationships may have been missed, such as peers outside of the classroom and
school environment.
Due to the cross-sectional nature of the data, the statistical models used in the three
studies could not establish causation and only identified associations between social network
influences and depression. The results can suggest avenues for future research, but cannot
explain causal mechanism leading from social network status to depressive outcomes. Further
study with longitudinal data is necessary.
Another limitation concerns effect size; while the associations reported were significant,
the effects were modest. This could potentially limit the meaningfulness of the findings;

101
however some have suggested that small effect sizes still have practical applicability (Rosenthal,
Rosnow, & Rubin, 2000). It is also possible that modest effects were observed because these
studies used measures of actual peer-reported behavior depression rather than measures
assessing targets’ perceptions of peer behavior. It is probable that some of the mechanisms
governing social influences originate from perceptions of peer behavior rather than actual peer
behavior. The stronger effects seen for the measure of perceived social self-efficacy on
depression and depressive symptoms confirm this (see results from Chapters 2 and 3). The
effects observed are congruent with the idea that social network influences are just one of a
number of factors that contribute to adolescent depressive symptoms. These studies were able
to determine that even after controlling for these stronger cognitive-perceptual influences,
social network influences were still associated with depression, though on a more modest scale.
This, in conjunction with fairly conservative analytic strategies, suggests that the results do
represent meaningful processes for adolescents across the social environments sampled for this
study.
Finally, limitations regarding generalizability need to be addressed. The CES-D is a self
report assessment of depressive symptoms, rather than a clinical assessment of syndrome or
disorder, therefore the findings of these studies does not generalize to populations who have
diagnosable disorders and it is not possible to determine whether the suggested social network
effects are important mechanisms operating in more clinical populations. Generalizability is also
limited to Chinese adolescents attending either professional or academic high school in a large
urban environment.

102
Implications and Future Directions
The results of these studies contribute to a better understanding of how peer relations are
related to adolescent depressive outcomes. The ultimate goal of such work is to glean useful
constructs upon which prevention interventions can operate. These studies also further refined
and elucidated methods that could be used to test potential intervention program mediators
other than depression-related outcomes, such as aggression, sensation seeking, and
inattentiveness; which have all been linked to health behavior outcomes (Barman, Pulkkinen,
Kaprio, & Rose, 2004; Greene et al., 2000; Loeber, 1997). Some next steps could be to test these
models with these other outcomes and expand the models to examine how health behaviors
such as substance use, risky sexual behavior, and disordered eating are associated with social
network influences and depression. The primary motivation behind proposing these studies was
to inform future prevention work with adolescents that include a greater understanding of how
their social world operates and impacts health behavior in order to tap into these powerful
social influences. School-based intervention work almost universally aims to change or
somehow address social norms, but to date there are few programs that do a good job of even
operationalizing what these norms are, let alone effectively intervening on them. Traditional
self-report-only measures cannot paint a true picture of the social processes that occur in an
adolescent’s environment, because they rely solely on the individual’s interpretation, which may
not be accurate, particularly when psychological distress and dysfunction may be involved. Also,
by their very nature, social processes require the input of more than a single participant;
therefore, to accurately assess them, it should also be necessary to utilize information gathered
from the other participants in the social interaction.
The findings presented in this dissertation suggest modifications for improvement of the
adapted main effects model presented in Chapter one (see Figure 1). As all three studies

103
discovered, social relationships and integration, as measured by social network metrics, can be
conceptualized in different ways and assessed in relation to an institution (such as a school),
social network, or another individual. The findings from this dissertation suggest boys and girls
and those in different school types have different conceptions of social worth and influence.
They may be oriented to different contexts and types of social networks (e.g. friend versus most
liked networks); therefore, specific conceptualizations of social network influences should be
carefully formulated and assessed in relation to the particular contexts that are relevant and
meaningful to the adolescent.
Another contribution of this work is a greater understanding of the limitation that much of
what has been learned about social network influences on depression and the gender
differences in this association is from studies primarily using adult and Western culture samples.
Considering that Chinese adolescents, even at the ages of 15 and 16, still maintain mostly
gender-segregated friendship networks, they may not yet realize the full effects of gender
socialization. This could be a contributing factor in the findings from these studies and suggests
further exploration of these same associations as the gender socialization process continues
across adolescent development.
This work helps to further understand the complicated social environments of adolescents.
It further clarifies the association between social network influences and depression during
adolescence. However, in the course of examining the research aims of this dissertation, further
questions were revealed. These additional questions serve as worthy areas of future research.
For example, these results demonstrate the need to further investigate the relationship
between school types and gender and depression during adolescence for Chinese youth. Future
research is necessary to determine why academic school girls have higher depressive symptoms
and why depression is more socially driven for professional school girls.

104
Another area of future research could be examining how isolation in more than one social
environment is related to depression outcomes. For example, if an adolescent is on the social
fringes of their peer groups at school and also does not find support at home in their family
environment - how do the social influences from both of these different realms interact to
impact depression outcomes. Is the risk of depression associated with social isolation in more
than one social environment additive or is social isolation in multiple realms multiplicative or
even exponential? These are interesting questions for future research that could help to inform
comprehensive prevention work including school and family components.
A third future direction for further investigation includes examining the idea of thresholds.
One question that arises from the associations found in these studies between measures such as
in-degree centrality, or the number of friend nominations an individual receives, and depression
is whether there is some threshold for social contact or number of friends or exposure to friends
who are depressed beyond which an adolescent is at risk for compromised psychological well-
being. Social network analysis has been closely linked to the theory of diffusion of innovations
(Valente, 1995; 1996), which would be the model by which these questions can be investigated.
Another area of future research concerns cross cultural comparisons and the unique
implications for mental health among Chinese youth. A comparison between these findings and
similar studies with adolescents from different cultures, such as comparison to a sample of
Western adolescents from the United States, would be beneficial and informative to the field.
Are these associations sensitive to cultural differences, and if so, how much of a difference does
different cultural context make? Comparing traditionally individualistic and collectivistic cultures
could answer some questions from the field of cross-cultural psychology, such as whether social
relationships are more influential for those living in collectivistic cultures. Another cultural
consideration that could be examined over time within this or other Chinese samples is whether

105
the rapid social and cultural changes occurring in China over the past several decades has any
influence on these relationships. Rates of depression have increased in China over the past
several years (Kleinman, 2004) and the suicide rates have also risen dramatically (Daly, 2009),
particularly among young people. Teachers involved with this study expressed interest in
learning more about mental health and how to better help their students’ psychological
adjustment and were very concerned about depression and suicide. This increased awareness of
the importance of psychological health and well-being is a fairly recent development in the
Chinese educational system, and work such as that presented in this dissertation could help
inform future policies, prevention programs, and training for Chinese teachers.
Another context that bears further study and consideration is the different school types.
There are clear differences between the two school types in the relationship between social
network status and depressive symptoms. The associations are stronger for professional school
students. The Chinese school system is set up differently than Western school systems, with
academic school students required to spend many more hours per week studying for college
entrance exams. They also have more pressure put on them to do well academically by their
parents and teachers and more competition from their classmates. They have little free time to
socialize, which may be one reason for the differences noted in these studies. Professional
school students have more free time and fewer academic pressures. While academic school
students, particularly the girls, report the highest levels of depressive symptoms, they do not
appear to be that closely associated with what occurs in their social environment. Dealing with
school and family stress as well as gender role expectations could be a larger factor in their
depression outcomes; however, further study would be required to examine this.
Finally, future prevention of depressive symptoms in adolescents will need to be driven by
theory and firmly grounded in empirical research. The studies presented in this dissertation

106
provide empirical support for promising theoretical constructs that can be used to develop
successful health promotion and prevention programs. As the theoretical evidence suggests,
although depression is generally conceptualized as an individual disorder with a strong focus on
problematic thoughts and perceptions, maladaptive interpersonal functioning also contributes
to the development and progression of depression and depressive symptoms (Brown & Lohr,
1987; Coyne, 1976 ; Lewinsohn, 1974; Cole,1990). Preventing distress and dysfunction in the
social networks of adolescent’s at risk for depression, as well as improving interpersonal
competencies, should be a key aim of future prevention work focusing on this developmental
period. It is equally important that promotion efforts are designed from a developmental
perspective that accounts for the unique experiences and influences of adolescence in order to
promote healthy growth and development. Interventions that have been developed to address
co-occurring health behaviors, such as substance abuse and anxiety, and for which significant
associations with social network influences have also been found (Ennett, et al., 2006; Hirsch &
Reischl, 1985; La Greca & Harrison, 2005), could influence depression-related outcomes as well.
However, past interventions targeting single problems have typically not measured multiple
outcomes. And, despite the promise that broad-based preventive intervention could result in
reduction in the incidence of depression, no controlled evaluations of such programs have been
published (NRCIOM, 2009). Interventions that are single problem-focused are most likely limited
in their potential effects. Comprehensive and integrated programs focusing on multiple levels of
adolescent functioning and using both cognitive and interpersonal approaches that address
multiple, co-occurring and interrelated outcomes are more likely to be effective in addressing
adolescent depression. Therefore, future prevention work could test the effects of interventions
that have primarily been developed to address related, co-occurring behaviors, on depression
and depressive symptoms among adolescents. This dissertation highlights the importance of

107
attending in greater detail to the social context in which adolescent depressive symptoms
develop and the potential impact integrating peer-oriented intervention strategies has for
future prevention and treatment efforts.

Conclusions
Once in school, children and adolescents spend a great deal of time interacting with peers in
different ways, so it is not surprising that important markers of healthy adjustment and
adaptation during adolescents include development, maintenance, and participation in healthy
relationships with peers.
The potential of peer relations to influence the etiology, prevention, and treatment of
depression and depressive symptoms stems from their significance in development (Hartup,
1996). Interventions can attempt to harness the power of positive peer influence, and
compensate for negative influences, in preventive interventions with a variety of strategies.
Interventions have engaged peers as counselors, coaches, and role models in efforts to change
risk or behavior in individuals and groups (Pearlman, Camberg, Wallace, Symons, & Finison,
2002; Valente, Hoffman, Ritt-Olson, Lichtman, & Johnson, 2003; Valente, 2006). Intervention
researchers have also experienced tough lessons regarding the power of peer influences to
undermine the good intentions of peer-based interventions under certain conditions (Dishion,
McCord, & Poulin, 1999). This work can hopefully help to further inform such work in order to
maximize program effectiveness.
Social network influences were identified as significant factors, both risk and protective, for
adolescent depression and depressive symptoms. General results indicate that individuals who
are better integrated into their social networks, who have fewer depressed friends, and whose
classroom social networks contain fewer peripheral members are less likely to report depressive

108
symptoms and depression. This work represents further steps in the investigation of the
complex social environments in which adolescents must navigate and which are so meaningful
for healthy development.

109
BIBLIOGRAPHY
Aseltine, RH (1995). A reconsideration of parental and peer influences on adolescent deviance.
Journal of Health and Social Behavior, 36, 103-121.
Aseltine, RH, Gore, S, & Colton, ME (1994). Depression and the social developmental context of
adolescence. Journal of Personality and Social Psychology, 67, 252–263.
Barman, SK, Pulkkinen, L, Kaprio, J, & Rose, RJ (2004). Inattentiveness, parental smoking and
adolescent smoking initiation. Addiction, 99(8): 1049-1061.
Barrera, M, & Garrison-Jones, C (1992). Family and peer social support as specific correlates of
adolescent depressive symptoms. Journal of Abnormal Child Psychology, 20(1), 1-16.
Berkman, LF (1995). The role of social relations in health promotion. Psychosomatic Medicine.
57:245-254.
Berkman, LF, Glass, T, Brissette, I, Seeman, TE (2000). From social integration to health:
Durkheim in the new millennium. Social Science & Medicine; 51: 843-857.
Berndt, TJ (1979). Developmental changes in conformity to peers and parents. Developmental
Psychology, 15(6): 608-616.
Birmaher, B, Ryan, ND; Williamson, DE, Brent, DA (1996a). Childhood and adolescent
depression: A review of the past 10 years, Part I. Journal of the American Academy of Child &
Adolescent Psychiatry, 35(12), 1427-1439.
Birmaher, B, Ryan, ND; Williamson, DE, Brent, DA (1996b). Childhood and adolescent
depression: A review of the past 10 years, Part II. Journal of the American Academy of Child &
Adolescent Psychiatry, 35(12), 1575-1583.
Boivin, Coie, Dodge, KA (1995). Individual-group behavioral similarity and peer status in
experimental play groups of boys: the social misfit revisited. The Journal of Personality and
Social Psychology, 69(2): 269-279.
Boivin, M, Hymel, S, & Bukowski, WM (1995). The roles of social withdrawal, peer rejection, and
victimization by peers in predicting loneliness and depressed mood in childhood.
Development and Psychopathology, 7, 765–785.
Bolton W, & Oakley, K (1987). A longitudinal study of social support and depression in
unemployed men. Psychological Medicine, 17: 453-460.
Borgatti, SP, Everett, MG, and Freeman, LC (1999). UCINET 6.0 Version 6.232. Natick: Analytic
Technologies.
Bowlby, J (1951). Maternal Care and Mental Health. New York: Schocken.

110
Bowlby, J (1980). Loss: Sadness & Depression. Attachment and Loss (vol. 3); International
psychoanalytical library, no. 109). London: Hogarth Press
Bowlby, J (1988). A Secure Base: Parent-Child Attachment and Healthy Human Development.
Tavistock professional book. London: Routledge.
Brass, DJ, Galaskiewicz, J, Greve, HR, Tsai, W (2004). Taking stock of networks and organizations:
A multilevel perspective. Academy of Management Journal, 47(6), 795-817.
Brendgen, M, Vitaro, F, Bukowski, WM (2000). Deviant friends and early adolescents’ emotional
and behavioral adjustment. Journal of Research on Adolescence. 10(2):173–189.
Bronfenbrenner, U (1979). The Ecology of Human Development: Experiments by Nature and
Design. Cambridge, MA: Harvard University Press.
Bronfenbrenner, U (1999). Environments in developmental perspective: Theoretical and
operational models. In SL Friedman & TD Wachs (Eds.), Measuring environment across the life
span: Emerging methods and concepts (pp. 3-29). Washington, DC: American Psychological
Association.
Brown, BB (1990). Peer groups and peer cultures. In Feldman, SS & Elliott, GR (Eds.). At the
threshold: The developing adolescent (pp. 171-196). Harvard University Press, Cambridge,
MA.
Brown, BB, & Lohr, MJ (1987). Peer group affiliation and adolescent self-esteem: An integration
of ego- identity and symbolic interaction theories. Journal of Personality and Social
Psychology, 52, 47-55.
Brown, BB, Clasen, DR, & Eicher, SA (1986). Perceptions of peer pressure, peer conformity
dispositions, and self-reported behavior among adolescents. Developmental Psychology, 22:
521-530.
Buhrmester, D (1996). Need fulfillment, interpersonal competence, and the developmental
contexts of early adolescent friendship. In: Bukowski, WM, Newcomb, AF, and Hartup, WW
(Eds.). The company they keep: Friendship in childhood and adolescence (pp. 158–185).
Cambridge University Press: Cambridge.
Burke, KC, Burke, JD, Rae, DS, & Regier, DA (1991). Comparing age of onset of major depression
and other psychiatric disorders by birth cohorts in five U.S. community populations. Archives
of General Psychiatry, 48, 789–795.
Burt, R (1980). Autonomy in a social topology. American Journal of Sociology, 85: 892-925.
Cairns, RB, Cairns, BD, Neckerman, HJ, Gest, SD, & Gariepy, JL (1988). Social networks and
aggressive behavior: Peer support or peer rejection? Developmental Psychology, 24, 815-823

111
Cairns, RB, Leung, MC, Buchanan, L, Cairns, BD (1995). Friendships and social networks in
childhood and adolescence: fluidity, reliability, and interrelations. Child Development, 66(5):
1330-1345.
Caldwell, MS, Rudolph, KD, Troop-Gordon, W, and Kim, D (2004). Reciprocal influences among
relational self views, social disengagement, and peer stress during adolescence. Child
Development, 75(4): 1140-1154.
Carley, KM (1991). A theory of group stability. American Sociological Review, 56(3): 331-354.
Cassel, J (1976). The contribution of the social environment to host resistance. American Journal
of Epidemiology, 104: 107-123.
Cauce, AM (1986). Social networks and social competence: Exploring the effects of early
adolescent friendships. American Journal of Community Psychology, 14(6), 607-628.
Cerel, J, Roberts, TA, Nilsen, WJ (2005). Peer suicidal behavior and adolescent risk behavior. The
Journal of Nervous and Mental Disease, 193(4), 237-243.
Chan, A, & Poulin, F (2009). Monthly instability in early adolescent friendship networks and
depressive symptoms. Social Development, 18(1), 1-23.
Chang, L (2004). The role of classroom norms in contextualizing the relations of children’s social
behaviors to peer acceptance. Developmental Psychology, 40(5), 691-702.
Chassin, L, Presson, CC, Sherman, SJ, Montello, D, McGrew, J (1986). Changes in peer and parent
influence during adolescence: Longitudinal versus cross-sectional perspectives on smoking
initiation. Developmental Psychology, 22(3), 327-344.
Cheung, CK, & Bagley, C (1998). Validating an American scale in Hong Kong: the Center for
Epidemiological Studies Depression Scale (CES-D). Journal of Psychology, 132(2), 169-186.
Cillessen, AHN (2007). New perspectives on social networks in the study of peer relations. New
Directions for Child and Adolescent Development, 118, 91-100.
Coates, D (1985). Relationships between self-concept measures and social network
characteristics for black adolescents. Journal of Early Adolescence, 5, 319–338.
Cohen, GL, Prinstein, MJ (2006). Peer contagion of aggression and health risk behavior among
adolescent males: An experimental investigation of effects on public conduct and private
attitudes. Child Development, 77 (4), 967 – 983.
Cohen, J (1977). Sources of peer group homogeneity. Sociology of Education 50: 227-41.
Cohen, S (2004). Social relationships and health. American Psychologist, 59(8): 676-684.
Cohen, S, & Wills, T.A (1985). Stress, social support, and the buffering hypothesis. Psychological
Bulletin, 98, 310–357.

112
Cohen, S, Gottlieb, BH, & Underwood, LG (2000). Social relationships and health. Social support
measurement and intervention: A guide for health and social scientists. (pp. 3-25). New York,
NY.
Coie, J, Terry, R, Lenox, K, Lochman, J, & Hyman, C (1995). Childhood peer rejection and
aggression as predictors of stable patterns of adolescent disorder. Development and
Psychopathology, 7, 697–713.
Cole, DA (1990). Relation of social and academic competence to depressive symptoms in
childhood. Journal of Abnormal Psychology, 99, 429.
Costello, EJ, Erkanli, A, Angold, A (2006). Is there an epidemic of child or adolescent depression?
Journal of Child Psychology and Psychiatry, 47(12), 1263-1271.
Coyne, JC (1976). Toward an interpersonal description of depression. Psychiatry, 38, 28-40.
Coyne, JC (1999). Thinking interactionally about depression: A radical restatement. In Joiner, T,
Coyne, JC (Eds.). Advances in Interpersonal Approaches: The Interactional Nature of
Depression (pp. 365-392). American Psychological Association: Washington, DC.
Crosnoe, R, McNeely, C (2008). Peer relations, adolescent behavior, and public health research
and practice. Family and Community Health, 31(1), 71-80.
Csikszentmihalyi, M, & Larson, RW (1984). Being adolescent. New York: Basic Books.
Daly, R (2009). Depression biggest contributor to global disease burden. Psychiatric News;
44(1):7.
Davidson, KW, Rieckmann, N, Lesperance, F (2004). Psychological theories of depression:
Potential application for the prevention of acute coronary syndrome recurrence.
Psychosomatic Medicine, 66, 165-173.
Dishion, TJ, McCord, J, & Poulin, F (1999). When interventions harm: Peer groups and problem
behavior. American Psychologist, 54, 755–764.
Ellen, J, Dolcini, M, Bir, ND, Harper, G, Watson, S, & Valente, T (2001). Social network centrality
and sexual activity among a household sample of urban African American adolescents.
Connections , 24 (2), 76-81.
Ennett, ST, Bauman, KE, Hussong, A, Faris, R, Foshee, VA, Cai, L, DuRant, RH (2006). The peer
context of adolescent substance use: Findings from social network analysis. Journal of
Research on Adolescence, 16(2):159-186.
Espelage, DL, Holt, MK, Henkel, RR (2003). Examination of peer-group contextual effects on
aggression during early adolescence. Child Development, 74(1), 205-220.
Falci, C, McNeely, C (2009).Too many friends: social integration, network cohesion and
adolescent depressive symptoms. Social Forces. 87(4): 2031-2062

113
Farmer, TW, & Rodkin, PC (1996). Antisocial and prosocial correlates of classroom social
positions: the social network centrality perspective. Social Development, 5: 174-188.
Field, T, Diego, M, & Sanders, C (2001). Adolescent depression and risk factors. Adolescence,36,
491-498.
Fischer, S (1982). To Dwell Among Friends. Chicago: University of Chicago Press.
Fowler, NA, & Christakis, JH (2008). Dynamic spread of happiness in a large social network:
longitudinal analysis over 20 years in the Framingham Health Study. British Medical Journal,
337: 2338-2347.
Franko, DL, Striegel-Moore, RH, Bean, J (2005). Psychosocial and health consequences of
adolescent depression in black and white young adult women. Health Psychology, 24, 586-93.
Furman W, & Buhrmester, D (1992). Age and sex differences in perceptions of networks of
personal relationships. Child Development, 63(1), 103-115.
Garber, J, & Flynn, C (2001). Predictors of depressive cognitions in young adolescents. Cognitive
Therapy and Research, 25, 353–376.
Gest, SD, Graham-Bermann, SA, & Hartup, WW (2001). Peer experience: common and unique
features of number of friendships, social network centrality, and sociometric status, Social
Development, 10, 23–40.
Gibbs, JP, & Martin, WT (1958). A theory of status integration and its relationship to suicide.
American Sociological Review. 23:14-147.
Glynn, TJ (1981). From family to peer: A review of transitions of influence among drug-using
youth. Journal of Youth and Adolescence, 10(5), 363-383.
Gotlib, IH, Lewinsohn, PM, & Seeley, JR (1995). Symptoms versus a diagnosis of depression:
differences in psychosocial functioning. Journal of consulting and clinical psychology, 63(1),
90-100.
Greene, K, Mrcmar, M, Walters, LH, Rubin, DL, Hale, J, Hale, L (2000). Targeting adolescent risk-
taking behaviors: the contributions of egocentrism and sensation-seeking. Journal of
Adolescence, 23(4): 439-461.
Haarasilta, LM, Marttunen, MJ, Evans, DL, Aro, HM (2004). Correlates of depression in a
representative sample of adolescents and young adults. European Journal of Public Health,
14, 280-285.
Hallfors, DD, Waller, MW, Ford, CA, Halpern, CT, Brodish, PH, Iritani, B (2004). Adolescent
depression and suicide risk: Association with sex and drug behavior. American Journal of
Preventive Medicine, 27(3), 224-231.

114
Hammen, CL (2009). Adolescent depression: Stressful interpersonal contexts and risk for
recurrence. Current Directions in Psychological Science, 18(4), 200-205.
Hammen, CL, & Brennan, PA (2001). Depressed adolescents of depressed and non-depressed
mothers: Tests of an interpersonal impairment hypothesis. Journal of Consulting and Clinical
Psychology, 69(2), 284-294.
Hammer, M (1981). Social support, social networks, and schizophrenia. Schizophrenia Bulletin, 7:
45-57.
Hammer, M (1983). ‘Core’ and ‘extended’ social networks in relation to health and illness. Social
Science and Medicine; 17(7): 405-411.
Hankin, BL & Abramson, LY (2001). Development of gender differences in depression: An
elaborated cognitive vulnerability-transactional stress theory. Psychological Bulletin, 127,
773-796.
Hansell, S (1985). Adolescent friendship networks and distress in school. Social Forces, 63, 698–
715.
Harrington, A (2005). Modern social theory: An introduction. Oxford, UK: Oxford University
Press.
Harrington, NR (1998). Bipolar versus unipolar mood disorders: The role of cognitive processes
and life stress in differential diagnosis, symptom changes, family history, and comorbidity.
Dissertation Abstracts International: Section B: The Sciences and Engineering, 59(3-B), 1367.
Harter, S, Stocker, C, & Robinson, NS (1996). The liabilities of a looking glass self-orientation
among young adolescents. Journal of Research on Adolescence, 6, 285-308.
Hartup, WW (1996). The company they keep: Friendships and their developmental significance.
Child Development, 67(1), 1-13.
Hawker, DSJ, & Boulton, MJ (2000). Twenty years’ research on peer victimization and
psychosocial maladjustment: A meta-analytic review of cross-sectional studies. Journal of
Child Psychology and Psychiatry, 41, 441-455.
Hecht, DB, Inderbitzen, HM, Bukowski, AL (1998). The relationship between peer status and
depressive symptoms in children and adolescents. Journal of Abnormal Child Psychology, 26,
153-160.
Heilbron, N, Prinstein, MJ (2008). Peer influence and adolescent non suicidal self-injury: A
theoretical review of mechanisms and moderators. Applied and Preventive Psychology, 12(4),
69-177.
Hinde, RA (1976). Interactions, relationships and social-structure. Man, 11, 1–17.

115
Hirsch, BJ, Reischl, TM (1985). Social networks and developmental psychopathology: a
comparison of adolescent children of a depressed, arthritic, or normal parent. Journal of
Abnormal Psychology, 94(3): 272-281.
Ho, DYF, & Chiu, CY (1998). Collective representations as a metaconstruct: An analysis based on
methodological relationism. Culture and Psychology, 4(3): 349-369.
Hoffman, BR, Sussman, S, Unger, JB, Valente, TW (2006). Peer influences on adolescent cigarette
smoking: A theoretical review of the literature. Substance Use and Misuse, 41(1), 103-155.
Hogue, A, Steinberg, L (1995). Homophily of internalized distress in adolescent peer groups.
Developmental Psychology, 31(6), 897-906.
Holahan, CK, & Holahan, CJ (1987). Self-efficacy, social support, and depression in aging: A
longitudinal analysis. The Journal of Gerontology, 42(1), 65-68.
House, JS, Landis, KR, Umberson, D (1988). Social relationships and health. Science, 241(4865):
540-545.
Inkeles, A (1959). Personality and Social Structure. In RK Merton, L Broom, & LS Cottrell (Eds.),
Sociological Today (pp. 249-76). New York: Basic Books.
Institute of Medicine (1994). Reducing risks for mental disorders: Frontiers for preventive
intervention research. PJ Mrazek and RJ Haggerty (Eds). Committee on Prevention of Mental
Disorders, Division of Biobehavorial Sciences and Mental Disorders. Washington, DC:
National Academy Press.
Jaccard, J, Dodge, T, & Blanton, H (2005). Peer influences on risk behavior: an analysis of the
effects of a close friend. Developmental Psychology, 41(1): 135-147.
Jacobs, RH, Reinecke, MA, Gollan, JK, Kane, P (2008). Empirical evidence of cognitive
vulnerability for depression among children and adolescents: A cognitive science and
developmental perspective. Clinical Psychology Review; 28(5): 759-782.
Jacobson, JM (2007). The Relationship between social support and depression in adolescents.
Journal of Child and Adolescent Psychiatric Nursing, 4(1), 20-24.
Kadushin, C (1983). Mental health and the interpersonal environment: Reexamination of some
effects of social structure and mental health. American Sociological Review, 48, 188–198.
Kalish, Y, Robins, G (2006). Psychological predispositions and network structure: The
relationship between individual predispositions, structural holes, and network closure. Social
Networks, 28, 56-84.
Kandel, DB (1978). Homophily, selection, and socialization in adolescent friendships. American
Journal of Sociology, 84(2), 427-436.

116
Kandel, DB (1996). The parental and peer contexts of adolescent deviance: An algebra of
interpersonal influences. Journal of Drug Issues, 26, 2, 289–315.
Kawachi, I, Berkman, LF (2001). Social ties and mental health. Journal of Urban Health; 78(3):
458-467.
Keenan, K, Loeber, R, Zhang, Q, Stouthamer-Loeber, M, & Van Kammen, WB (1995), The
influence of deviant peers on the development of boys' disruptive and delinquent behavior: a
temporal analysis. Development & Psychopathology, 7: 715-726.
Keenan-Miller, D, Hammen, CL, Brennan, PA (2007). Health outcomes related to early
adolescent depression. Journal of Adolescent Health, 41, 256-262.
Kiesner, J (2002). Depressive symptoms in early adolescence: Their relations with classroom
problem behavior and peer status. Journal of Research on Adolescence, 12, 463–478.
Kindermann, TA (1996). Strategies for the study of individual development within naturally
existing peer groups. Social Development, 5(2),158-173.
Kistner, J (2006). Children’s peer acceptance, perceived acceptance, and risk for depression. In
Joiner, TE, Brown, JS, & Kistner, J (Eds.) The interpersonal, cognitive, and social nature of
depression (pp. 1-21). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Klein, KJ, Lim, B, Saltz, JL, & Mayer, DM (2004). How do they get there? An examination of the
antecedents of centrality in team networks. Academy of Management Journal, 47, 952-963.
Kleinman, A (2004). Culture and depression. New England Journal of Medicine, 351(10): 951-
952.
Klerman, GL (1988). Depression and related disorders of mood (affective disorders). In AM
Nicholi, Jr., (Ed.) The New Harvard Guide to Psychiatry. Cambridge, MA: Harvard University
Press.
Kobus, K (2003). Peers and adolescent smoking. Addiction, 98(1), 37-55.
Kupersmidt, JB, & Patterson, CJ (1991). Childhood peer rejection, aggression, withdrawal, and
perceived competence as predictors of self-reported behavior problems in preadolescence.
Journal of Abnormal Child Psychology, 19, 427–449.
La Greca, AM & Lopez, N (1998). Social anxiety among adolescents: Linkages with peer relations
and friendships. Journal of Abnormal Child Psychology, 26(2), 83-94.
La Greca, AM & Prinstein, MJ (1999). Peer group. In WK Silverman & TH Ollendick (Eds.),
Developmental issues in the clinical treatment of children and adolescents. New York: Wiley.
La Greca, AM, & Harrison, HM (2005). Adolescent peer relations, friendships, and romantic
relationships: Do they predict social anxiety and depression? Journal of Clinical Child and
Adolescent Psychology, 34, 49–61.

117
Lam , CK, Lim, PPJ, Low, BL, Ng, LL, Chiam, PC, Sahadevan, S (2004). Depression in dementia: a
comparative and validation study of four brief scales in the elderly Chinese. International
Journal of Geriatric Psychiatry, 19(5), 422-428.
Lansford, JE, Capanna, C, Dodge, KA, Caprara, GV, Bates, JE, Pettit, GS, Concetta Pastorelli, C
(2007). Peer social preference and depressive symptoms of children in Italy and the United
States. International Journal Behavioral Development, 31(3), 274–283.
Larson, RW, & Richards, MH (1991). Boredom in the middle school years: Blaming schools versus
blaming students. American Journal of Education, 99, 418-443.
Larson, RW, Richards, MH, Moneta, G, Holmbeck, G, & Duckett, E (1996). Changes in
adolescents’ daily interactions with their families from ages 10 to 18: Disengagement and
transformation. Developmental Psychology, 32, 744-754.
Lewinsohn, PM (1974). A behavioral approach to depression. In RM Friedman, & MM Katz (Eds.),
The psychology of depression: Contemporary theory and research (pp. 132-149). New York:
Wiley.
Li, F, Barrera, M, Hops, H, Fisher, KJ (2002). The longitudinal influence of peers on the
development of alcohol use in late adolescence: A growth mixture analysis. Journal of
Behavioral Medicine, 25(3), 293–315.
Li, X, Stanton, B, Chen, X, Hong, Y, Fang, X, Lin, D, Mao, R, Wang, J (2006). Health indicators and
geographic mobility among young rural-to-urban migrants in China. World Health and
Population, 8(2), 5-21.
Lin, HC, Tang, TC, Yen, JY, Ko, CH, Huang, CF, Liu, SC, Yen, CF (2008). Depression and its
association with self-esteem, family, peer, and school factors in a population of 9586
adolescents in southern Taiwan. Psychiatry and Clinical Neuroscience, 62, 412-420.
Liu, X, & Kaplan, H (1999). Explaining gender differences in symptoms of subjective distress in
young adolescents. Stress Medicine, 15(1): 41-51.
Loeber, R (1997). Key issues in the development of aggression and violence from childhood to
early adulthood. Annual Review of Psychology, 48: 371-410.
Luke DA, & Harris JK (2007). Network analysis in public health: history, methods, and
applications. Annual Review of Public Health, 28: 69-93.
Mahon, NE, Yarcheski, A (2001). Outcomes of depression in early adolescents. Western Journal
of Nursing Research, 23(4), 360-375.
Mehra, A, Kilduff, M, & Brass, DJ (2001). The social networks of high and low self-monitors:
Implications for workplace performance. Administrative Science Quarterly, 46, 121-146.

118
Mercy, JA, Kresnow, MJ, O’Carroll, PW, Lee, RK, Powell, KE, Potter, LB, Swann, AC, Frankowski,
RF, Bayer, TL (2001). Is suicide contagious? A study of the relation between exposure to the
suicidal behavior of others and nearly lethal suicide attempts. American Journal of
Epidemiology, 154(2), 120-127.
Michell, L, & Amos, A (1997). Girls, pecking order and smoking. Social Science and Medicine, 44,
1861–1869.
Miller, G (2006). Mental health in developing countries: China: Healing the metaphorical heart.
Science, 311(5760), 462-563.
Mondimore, FM (1990). Depression: The mood disease. Baltimore, MD: Johns Hopkins University
Press.
Nangle, DW, Erdley, CA, Newman, JE, Mason, CA, Carpenter, EM (2003). Popularity, friendship
quantity, and friendship quality: Interactive influences on children’s psychological
adjustment. Journal of Clinical Child and Adolescent Psychology, 32, 546-555.
National Research Council and Institute of Medicine (NRCIOM). (2009). Preventing Mental,
Emotional, and Behavioral Disorders Among Young People: Progress and Possibilities.
Committee on Prevention of Mental Disorders and Substance Abuse Among Children, Youth
and Young Adults: Research Advances and Promising Interventions. Mary Ellen O’Connell,
Thomas Boat, and Kenneth E. Warner, Editors. Board on Children, Youth, and Families,
Division of Behavioral and Social Sciences and Education. Washington, DC: The National
Academies Press.
Nolen-Hoeksema, S (1990). Sex differences in depression. Stanford, CA: Stanford University
Press.
Nolen-Hoeksema, S (1991). Responses to depression and their effects on the duration of
depressive episodes. Journal of Abnormal Psychology, 100(4): 569-582.
Nolen-Hoeksema, S, & Girgus, JS (1994). The emergence of gender differences in depression
during adolescence. Psychological Bulletin, 115(3), 424-443.
O’Hara, MW, Rehm, LP, Campbell, SB (1983). Postpartum depression: A role for social network
and life stress variables. The Journal of Nervous and Mental Disease, 171(6), 336-341.
Oetting, ER, & Beauvais, F (1986). Peer cluster theory: Drugs and the adolescent. Journal of
Counseling and Development, 65, 17–22.
Okun, MA, & Keith, VM (1998). Effects of positive and negative social exchanges with various
sources on depressive symptoms in younger and older adults. Journal of Gerontology, 53: 4–
20.
Ostberg, V (2003). Children in classrooms: Peer status, status distribution, and mental well-
being. Social Science & Medicine, 56, 17-29.

119
Panak, WF, & Garber, J (1992). Role of aggression, rejection, and attributions in the prediction of
depression in children. Development and Psychopathology, 4, 145–165.
Pearlman, D, Camberg, L, Wallace, L, Symons, P, Finison, L (2002). Tapping youth as agents for
change: evaluation of a peer leadership HIV/AIDS intervention. Journal of Adolescent Health,
31(1): 31-39.
Pearson, V, Phillips, MR, He, F, & Ji, H (2002). Attempted suicide among young rural women in
the People’s Republic of China: Possibilities for prevention. Suicide and Life-Threatening
Behavior, 32(4), 359-369.
Peirce, RS, Frone, MR, Russell, M, Cooper, ML, Mudar, P (2000). A longitudinal model of social
contact, social support, depression, and alcohol use. Health Psychology, 19(1), 28-38.
Petersen, AC, Compas, BE, Brooks-Gunn, J, Stemmler, M, Ey, S, Grant, KE (1993). Depression in
adolescence. American Psychologist, 48(2), 155-168.
Phillips, MR, Liu, H, Zhang, Y (1999). Suicide and social change in China. Culture, Medicine, &
Psychiatry, 23, 25-50.
Pine, DS, Cohen, P, Gurley, D, Brook, J, & Ma, Y (1998). The risk for early-adulthood anxiety and
depressive disorders in adolescents with anxiety and depressive disorders. Archives of
General Psychiatry, 55, 56–64.
Polister, PE (1980). Network analysis and the logic of social support. In RH Price & PE Polister
(Eds.), Evaluation and Action in the Environment. New York: Academic Press.
Possel, P, Baldus, C, Horn, AB, Groen, G, Hautzinger, M (2005). The influence of general self-
efficacy on the effects of a school-based universal primary prevention program of depressive
symptoms in adolescents: A randomized and controlled follow-up study. Journal of Child
Psychology and Psychiatry, 46(9), 982-994.
Prinstein, M J, Meade, C S, & Cohen, G L (2003). Adolescent oral sex, peer popularity, and
perceptions of best friends’ sexual behavior. Journal of Pediatric Psychology, 28, 243–249.
Prinstein, MJ & La Greca, AM (2002). Peer crowd affiliation and internalizing distress in
adolescence: A longitudinal follow-back study. Journal of Research on Adolescence, 12, 325-
351.
Prinstein, MJ (2007). Moderators of peer contagion: A longitudinal examination of depression
socialization between adolescents and their best friends. Journal of Clinical Child and
Adolescent Psychology, 36(2): 159-170.
Prinstein, MJ, & Aikens, JW (2004). Cognitive moderators of the longitudinal association
between peer rejection and adolescent depressive symptoms. Journal of Abnormal Child
Psychology, 32(2), 147-158.

120
Prinstein, MJ, Boergers, J, Spirito, A (2001). Adolescents’ and their friends’ health-risk behavior:
factors that alter or add to peer influence. Journal of Pediatric Psychology, 26(5): 287-298.
Prinstein, MJ, Cheah, CSL, Guyer, AE (2005). Peer victimization, cue interpretation, and
internalizing symptoms: Preliminary concurrent and longitudinal findings for children and
adolescents. Journal of Clinical Child and Adolescent Psychology, 34(1), 11-24.
Prinstein, MJ, Wang, SS (2005). False consensus and adolescent peer contagion: Examining
discrepancies between perceptions and actual reported levels of friends’ deviant and health
risk behaviors. Journal of Abnormal Child Psychology, 33(3), 293-306.
Radloff, L (1977). The CES-D Scale: a self-report depression scale for research in the general
population. Applied Psychological Measurement, 1, 385-401.
Roberts, RE, Lewinsohn, PM, & Seeley, JR (1991). Screening for adolescent depression: A
comparison of depression scales. Journal of the American Academy of Child & Adolescent
Psychiatry, 30(1), 58-66.
Roff, D, & Wirth, RD (1984). Childhood social adjustment, adolescent status, and young adult
mental health. American Journal of Orthopsychiatry, 54, 595–602.
Rook, K (1992). Detrimental aspects of social relationships: Taking stock of an emerging
literature. In HO Veiel & U Baumann (Eds.), The meaning and measurement of social support
(pp. 157–169). New York: Hemisphere.
Rose, A, & Rudolph, KD (2006). A review of sex differences in peer relationship processes:
Potential trade-offs for the emotional and behavioral development of girls and boys.
Psychological Bulletin, 132: 98-131.
Rosenberg, M & McCullough, BC (1981). Mattering: Inferred significance and mental health
among adolescents. Research in Community and Mental Health, 2, 163-182.
Rosenblatt, A, & Greenberg, J (1991). Examining the world of the depressed: Do depressed
people prefer others who are depressed? Journal of Personality and Social Psychology, 60:
620–629.
Rosenthal, R, Rosnow, RL, Rubin, DB (2000). Contrasts and effect sizes in behavioral research: A
correlational approach. Cambridge, UK: Cambridge University Press.
Rudolph, KD, & Hammen, C (1999). Age and gender as determinants of stress exposure,
generation, and reactions in youngsters: A transactional perspective. Child Development, 70,
660-677.
Sastry, J, & Ross, CE (1998). Asian ethnicity and the sense of personal control. Social Psychology
Quarterly, 61: 101-120.
Schraedley, PK, Gotlib, IH, & Hayward, C (1999). Gender differences in correlates of depressive
symptoms in adolescence. Journal of Adolescent Health, 25: 98-108.

121
Seeman, TE (1996). Social ties and health: The benefits of social integration. Annals of
Epidemiology.6: 442-451.
Seeman, TE, Kaplan, GA, Knudsen, L, Cohen, R, Guralnik, J (1987). Social network ties and
mortality among tile elderly in the Alameda County study. American Journal of Epidemiology,
126(4), 714-723.
Sih, A, Hanser, SF, McHugh, KA (2009). Social network theory: New insights and issues for
behavioral ecologists. Behavioral Ecology and Sociobiology, 63, 975-988.
Smith, CJ (1998). Modernization and health care in contemporary China. Health Place, 4(2), 125–
139.
Smith, HM, & Betz, NE (2000). Development and validation of a scale of perceived social self-
efficacy. Journal of Career Assessment, 8, 283-301.
Stapley, JC, & Jeannette M. Haviland, JM (1989). Beyond depression: Gender differences in
normal adolescents' emotional experiences. Sex Roles, 20(5-6): 295-308.
StataCorp (2009). Stata Statistical Software: Release 11. College Station, TX: StataCorp LP.
Stevens, EA, Prinstein, MJ (2005). Peer contagion of depressogenic attributional styles among
adolescents: A longitudinal study. Journal of Abnormal Child Psychology, 33(1), 25-37.
Stice, E, Ragan, J, Randall, P (2004). Prospective relations between social support and
depression: Differential direction of effects for parent and peer support? Journal of Abnormal
Psychology, 113(1), 155-159.
Stokes, JP (1983). Predicting satisfaction with social support from social network structure.
American Journal of Community Psychology, 11, 141-152.
Stormshak, EA, Bierman, KL, Coie, JD, Dodge, K, Greenberg, MK, Lochman, JE., et al. (1998). The
implications of different developmental patterns of disruptive behavior problems for school
adjustment. Development and Psychopathology, 10(3), 451-467.
Swap, RL, & Rubin, JZ (1983). US-China trade negotiations: practices, procedures, and outcomes.
Journal of International Business Studies, 13: 25-38.
Tafarodi, RW, & Smith, AJ (2001). Individualism-collectivism and depressive sensitivity to life
events: The case of Malaysian sojourners. International Journal of Intercultural Relations, 25:
73-88.
Thoits, PA (1983). Multiple identities and psychological well-being: A reformation and test of the
social isolation hypothesis. American Sociological Review, 48, 174–187.
Ueno, K (2005). The effects of friendship networks on adolescent depressive symptoms. Social
Science Research, 34, 484-510.

122
United Nations Population Division (UNPD; 2008). World Urbanization Prospects: The 2008
Revision. New York: United Nations.
Urberg, KA, Degirmencioglu, SM, & Pilgrim, C (1997). Close friend and group influence on
adolescent cigarette smoking and alcohol use. Developmental Psychology, 33, 834-844.
Valente, TW (1995). Network models of the diffusion of innovations. Cresskill, N.J.: Hampton.
Valente, TW (1996). Social network thresholds in the diffusion of innovations. Social Networks,
18(1):69-89.
Valente, TW (2006). Opinion leader interventions in social networks. BMJ, 333: 1082-1083.
Valente, TW, Gallaher, P, & Mouttapa, M (2004). Social network analysis for understanding
substance abuse: A transdisciplinary perspective. Substance Use & Misuse, 39, 1685-1712.
Valente, TW, Hoffman, BR, Ritt-Olson, AR, Lichtman, K, Johnson, CA (2003). Effects of a social-
network method for group assignment strategies on peer-led tobacco prevention programs
in schools. American Journal of Public Health, 93(11): 1837-1843.
Valente, TW, Unger, J, & Johnson, CA (2005). Do popular students smoke? The association
between popularity and smoking among middle school students. Journal of Adolescent
Health, 37: 323-329.
Vega, WA, Kolody, B, Weir, J (1991). Social networks, social support, and their relationship to
depression among immigrant Mexican women. Human Organization, 50(2), 154-162.
Wade, TD, & Kendler, KS (2000). The relationship between social support and major depression:
Cross-sectional, longitudinal, and genetic perspectives. Journal of Nervous and Mental
Disease, 188(5), 251-258.
Wasserman, D, Cheng, Q, Jiang, GX (2005). Global suicide rates among young people aged 15-19.
World Psychiatry, 4(2), 114-120.
Wills, T A (1985). Supportive functions of interpersonal relationships. In S Cohen & SL Syme
(Eds.), Social support and health (pp. 61–82). New York: Academic Press.
World Health Organization, 2008. The global burden of disease: 2004 update. Geneva,
Switzerland: WHO Press. Retrieved August 2, 2009, from
http://www.globalburden.org/GBD_Study_Operations_Manual_Jan_31_2008.pdf.
Wright, JC, Giammarino, M, & Parad, HW (1986). Social status in small groups: Individual-group
similarity and the social “misfit.” Journal of Personality and Social Psychology, 50, 523-536.
Xie, H, Cairns, RB, & Cairns, BD (2002). The development of social aggression and physical
aggression: A narrative analysis of interpersonal conflicts. Aggressive Behavior, 28: 341-355.

123
Yang HJ, Soong, WT, Kuo, PH, Chang, HL, Chen, WJ (2004). Using the CES-D in a two-phase
survey for depressive disorders among non referred adolescents in Taipei: A stratum-specific
likelihood ratio analysis. Journal of Affective Disorders, 82, 419-430.Lin et al., 2008.
Yip, PSF (2001). An epidemiological profile of suicides in Beijing, China. Suicide and Life-
Threatening Behavior, 31(1), 62–70.
Zahn-Waxler, C, Klimes-Dougan, B, & Slattery, M (2000). Internalizing disorders of childhood and
adolescence: Progress and prospects for advances in understanding anxiety and depression.
Development and Psychopathology, 12, 443-466.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The role of depression symptoms on social information processing and tobacco use among adolescents

PDF

Contextualizing social network influences on substance use among high risk adolescents

PDF

A network analysis of online and offline social influence processes in relation to adolescent smoking and alcohol use

PDF

Adolescent social networks, smoking, and loneliness

PDF

Using observed peer discussions to understand adolescent depressive symptoms and interpersonal interactions

PDF

Heterogeneity among adolescents with elevated depressive symptoms: distinct patterns of social and academic attributes

PDF

Social selection and influence of alcohol & marijuana implicit cognitions and behaviors: a longitudinal investigation of peer social network dynamics

PDF

Anxiety symptoms and nicotine use among adolescents and young adults

PDF

Investigating factors that influence peer relationships and obesity during middle childhood

PDF

Untangling the developmental relations between depression and externalizing behavior among maltreated adolescents

PDF

Exploring the role of peer influence, linguistic acculturation, and social networks in substance use

PDF

The effects of mindfulness on adolescent cigarette smoking: Measurement, mechanisms, and theory

PDF

The link between maternal depression and adolescent daughters' risk behavior: the mediating and moderating role of family

PDF

Cultural risk and protective factors for tobacco use behaviors and depressive symptoms among American Indian adolescents in California

PDF

Sociocultural stress, coping and substance use among Hispanic/Latino adolescents

PDF

The formation and influence of online health social networks on social support, self-tracking behavior and weight loss outcomes

PDF

Influences of specific environmental domains on childhood obesity and related behaviors

PDF

Evaluating social networks and impact of micro-influencers who promote e-cigarettes on social media

PDF

Social self-control and adolescent substance use

PDF

Friendship network position on adolescent behaviors: an examination of a broker position and the likelihood of alcohol and cigarette use

Asset Metadata

Creator Okamoto, Janet (author)

Core Title Social network influences on depressive symptoms among Chinese adolescents

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Preventive Medicine (Health Behavior)

Publication Date 06/01/2010

Defense Date 04/29/2010

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Adolescent,China,depressive symptoms,OAI-PMH Harvest,social networks

Place Name China (countries)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Valente, Thomas W. (committee chair), Leventhal, Adam (committee member), Milam, Joel (committee member), Pentz, Maryann (committee member), Schwartz, David (committee member)

Creator Email jokamoto@usc.edu,jokamoto4646@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3106

Unique identifier UC163392

Identifier etd-Okamoto-3780 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-339311 (legacy record id),usctheses-m3106 (legacy record id)

Legacy Identifier etd-Okamoto-3780.pdf

Dmrecord 339311

Document Type Dissertation

Rights Okamoto, Janet

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu