University of Southern California Dissertations and Theses

The role of school climate in the mental health and victimization of students in military-connected schools

(USC Thesis Other)

The role of school climate in the mental health and victimization of students in military-connected schools

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

THE ROLE OF SCHOOL CLIMATE IN THE MENTAL HEALTH AND VICTIMIZATION
OF STUDENTS IN MILITARY-CONNECTED SCHOOLS

by

Kris Matthew Tunac De Pedro

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
(EDUCATION)

December 2012

Copyright 2012 Kris Matthew Tunac De Pedro

ii

Dedication

First, I dedicate this dissertation to my parents, Maximino and Mayda De Pedro. They
gave me life, love, and support. Throughout my academic and professional career, they provided
me with the encouragement to persist even though the most difficult challenges. Without them, I
would never be where I am today. I pursue great things in my career and life because they inspire
me to dream big.
I dedicate this dissertation to my family—cousins Camille and Melissa, Auntie Erna, and
Auntie Nelia—for giving me the emotional support and courage to move forward in graduate
school. I am so fortunate to have the support of a close-knit family who love and care about me
so much! I also dedicate this dissertation to my four best friends in the world, April, Grace,
Karolina, and Joel. Few people in the world have lifelong friends, who are incredibly loyal and
dedicated. I thank them for their emotional support and life lessons.
This dissertation is also dedicated to Aaron, the love of my life. I am grateful for his love
and support in the final stages of my dissertation. Everyday, he teaches me about what it means
to have strong character—integrity, trust, honesty, passion, and love. It is these qualities that will
help me fulfill my truest potential. I am excited for our life journey together.
I would also like to dedicate this dissertation to my Ph.D. colleagues past and present—
Misty, Monica, Icela, Jonathan, Connie, Joey, Anthony, Robin, Katie, Tiffany, Cait, Dara, and
Megan. It was so wonderful to be around such intelligent, inspiring, and enthusiastic educators. I
am so grateful for all the opportunities you gave me to talk about the challenges of this PhD
program, for your patience and understanding, and for the all the emotional support, love. Most
importantly, I am grateful for the endless laughter, which was really good for my soul and spirit.

iii
I would like to dedicate this dissertation to all my mentors, Dr. Tamika Gilreath, Dr.
Shafiqa Ahmadi, Dr. Hazel Atuel, Dr. Darnell Cole, and Dr. Julie Cederbaum. I am grateful for
their emotional support throughout my academic journey at USC. They gave me the
encouragement and support throughout my dissertation. They are true examples of what it means
to be a passionate academic truly dedicated to knowing more about the world around us.
Last but not least, this dissertation is dedicated to Dr. Ron Avi Astor. I am forever
indebted to his personal and professional guidance at USC. He gave me amazing opportunities
and the love and emotional support that I needed throughout this process. I am so fortunate to
have had his mentorship and friendship. Every day, he inspires me to change the world.

iv
Table of Contents
Dedication ii

List of Tables v

Abstract vi

Chapter One: Introduction, Rationale, and Overview of the Three Studies 1
Introduction 1
Theoretical Framework 4
Structure of Dissertation 14

Chapter Two: School Climate Perceptions among Students in Military-Connected Schools:
A Comparison of Military and Non-Military Students from the Same Schools 18
Introduction 18
Methods 23
Results 30
Discussion 39
Limitations 42
Implications 43

Chapter Three: The Role of School Climate in the Mental Health of Secondary Students: A
Study of Students in Military-Connected Schools 45
Introduction 45
Methods 52
Results 66
Discussion 90
Implications 96

Chapter Four: Victimization Rates of Secondary Students: The Role of School Climate in
Military-Connected Schools 99
Introduction 99
Methods 108
Results 122
Discussion 134
Implications 140

Chapter Five: Integration and Implementation of Findings from Three Studies
Purpose of the Studies 143
Overall Findings and Links to Research 147

Bibliography 160

v
List
of
Tables

Table 1. Domains, Scales and Items for the Demographic Variables 26
Table 2. Domains, Scales and Items for the School Climate Variables 29

Table 3. Sample Characteristics for Study #1 31
Table 4. Perceptions of Belonging by Military Connection Status 33
Table 5. Perceptions of School Safety by Military Connection Status 34
Table 6. Perceptions of Relationships by Military Connection Status 35
Table 7. Perceptions of Meaningful Participation by Military Connection Status 36
Table 8. Perceptions of Risky Behavior Approval by Military Connection Status 37
Table 9. Perceptions of Respect for Family by Military Connection Status 38
Table 10. Domains, Scales and Items for the Mental Health 56
Table 11. Domains, Scales and Items for the School Climate Variables 59
Table 12. Deployment Item 60
Table 13. Domains, Scales and Items for the Demographic Variables 62
Table 14. Sample Characteristics for Study #2 67
Table 15. Means and Standard Deviations of Mental Health Scales 69
Table 16. Depression, Well-Being, and Suicidal Ideation by Military Connection 69
Table 17. Well-Being by Military Connection Status 71
Table 18. Depression by Military Connection Status 73
Table 19. Correlational Analyses for Study #2 76
Table 20. Logistic Regressions of Mental Health by Demographic and School Climate
Variables among All Students 79

Table 21. Logistic Regressions of Mental Health by Demographics, School Climate and
Deployment 82

vi
Table 22. Logistic Regressions of Well-Being by Demographics and School Climate among
Non-Military Students and Military Students 84

Table 23. Logistic Regressions of Depression by Demographics and School Climate among Non-
Military Students and Military Students 86

Table 24. Logistic Regressions of Suicidal Ideation by Demographics and School Climate among
Non-Military Students and Military Students 88

Table 25. Domains, Scales and Items for the Victimization 112

Table 26. Domains, Scales and Items for the School Climate Variables 115

Table 27. Deployment Item 116

Table 28. Domains, Scales and Items for the Demographic Variables 118
Table 29. Sample Characteristics for Study #3 123
Table 30. Victimization Rates by Military Connection Status 125
Table 31. Correlational Analyses of School Climate, Deployment and Victimization 126
Table 32. Logistic Regressions of Victimization by Demographics and School Climate 128

Table 33. Logistic Regressions of Victimization by Demographics, Deployment, and School
Climate 130

Table 34. Logistic Regressions of Victimization by School Climate 133

vii
Abstract
Research on school climate has found that a supportive school climate promotes positive
social, emotional, psychological and academic outcomes among students, even student
populations experiencing stressors in the family and community context. Studies have shed light
on the stressors (i.e. deployment) and the negative mental health outcomes of children in military
families. Given the significant presence of military students in over 200 civilian public school
districts in the United States, it is surprising that school climate researchers have largely ignored
the role that a supportive school climate plays in the social and emotional outcomes of military
students. This multiple manuscript dissertation utilizes a population sample of secondary
students in eight military-connected school districts to examine the role that a supportive school
climate plays in two key social and emotional outcomes known to impact academic
achievement—mental health and victimization. The first study provides detailed descriptive
analyses of school climate perceptions (belonging, caring relationships, safety, meaningful
participation, respect for student’s family, and risky behavior disapproval) among military and
non-military students. The second study examines associations between multiple components of
school climate and three mental health indicators—well-being, depression, and suicidal
ideation—among all students as well as within military and non-military students. The third
study assesses associations between school climate and victimization among all students as well
as within military and non-military students. Findings from the descriptive analyses indicate that
military students—those with a parent or those with a sibling in the military—consistently have
more negative school climate perceptions than non-military students. In addition, findings from
multivariate analyses reveal that school climate promotes well-being and curbs rates of
depression, suicidal ideation and victimization among all students and within military and non-

viii
military student populations. This was true even accounting for deployment. This means that a
positive school climate can have healing effects that improve mental health and reduce violence
and suicide for both military and nonmilitary students. This dissertation concludes with
implications for future educational reform, practice and policy.

1
Chapter One:
Introduction, Rationale, and Overview of the Three Studies
Introduction
A positive school climate is one where students have caring relationships with peers and
adults, a sense of belonging, feel a high degree of personal safety, participate meaningfully in
school policies and activities, disapprove of risky behaviors among peers, and feel that peers and
school staff are aware of their family issues (Brand, Feldner, Shim, Seitsinger, & Dumas, 2003;
Cohen, McCabe, Michelli, & Pickeral, 2009; Zullig, Koopman, Patton, & Ubbes, 2010). A large
body of literature has found that a positive school climate can promote a wide range of positive
academic, social and emotional, and psychological student outcomes (Comer, 1984; Hoy &
Hunnum, 1997; Jia, Way, Ling, Yoshikawa, Chen, Hughes, Ke & Lu, 2009; Modin & Ostberg,
2009). Several of these studies have found that a nurturing and caring school climate affects two
key outcomes known to impact academic functioning and achievement: mental health and
victimization (Cohen et al., 2009; Zullig et al., 2010). Research suggests that the role that school
climate plays in reducing mental health problems and victimization rates can be generalized to
different student populations in different geographic contexts, even among student populations
experiencing significant psychological strain (i.e. students with war trauma, students in highly
violent neighborhoods) (Comer, 1984; Elliot, Cornell, Gregory, & Fan, 2010; Hanish & Guerra,
2000; Khoury-Kassabri, Benbenishty, Astor, & Zeira, 2004).
Given the long wars in Iraq and Afghanistan since 2002, it is surprising that one
population of students has been largely ignored by school climate researchers: the children of
military service personnel. Military children in the current war context experience significant
psychological strain as a result of the deployment cycle (De Pedro et al., 2011). When parents

2
and/or other family members are deployed to war, military children experience multiple stressors
including parental separation, geographic relocation and school transitions, shifting household
roles and responsibilities, and exposure to the stress of a left-behind parent (Chandra, Martin,
Hawkins, & Richardson, 2010; De Pedro et al., 2011; Flake, Davis, Johnson, & Middleton,
2009). In addition, even when parents or family members return from war and reintegrate into
civilian society, military children continue to encounter stressful experiences related to veteran
war trauma, re-adjusting family dynamics, and repeated, multiple deployments. A large body of
research has found that military-specific stressors adversely impact the psychological and
emotional well-being of military children. For instance, studies have found that when compared
to children in civilian families, military children have more negative mental health outcomes and
experience higher rates of physical and emotional maltreatment and abuse during the current war
context (Chandra, Martin, Hawkins, & Richardson, 2010; Flake et al., 2009; Gorman, Eide, &
Hisle-Gorman, 2010; Huebner, Mancini, Bowen, & Orthner, 2009; Mmari, Roche, Sudhinaraset,
& Blum, 2009).
Despite these outcomes, recent studies have found that supportive and responsive
communities (i.e. military bases, military-impacted communities) and households can help
facilitate the healthy coping of military children and their left-behind parents, experiencing
deployment and other military life events (Amen, Jellen, Merves, & Lee, 1988; Hoshmand &
Hoshmand, 2007; MacDermid, Samper, Schwartz, Nishida, & Nyaronga, 2008). However, this
area of research has ignored the potentially significant role of supportive school environments in
promoting healthy social and emotional outcomes among military children. Military children
have a numerically significant presence in U.S. public schools in civilian communities (Kitmitto
et al., 2011; Yin, Kitmitto, Shkolnik, Hoshen, Hannan, & Arellanes, 2011). According to recent

3
estimates from the Department of Defense Educational Activity (DoDEA), there are an estimated
400,000 military children enrolled in American public schools (Kitmitto et al., 2011; Yin et al.,
2011). In addition, 214 civilian public school districts serve a significant proportion of military
children (about 4% of total student enrollment or 400 students) (Kitmitto et al., 2011; Yin et al.,
2011). Surprisingly, to date, only a few recent qualitative studies have examined the social and
emotional experiences of military students in civilian public schools (Bradshaw, Sudhinaraset,
Mmari, & Blum, 2010; Chandra et al., 2010; Mmari et al., 2009). These investigations have
suggested that civilian schools serving significant concentrations of students from military
families, also known as military-connected schools, are struggling to respond to the mental
health challenges and victimization of military children. These studies, however, had exploratory
objectives and small samples. Future studies are needed to assess the relationship between school
climate and the mental health and school victimization of military children.
Overall, there is a gap in research regarding the role that school climate plays in the
social and emotional outcomes of military children. It is unclear how military children perceive
their school’s climate and if these perceptions are generally more negative or positive than their
civilian peers. Also, the research literature has not yet assessed the role of a supportive school
climate in the mental health outcomes and victimization rates of military children. In addition, no
studies have examined if a positive school climate affects the social and emotional outcomes of
military children, given their experiences surrounding deployment. Further, no studies to date
have utilized population samples to address the preceding research objectives. A population
sample could help educational researchers make valid generalizations about the outcomes of
military students and inform school-based interventions to fit the needs of military students.

4
Given these gaps in school climate research and the research literature on military
children, this multiple manuscript dissertation will accomplish the following aims:
1. The first aim is to compare the school climate perceptions of students by military
connection (non-military students, students with a military parent, and students with a
military sibling) in military-connected schools.
2. The second aim is to examine associations between school climate and mental health
outcomes among military and non-military students in military-connected schools.
Associations between school climate and mental health outcomes, controlling for
deployment will also be examined.
3. The third aim is to examine associations between school climate and victimization
outcomes among military and non-military students in military-connected schools. The
relationship between school climate and victimization, controlling for deployment will
also be examined.
4. The fourth aim is integrate the joint findings and synthesize the implications of the
studies’ findings for school climate research and practice as they relate to the mental
health outcomes and victimization of students in military-connected schools.
Theoretical Framework
The Role of Supportive School Environments in the Outcomes of Military Students
De Pedro and colleagues (2011) developed a heuristic that provides the context for these
studies within a broader educational research agenda. As seen in the left hand side of Figure 1,
studies have primarily focused on the external risk issues of military children (e.g. deployment),
the protective role of supportive social environments outside the school context (i.e. household,
military community), and their social, emotional, and psychological development (De Pedro et

5
al., 2011). In general, these studies have found that military children experience significant
psychological strain and stress from military-related life events, including parental separation
during deployment, multiple school transitions, veteran-war physical and psychological trauma,
the shifting of household roles and responsibilities, and the visible stress of a left-behind parent
(De Pedro et al., 2011). In the Iraq and Afghanistan war context, these studies have found that
when compared to their civilian peers, military students have significantly more adverse mental
health and child maltreatment outcomes (De Pedro et al., 2011; Huebner et al., 2009; Mmari et
al., 2009). This area of research also includes some studies on the resiliency of military children,
which results from the experience of multiple school transitions and repeated and prolonged
deployments. In these studies, military children and adolescents developed healthy coping skills
and resiliency despite repeated parental separation and school transitions when left-behind
parents were financially stable and families had access to military-specific physical and mental
health services (Bradshaw et al., 2010; Hoshmand & Hoshmand, 2007; MacDermid et al., 2008;
Morris & Age, 2009). This area of research has provided a knowledge base surrounding the
protective role of supportive contexts (households, civilian communities, military bases), but
research on the potential role that supportive public schools play in the social, emotional, and
academic outcomes of military children and adolescents still remains scarce.
De Pedro and colleagues (2011) called for future research to focus on the role of
supportive school environments in promoting positive social, emotional, and academic outcomes
among military students (see right hand side of Figure 1). Within the school environment, De
Pedro and colleagues (2011) posit that studies need to focus specifically on the role of school-
level factors such as principal leadership, teacher and peer awareness and support, and school
climate on the social, emotional, and academic outcomes of military students. De Pedro and

6
colleagues (2011) also addressed the need for future research on contextual factors that influence
a civilian school’s capacity to respond effectively to the needs of military students. As seen on
the right hand side of Figure 1, these contextual factors include school reform surrounding
military-connected schools and support from universities, military installations, and community
organizations.
Figure 1. Dissertation Studies within the Context of a Research Agenda

The studies in this dissertation are situated within the research agenda outlined in Figure
1. This dissertation initiates an area of research that examines the role of a supportive school
environment on the social, emotional, and academic outcomes of military students. Specifically,
these studies examine the school climate of military-connected schools and assess relationships
between school climate and two key social and emotional outcomes among military students,
mental health and victimization.

Military-
specific
External
risk issues
Family and
community
supports
Social,
emotional,
and
psychological
development
of military
children
Principal
Leadership
Teacher
Awareness
and
Support
Peer
Awareness
support
Social
Outcomes
Academic
Outcomes
Emotional
Outcomes
Supportive
School
Climate
School Reform for Supportive School Climates and Military Cultural Group
Support from Universities, Community Organizations, and Military Bases
Societal Contextual Factors (e.g., popular attitudes
toward military)
Military Contextual Factors (e.g. war, military
branch)

7
Theoretically-Driven Model of School Climate
Research has found that the social, emotional and affective qualities of a school
environment—a school’s climate—influence a wide range of student outcomes. These include
academic outcomes (e.g. attendance, motivation, cooperative learning, test scores), social and
emotional outcomes (e.g. bullying and victimization, risky sexual, violence, alcohol, and drug
use behaviors), and psychological outcomes (e.g. psychiatric problems, depression, anxiety)
(Cohen et al., 2009). The bulk of these studies have focused on students’ mental health outcomes
and rates of school violence victimization.
Though hundreds of studies on school climate have been conducted, it must be noted that
researchers have defined and measured school climate in multiple ways and have examined
school climate from diverse disciplinary perspectives (i.e. public health, psychology, educational
research, criminology) (Cohen et al., 2009; Loukas, Suzuki, & Horton, 2006; Rhodes, Camic,
Milburn, & Lowe, 2009; Wang, Selman, Dishion, & Stormshak, 2009; Zullig et al. 2010). For
instance, one study may define school climate as caring relationships, while another study may
define school climate as academic expectations and meaningful participation in school activities.
The lack of commonly used definitions and measures of school climate across academic
disciplines has created barriers for researchers and educators interested in improving school
climate at a practice level. Without a clear idea of what school climate is, how it should be
measured, and what role it plays in improving student outcomes (i.e. academic, social, and
emotional), it is challenging for practitioners and researchers to develop clear and consistent
strategies for improving school climate on a large scale (i.e. in schools, district-wide, state-wide
and nationally) and to change particular elements of a school’s climate, all to promote more
positive outcomes among specific groups of students (i.e. military students)

8
This dissertation utilizes a theoretically-driven school climate framework and measures
representing each construct. Recent reviews of school climate research have asserted the need for
educational researchers to utilize a multidimensional conceptual model and accompanying
measures. Both reviews have posited that there are multiple components of school climate
(Cohen et al., 2009; Zullig et al., 2010). These components include relationships with adults and
peers, peers and adults awareness of student’s family issues, approval of risky behavior, sense of
safety and order, meaningful participation, and belonging (Cohen et al., 2009; Zullig et al.,
2010). Caring relationships is defined as the degree to which students perceive that teachers and
other school adults care about them and provide social and emotional support. Safety involves a
student’s sense of physical security, and sense of social-emotional security, while meaningful
participation is defined as the involvement of the student in relevant, engaging and interesting
opportunities for responsibility and contribution. Risky behavior disapproval is defined as a
student’s perception of peer risky behaviors such as substance use, bullying, weapons
possession, and other risky behaviors in the school context. Belonging involves the extent to
which students feel a sense of belonging and a positive connection with peers and adults in a
school community. Respect for student’s family is defined by the student’s perception that
teachers and peers are aware student’s unique family issues and background. Given the research
on school climate, it is expected that a positive school climate promotes positive mental health
outcomes and lower victimization rates among military and non-military students. The following
sections will describe the multiple dimensions of school climate in detail.
The Multiple Dimensions of School Climate
Caring Relationships. Overall, research suggests that students who have nurturing and
caring social relationships at school have positive long and short-term social-emotional outcomes

9
and academic achievement (Brand et al., 2003; Modin & Ostberg, 2009; Zullig et al., 2010).
Relationships include student-adult school staff member relationships, student-to-student
relationships, and relationships among staff members (Brand et al., 2003; Hoy, Smith &
Sweetland, 2002; Modin & Ostberg, 2009; Zullig et al., 2010).
Supportive relationships with teachers have been shown to help students adjust
behaviorally and emotionally to constantly changing school and classroom contexts (i.e.
changing rules and procedures, academic standards). In a study of middle school and early high
school students in urban schools, Brand & colleagues (2003) found that a high level of teacher
emotional support is positively associated with students’ ability to adjust in school, specifically
in decision-making skills and compliance with school rules and procedures. Brand and
colleagues (2003) also found that a high level of teacher social support was associated with a
student's commitment to academic achievement and persistence.
Supportive teacher relationships are also associated with positive mental health outcomes
among students. Classrooms where teachers provide both social and emotional support can serve
as protective spaces for student self-esteem. In a comparative study of Chinese and American
high school students, Jia and colleagues (2009) found that students' perceptions of teacher
support were positively associated with adolescents' self-esteem. Similarly, in a study of high
school students in Sweden, Modin & Ostberg (2009) found that less frequent psychosomatic
complaints occurred in classes with a high frequency of immediate teacher help with their
schoolwork. In contrast, schools characterized as conflictual or supporting interpersonal
informality/familiarity among students and teachers were related to increases in cluster C
(avoidant, dependent, and obsessive-compulsive personality disorder) symptoms.

10
Together, supportive teacher and peer relationships are associated with lower bullying
and victimization rates. In a study of Israeli high school students, Marachi, Astor, & Benbenishty
(2006) found that higher levels of teacher support were associated with lower rates of
victimization. Further, when students participated in decision-making they were also less likely
to be victimized in school. Similarly, Wilson (2004) found that caring student-teacher
relationships, student-peer relationships, and respect for authority were inversely related to
physical aggression, and verbal harassment. Research has suggested that in these school
environments, students adopt attitudes and behaviors that prevent school violence. In a study of
high schools in Virginia, Eliot and colleagues (2010) found that in schools where students
perceived more peer and teacher social and emotional support, students tended to endorse
positive attitudes toward seeking help for bullying and threats of violence. Other studies have
also found that supportive relationships among peers and school staff facilitated a sense of trust
and an obligation to be fair, and thus, prevented school violence (Flanagan & Stout, 2010; Hoy et
al., 2002). Given the research on relationships with peers and adults, it is expected that
relationships with peers and adults reduce victimization rates and promote positive mental health
outcomes among all students and within military and non-military student populations.
Safety and Meaningful Participation. The safety and order of a school environment is
comprised of individual and collective perceptions of safety, respect for peers and authority,
acknowledgement of disciplinary policies as fair, and a minimal presence of gangs (Zullig et al.,
2010). Research has found that in schools where violence and discipline policies procedures are
perceived by students as fair and consistently implemented, students are likely to have positive
academic, mental health, and behavioral outcomes. In addition, in schools where students have a
high degree of involvement and decision-making in creating and enforcing rules and procedures,

11
students are less likely to engage in violent behaviors and be less victimized (Brand et al., 2003;
Gottfredson et al., 2005).
Not surprisingly, in schools with highly punitive discipline policies, research has found
that students are likely to be fearful of schools and be victims or perpetrators of bullying, which
can adversely affect their academic achievement (Fenning & Rose, 2007; Zullig et al., 2010). In
their research on elementary and high schools students in Israel, students who were personally
victimized by teachers and staff and had a general perception of their school as highly violent
also had a fear of attending. Research literature on the school-to-prison pipeline has suggested
that students, especially African-American and Latino male students, in highly punitive
environments with zero tolerance discipline policies and procedures, are disproportionately
placed in lower tracks and special education, drop out of school, and in the long term, engage in
risky behaviors that lead to imprisonment (Solomon & Palmer, 2006; Opportunity Suspended,
2000).
Research has suggested that in schools where students are active participants in school
violence prevention and safety and violence procedures are consistently implemented and
commonly understood, there are substantial reductions in the rate of bullying and victimization
as well as mental health problems (Brand et al., 2003; Khoury-Kasabri, Benbenishty, & Astor,
2005). In a national study of Israeli secondary students, Khoury-Kassabri & colleagues (2005)
found that in both Jewish and Arab schools, a student’s understanding of school violence policies
and procedures and positive relationships with teachers was negatively associated with serious
physical victimization. In addition, when students are involved in the decision-making process
for responding to school violence, Khoury-Kassabri and colleagues (2005) found significantly
lower rates of all victimization types (e.g. physical, verbal). In a study of American secondary

12
students in urban schools, Brand and colleagues (2003) found that consistently implemented
school discipline policies and procedures and students’ perceptions of a clarity of rules and
expectations facilitated students' behavioral adjustment.
Risky Behavior Disapproval. School climate research has also focused on the influence
of student disapproval of risky behavior among peers (i.e. substance use, weapon carrying) and
shared knowledge of discipline policies and procedures. Studies have found that risky behavior
disapproval can serve as protective factors for school violence and negative mental health
problems. In a study of middle school students, Kuperminc, Leadbeater, & Blatt (2001) found
that favorable perceptions of fairness, order and discipline, and a school culture that disapproves
of drug use moderated the negative effects of a lack of self-efficacy on internalizing and
externalizing problems and of a lack of efficacy on internalizing problems. In addition, research
has suggested schools that have a shared perception of fairness and clarity of rules and discipline
can curb the risk of students’ risky behaviors and delinquency in schools within highly violent
communities. In a national study of secondary students, Gottfredson, Gottfredson, Payne, &
Gottfredson (2005) found that students’ perceptions of fairness of rules, clarity of rules,
organizational focus, morale, planning, and administrative leadership explained a substantial
percentage of the variance in all measures of school disorder (i.e. student drug use and violent
behavior), controlling for the effects of community characteristics. Overall, it is expected that a
school community that disapproves of risky behaviors and has clear and consistent discipline
procedures promotes positive mental health outcomes and reduces victimization rates of students
in military-connected schools.

13
Belonging and Respect for Family and Cultural Background. The school climate
literature has found that a sense of belonging and a perception that school staff and peers have
awareness and respect for students’ family and cultural backgrounds promote positive well-being
and reduced school violence victimization outcomes. In addition, studies suggest that students
with a great sense of belonging to school adults and peers in a community are less likely to have
psychological issues, are less victimized in school, and are more motivated to attain academic
success (Bond et al., 2007; Wilson, 2004; Bonny, Britto, Klostermann, Hornung, & Slap, 2000).
When compared to disengaged and alienated students, research has found that students
who feel highly connected to a school community have less negative mental health issues. For
instance, secondary students who experienced high social connectedness in school reported
fewer anxiety and depressive symptoms and conduct problems during early adulthood and in
post-secondary settings (i.e. college) (Bond et al., 2007; Loukas et al., 2006). In addition,
students who have positive connectedness to a school community also appear to have less
involvement in risky behaviors. Schools with positive connectedness, for instance, have lower
bullying rates, including physical aggression (Wilson, 2004). Positive belonging is also
associated with lower rates of substance use. For instance, Bond and colleagues (2000) found
that highly connected middle school students were less likely than less connected students to
report regular smoking, drinking, and marijuana use outside of school (Bond et al., 2004). In
addition, Bonny & colleagues (2000) found that highly connected middle school students in
urban districts students reported less use of cigarettes and alcohol.
Research has suggested that strategies and interventions that strengthen belonging may
need to be culturally relevant. Studies have found that disconnected and disaffected students are
disproportionately racial and ethnic minorities, found in urban schools, have low parental

14
educational attainment, and male (Bonny et al., 2000.; Skiba, Nardo, & Peterson, 2002).
Culturally relevant pedagogy theorists have posited that racial minority students feel more
connected to a classroom and school community when their cultures and histories are
represented in the curriculum, their languages are utilized in daily instruction, and they are
included in deciding classroom rules and procedures (Hernandez-Sheets, 2003 and 2009;
Ladson-Billings, 1994). Empirically, Brand, Felner, Shim, Seitsinger, & Dumas (2003) found
that schools that were rated by minority students as having higher levels of cultural pluralism and
student participation in decision making were ones in which minority students exhibited
significantly higher levels of behavioral adjustment. Given the preceding studies, it is expected
that a high degree of belonging and a school’s awareness and respect for a student’s family and
cultural background will promote positive mental health outcomes and reduce victimization rates
among students in military-connected schools.
Structure of Dissertation

This dissertation is in a multiple manuscript format as three separate but related studies.
The data in this project were collected in the context of Building Capacity in Military-Connected
Schools. Building Capacity is a consortium of eight military-connected public school districts.
This consortium aims to develop culturally-responsive and supportive school climates in 141
military-connected public elementary, middle, and high schools surrounding military-
installations in the San Diego metropolitan area. The data for these studies are student responses
to the California Healthy Kids Survey (CHKS). The CHKS is a comprehensive survey comprised
of several modules that cover demographic information, school climate, bullying and
victimization, resilience, and sexual health issues. In addition, the CHKS includes specialized
modules, including the Military-Connected Module, which is utilized in this dissertation. The

15
Military-Connected Module includes items on schooling experiences, mental health, and
demographic items anonymously identifying children with a military parent and/or sibling. The
data were collected in the spring of 2011 by researchers in the Building Capacity in Military-
Connected Schools consortium. The sample is a census of all students in the 5
th
, 7
th
, 9
th
, and 11
th

grades in the eight military-connected districts. The data for this study are responses in the core
module and military-connected module from the secondary students in the sample (7
th
, 9
th
, and
11
th
) grades in the eight military-connected districts.
The aim of this multiple manuscript dissertation is to generate multiple studies that can be
readily prepared for a research audience via manuscript submissions to peer-reviewed journals.
The goal of Chapter Two (Study 1) is to provide a macro-level overview of school climate
perceptions by military connection in eight military-connected school districts. Study 1 will
present detailed descriptive statistical analyses of school climate perceptions by military
connection. The school climate constructs under examination will be caring relationships,
belonging, safety, risky behavior disapproval, meaningful participation, and family and cultural
respect. Military-connection includes non-military students, students with a military parent, and
students with a military sibling. This study will examine if school climate perceptions differ by
military connection in a population sample of secondary students in eight military-connected
school districts.
Study 1 will be the first empirical investigation that assesses differences in school climate
perceptions by military connection. The study will explore potential similarities or differences
between military and non-military students across school climate indicators. In addition, unlike
previous studies on school climate, study 1 will be the first study to utilize a theoretically-driven,
multi-dimensional model of school climate. This study will discuss the need for school climate

16
studies to determine the perceptions of highly-stressed demographic groups, such as military-
connected students, across the multiple school climate indicators. Study 1 will conclude with a
discussion relating findings to school climate interventions and educational policy surrounding
military students and military-connected schools.
The goal of Chapter Three (Study 2) is to examine the relationship between school
climate and mental health—well-being, depression, and suicidal ideation. Study 2 will present
analyses on the relationship between school climate and mental health, The mental health
constructs under examination include three dimensions—well-being, depression, and suicidal
ideation. Additional analyses also include school-level variables, including school academic
achievement, school poverty, and school military children concentration.
Study 2 will be the first empirical investigation that evaluates the relationship between
school climate and mental health among students in military-connected schools. In addition,
study 2 will be the first study to examine associations between school climate and mental health
(well-being, depression, and suicidal ideation) within non-military and military student
populations. Analyses from this study will also include the relationship between school climate
and each mental health indicator, controlling for deployment. Study 2 will conclude with a
discussion relating the findings to current research and practice surrounding the mental health of
military students and school climate interventions.
The goal of Chapter Four (Study 3) is to examine the relationship between school climate
and school violence-related victimization among students in military-connected schools. Study 3
will present the relationship between school climate and victimization, including an investigation
of the the relationship between school climate and victimization, controlling for deployment.
Several victimization indicators will be under examination. These include being pushed, fear of

17
getting beaten up, mean rumors, sexual jokes, rumors on the internet, made fun of looks, having
property stolen and threatened with a weapon.
Study 3 will be the first empirical investigation that evaluates the relationship between
school climate and school violence victimization among military and nonmilitary students in a
military-connected school context. Study 3 will be also the first study to examine how school
climate affects students with either a military parent or a military sibling. Most studies have only
explored parental military connections. This study will also examine the relationships between
deployment, school climate, and victimization. Study 3 will conclude with a discussion relating
the findings to current research and practice surrounding the victimization of military students
and school climate interventions.
Overall, study 2 and study 3 will result in a deeper understanding of the role that school
climate plays in the mental health and school violence victimization of students in military-
connected schools. The last section will synthesize the findings of the three studies and discuss
implications of the findings as they relate to school climate interventions as well as the mental
health outcomes and victimization rates of students, military and non-military, in military-
connected schools.

18
Chapter Two:
School Climate Perceptions among Students in Military-Connected Schools: A Comparison
of Military and Non-Military Students from the Same Schools
Introduction
An estimated 1.2 million school-aged children of active-duty military service members,
625,000 children from National Guard families, and 705,000 children from Reservist families are
enrolled in civilian-operated schools. These students comprise a significant minority of students
(at least 400 students or more than 10% of total student enrollment) in 214 school districts in the
United States (Kitmitto et al., 2011). Public school districts that serve a significant number of
military students are called military-connected school districts. Although there is a significant
presence of military children in civilian public schools throughout the United States, only two
peer-reviewed publications in educational research have examined the issues of military children
(De Pedro et al., 2011; Esqueda, Astor, & De Pedro, 2012).
Psychological Outcomes of Military Children
In the past decade, research conducted outside of educational research (public health,
medicine, and psychology) has found that students from military families experience daily
stressors that could adversely impact their academic and social functioning (De Pedro et al.,
2011; Chandra, Martin, Hawkins, & Richardson, 2010; Mmari, Roche, Sudhinaraset, & Blum,
2009). In the current Iraq and Afghanistan war context, military children experience tremendous
stress in different social contexts (i.e. household, community, and school). Since the beginning of
the wars in Afghanistan and Iraq in 2002, studies in public health and psychology have found
that military children experience unique stressors and psychological strain as a result of military-
specific life events. These include the stress of left-behind parents (Chandra et al., 2010; Flake,

19
Davis, Johnson, & Middleton, 2009), the shifting of household responsibilities (Chandra et al.,
2010; Huebner, Mancini, Bowen, & Orthner, 2009; Mmari et al., 2009), the lack of military-
centric social supports for reservists (Chartrand & Seigel, 2007; Hoshmand & Hoshmand, 2007),
reintegration of the veteran (Galovski & Lyons, 2004; Jordan, Marmar, Fairbank, Schlenger,
Kulka, Hough, & Weiss, 1992; Rosenheck & Nathan, 1986; Solomon, 1988), and school
transitions (Medway & Marchant, 1987).
Research indicates that military life stressors negatively affect the psychological
outcomes of military children in the current context of the Iraq and Afghanistan wars. These
studies have found that military children have poorer mental health outcomes than civilian
children due to psychological strain in the present war context. A longitudinal study on military
children, deployments, and mental health found that in 2005 and 2006 behavioral disorders
increased by 19%, stress disorders increased by 18% among military children with a deployed
parent, and the number of mental and behavioral health visits increased by 11% (Gorman, Eide,
& Hisle-Gorman, 2010). In addition, rates increased particularly in adolescents and children of
married and male military parents (Gorman et al., 2010). During the deployment cycle (i.e. pre-
deployment, deployment, reunion, and reintegration), studies have found that left-behind parents
experience tremendous stress and anxiety (Cozza, Chun, & Polo, 2005; Morris & Age, 2009).
Recent research also indicated that military children engage in violent behaviors (i.e. weapon
carrying) and have higher rates of physical and non-physical victimization than their civilian
peers in the school context (Reed, Bell, & Edwards, 2011).

20
Research on School Climate’s Impact on At-Risk Student Populations
Teachers, principals, and other school staff interested in making school environments
more responsive and accommodating to the life challenges of military students could focus on
transforming a school’s social and emotional climate. Two recent literature reviews in
educational research have posited that a school’s climate is comprised of multiple components—
caring relationships with adults, a sense of safety, a sense of belonging to the school community,
disapproval of risky behaviors among peers, high levels of meaningful participation in school
procedures and activities, and respect for the family and cultural background of students (Cohen,
McCabe, Michelli, & Pickerall, 2009; Zullig, Koopman, Patton, & Ubbes, 2010). Studies have
indicated that a school’s social and emotional climate promote positive social, emotional, and
academic outcomes among students. This includes higher attendance, enhanced problem solving,
motivation to learn, higher self-esteem, and more frequent help seeking from teachers and other
school staff (Bond et al., 2007; Wilson, 2004; Bonny, Britto, Klostermann, Hornung, & Slap,
2000). In some cases, research has found that school climate protects students from negative
psychological and behavior outcomes. For example, in a study of secondary students in an urban
high school, supportive relationships with teachers were associated with lower levels of anxiety
and depression (Bond et al., 2007). In addition, in schools where students have a strong sense of
safety and order and understanding of school violence policies and procedures, the rate of school
violence and victimization, anxiety, and depression is substantially reduced (Marachi, Astor, &
Benbenishty, 2006). Furthermore, students who feel they belong to a school community have
reported fewer anxiety and depressive symptoms and conduct problems (Bond, Butler, Thomas,
Carlin, Glover, Bowes, & Patton, 2007; Loukas, Suzuki, & Horton, 2006).

21
Studies have also suggested that a supportive school climate can curb rates of negative
mental health outcomes and school violence among historically oppressed student populations
(Ladson-Billings, 1994; Lynch & Cicchetti, 1997; McGuire, Toomey, & Russell, 2010;
Townsend, 2000). In the United States, for instance, studies conducted in high poverty urban
schools primarily serving Latino and African-American students suggest that different elements
of school climate (i.e. caring relationships with teachers, belonging, meaningful participation in
creating school discipline policies and procedures, and staff respect and awareness of students’
family issues) significantly reduce mental health problems and victimization rates among
African-American and Latino students with low SES status (Ladson-Billings, 1994; Lynch &
Cicchetti, 1997; Townsend, 2000). Studies on the protective role of school climate in negative
mental health, school violence, and disciplinary issues have been conducted among different
student populations (Marachi, Astor, & Benbenishty, 2006; McEvoy & Welker, 2000; McGuire,
Anderson, Toomey, & Russell, 2010).
The Experiences of Military Children in Civilian Public School Environments
A large proportion of military-connected students, about 80%, attend schools in districts
where military-connected students are a significant minority of students (Kitmitto et al., 2011).
Public schools that serve a significant number or proportion of military students (about 400
military students or 4% of total student enrollment) are known as military-connected schools
(Kitmitto et al., 2011). In the context of military-connected schools, recent public health research
suggests that students in military families may have more negative social and emotional climate
perceptions than civilian students (Bradshaw et al., 2010; Chandra et al., 2010; Mmari et al.,
2009). Qualitative studies on military-connected schools have found that military-connected
school teachers, principals, and other school staff lack awareness and appropriate strategies to

22
address the social, emotional and academic needs of military children (Bradshaw et al., 2010;
Chandra et al., 2010; Mmari et al., 2009). For instance, military children reported feeling
misunderstood by school staff members and peers, having difficulty making friends, and
challenges with adjusting to the policies and procedures of a new school after a school transition
(Mmari et al., 2009). Research also indicates that military children are engaging in violent
behaviors in civilian public schools (Mmari et al., 2009; Reed et al., 2011). In addition, Mmari
and colleagues (2009) found that military adolescents with deployed parents exhibited behavior
problems, had fights with other students at school, and had frequent incidences of victimization
by anti-war perpetrators.
The preceding studies indicate that some military students have had negative perceptions
of the social and emotional qualities of their civilian schools. However, the preceding studies
were small in scale, had exploratory objectives, and did not provide detailed comparisons of the
school climate perceptions of military and non-military students in the same schools. In addition,
these studies had exploratory objectives and did not utilize multi-dimensional theoretical
frameworks to provide a systematic understanding of how military children perceive their school
environments. Hence, it is not clear if military students have more negative perceptions than non-
military students. To draw comparisons of school climate perceptions among military and non-
military students, future studies need to utilize large population samples of military and non-
military students in the same schools and employ a school climate conceptual model comprised
of multiple components of school climate (i.e. caring relationships, belonging, safety, meaningful
participation, risky behavior disapproval, and respect for family and cultural background of
students). Findings from these studies can better inform current public school reforms focused on
military-connected schools as well as current school climate interventions.

23
Hence, the goal of this study is to describe the school climate perceptions of students in
military-connected schools and to compare student perceptions of school climate by military
connection. This study utilizes a theoretically-driven school climate model, which includes the
following constructs: belonging, caring relationships, meaningful participation, respect for
family and cultural background, safety, and risky behavior disapproval. Educational researchers
and practitioners can utilize findings from this study to understand how military and non-military
students perceive the social and emotional qualities of public school environments.
Methods
The data from this dissertation is drawn from two sources. The main source of data is the
California Healthy Kids Survey. This study also incorporated some school-level data on
academic achievement and school demographics, publicly available from the California
Department of Education.
Core and Military-Connected Modules
One of the data sources utilized in this study is the 2011 California Healthy Kids Survey
(CHKS). The data from the California Healthy Kids Survey (CHKS) is comprised of individual
level data where individuals are also identified by school, county, and local educational agency
(i.e. public school district) (Austin, Bates, & Duerr, 2011). The CHKS is the largest statewide
survey of resiliency, protective factors, and risk behaviors in the United States and is collected
from students at the elementary, middle, and high school levels (Austin et al., 2011). The CHKS
is comprised of several modules, including the core module. The core module includes data on
demographics (e.g. age, grade, gender, race and ethnicity), health-related behaviors, tobacco use,
drug use, violence behaviors, bullying, victimization, and school climate (Austin et al., 2011).

24
The CHKS also includes the military-connected module. In 2010, a research team at the
University of Southern California and West Ed collaborated to create the CHKS military-
connected module in response to a growing awareness among researchers and educators on the
social, emotional, psychological and academic needs of military children. Congress directed the
U.S. Department of Education (DoE) and U.S. Department of Defense (DoD) to address these
gaps (U.S. Department of Defense, 2008). The military-connected module is administered to
parents, school staff, and students to gather their views of their schools and the educational,
developmental, and health-related needs of students (Austin et al., 2011). The military connected
module includes military demographic items as well as items on deployment, school transitions,
family dynamics, parent-school relationships, mental health, and positive well-being. The student
military connected module is administered to both military and non-military students in military-
connected schools (Austin et al., 2011).
The present study utilizes data from both the core module and military-connected
modules collected in Spring 2011. The CHKS core module included 115 items in the middle
school version and 126 items in the high school version.
Data Collection Procedure
The CHKS core module and the military-connected module were administered in the
Spring of 2011 in eight military-connected school districts serving elementary, middle, and high
school students, surrounding military bases in the San Diego metropolitan area. The participating
schools were required to survey all students in grades 5, 7, 9, and 11. Since the focus of this
dissertation is on secondary students, data collected from 7
th
, 9
th
, and 11
th
graders will be
analyzed. Prior to the survey administered at a secondary school site, parental consent was
gathered for each participating student. Surveys were also sealed in envelopes for classrooms in

25
schools. The envelopes reported the number of students absent in these classes at the day of the
survey. There were 599 envelopes. The surveys were administered by school staff members or
by USC researchers. Proctoring instructions and trainings were given to all survey administrators
and an introductory script was read to the participating students. Participants were encouraged to
answer questions honestly and assured their responses would remain anonymous. Participants
were allowed to withdraw from the survey at any time. The core and military-connected modules
took approximately two hours to complete.
The data used in this study is a subsample of 14,943 7
th
, 9
th
and 11
th
graders, The overall
number of enrolled students eligible to participate in the survey was 18,701. Absent students and
students whose parents refused permission to participate in the study were excluded from the
study sample. The final response rate was 86.73%.
Measures
The following variables and indices taken from the California Healthy Kids Survey
(CHKS) are constructed based on theoretically driven constructs. The following paragraphs
describe the scales and variables used in the model.
Race and Ethnicity. As seen in Table 1, two items were used to determine the racial and
ethnic backgrounds of the participants. Participants were asked, “What is your race?” the
possible responses were 1 = American Indian or Alaska Native, 2 = Asian, 3= Black or African-
American, 5 = Native Hawaiian or Pacific Islander, and 6 = White. Another item asked, “Are
you of Hispanic or Latino origin?” The possible responses were 1) Yes and 2) No. A variable
was computed for race and ethnicity and the categories were 1) Asian/AI/NH/PI 2) Black 3)
White 4) Mixed 5) Hispanic/Latino

26
Grade level. One item was used to determine the grade of the participants. Participants
were asked, “What grade are you in? The possible responses were 1) 6
th
grade, 2) 7
th
grade, 3) 8
th

grade, 4) 9
th
grade, 5) 10
th
grade, 6) 11
th
grade, 7) 12
th
grade, and 8) Other grade
Gender. Participants were asked to report their gender. They were asked, “What is your
sex?” and the possible responses were 1 = male and 2 = female. The items were recoded as 0 =
female and 1 = male.
Military Connection. One item was used to determine a respondent’s military
connection. Participants were asked to report their connection to the military in one item. The
item asks participants, “Who in your family is in the military (Army, Navy, Marines, Air Force,
national Guard or Reserves)?” the possible responses were 1 = No one in my family is in the
military, 2 = Father, 3 = Mother, 4 = Brother or sister, and 5 = Grandparent or other relative.
Based on the literature, military connection is defined as having a parent or sibling serving in the
military (De Pedro et al., 2011; Rodriguez & Margolin, 2011) . Thus, the variable was recoded
with the following categories: 0) Not in the military 1) Having a military parent and 2) Having a
military sibling.
Table 1. Domains, Scales and Items for the Demographic Variables
Variable Items
Gender “What is your sex?”
Respondents chose one of the following: A) Male and B) Female.
Grade “What grade are you in?”
Respondents chose one of the following: A) 6
th
grade; B) 7
th
grade;
C) 8
th
grade; D) 9
th
grade; E) 10
th
grade; F) 11
th
grade; G) 12
th
grade;
H) Other grade; I) Ungraded
Race and Ethnicity “What is your race?”
Possible responses included A) American Indian or Alaska Native;
B) Asian; C) Black or African American; D) Native Hawaiian or
Pacific Islander; E) White; F) Mixed (two or more) races.
“Are you of Hispanic or Latino origin?”
Respondents chose one of the following: A) No; B) Yes
(Table 1 continued)

27
(Table 1 continued)
Military
Connection
“Who in your family is in the military (Army, Navy, Marines, Air
Force, National Guard, or Reserves)?
Respondents could mark more than one of the following answers:
A) No one in my family is in the military; B) Father; C) Mother;
D) Brother or sister; E) Grandparent or other relative; F) Don’t
know

As seen in Table 2, this study covers multiple dimensions of school climate, including
sense of safety, relationship with adults, belonging, and peer risky climate, and meaningful
participation. In addition, the school climate literature has found that a school’s respect for a
student’s family life and cultural background moderates the effect of school climate on mental
health and victimization.

Safety. Two items asked respondents about their perceptions of safety in their schools (α
= .812). One item is “I feel safe in my school.” The responses to this item were on a five point
Likert scale and ranged from 1 = Strongly disagree, 2 = Disagree, 3 = Neither disagree or agree,
4 = Agree, and 5 = Strongly agree. The second item is “How safe do you feel when you are at
school?” The responses to this item were 1 = Very safe, 2 = Safe, 3 = Neither safe nor unsafe, 4
= Unsafe, and 5 = Very unsafe. The item responses were not congruent. This item was recoded
as 1 = Very Unsafe, 2 = Unsafe, 3 = Neither safe nor unsafe, 4 = Safe, 5= Very Safe. The second
item was reverse coded.
Caring Relationships. In this scale, participants were asked about their relationships
with adults in the school (α = .880). Six items were used for this construct: At my school, there is
a teacher or some other adult who tells me I do a good job; who tells me when I do good; who
notices when I’m not there; who always wants me to do my best; who listens to me when I have
something to say; who believes I will be a success. The responses to these items were on a four-

28
point Likert scale and ranged from 1 = Not at all true, 2= A little true, 3 = Pretty much true, and
4 = Very much true.
Belonging. In this scale, participants were asked about their sense of belonging (α =
.797). Three items were used for this construct: Strongly agree or disagree with, I feel close to
people at this school; I am happy to be at this school; I feel like I am part of this school The
possible responses to these items were on a five-point Likert scale and ranged from 1 = Strongly
disagree, 2= Disagree, 3 = Neither disagree or agree, 4 = Agree, and 5 = Strongly agree.
Risky Behavior Disapproval. In this scale, participants were asked about their
perceptions of risky behaviors (α = .900). Five items were used for this construct: How do you
feel about someone your own age smoking one or more packs of cigarettes a day; having one or
more drinks of alcohol every day; trying marijuana or hashish; using marijuana once a month or
more, carrying a weapon to school The possible responses to these items were on a three-point
Likert scale and ranged from 1 = Neither approve or disapprove, 2= Somewhat disapprove, and 3
= Strongly disapprove.
Meaningful Participation. In this scale, participants were asked about their meaningful
participation at school (α = .742). Three items were used for this construct: I do interesting
activities at school; At school, I help decide things like class activities; I do things that make a
difference. The possible responses to these items were on a four-point Likert scale and ranged 1
= Not at all true, 2= A little true, 3 = Pretty much true, and 4 = Very much true.
Respect for Family. One item asked participants to report their school peer’s
understanding for their family life. The item was “Other students in school do not really
understand my family life.” One item asked participants to report school adult’s understanding
for their family life. The item was “Adults in this school respect my family life.” The possible

29
responses to the two items were 1 = Not at all true, 2= A little true, 3 = Pretty much true, 4 =
Very much true. In order to maintain categorical consistency with the other variables, the item
was recoded as 1 = Very much true, 2 Pretty much true, 3 = A little true, and 4 = Not at all true.
Table 2. Domains, Scales and Items for the School Climate Variables
Variable Items
Belonging “Strongly agree or disagree with…
…I feel close to people at this school.”
…I am happy to be at this school.”
…I feel like I am part of this school.”
The possible responses to these items were A) Strongly disagree; 2)
Disagree; 3) Neither disagree or agree; 4) Agree; 5) Strongly agree.
Caring
Relationships
“At my school there is a teacher or some other adult who…
…tells me I do a good job.”
…tells me when I do good.”
…who notices me when I’m not there.”
…who always wants me to do my best.”
…who notices when I’m not there.”
…who always wants me to do my best.”
…who listens to me when I have something to say.”
…who believes I will be a success.”
The possible responses to these items were A) Not at all true; 2) A little
true; 3) Pretty much true; 4) Very much true;
Safety “I feel safe in my school.”
The possible responses to this item were
A) Strongly Disagree; 2) Disagree; 3) Neither disagree or agree; 4)
Agree; 5) Strongly agree.
“How safe do you feel when you are at school?”
The possible responses to this item were
A) Very safe; B) Safe; C) Neither safe nor unsafe; D) Unsafe; E) Very
unsafe.
Meaningful
Participation
“At school…
…I do interesting things at school.”
…I help decide things like class activities.”
…I do things that make a difference.”
The possible responses to this item were A) Not at all true; B) A little
true; C) Pretty much true; D) Very much true; E) Don’t know.
(Table 2 continued)

30
(Table 2 continued)
Risky
Behavior
Disapproval
“How do you feel about…
…someone your own age smoking one or more packs of cigarettes a day.”
…having one or more drinks of alcohol every day.”
…trying marijuana or hashish.”
…using marijuana once a month or more.”
…carrying a weapon to school.”
The possible responses to this item were A) Neither approve nor
disapprove; B) Somewhat disapprove; C) Strongly disapprove
Respect for
Family
“Adults in this school respect my family.”
“Other students in school do not really understand my family.”
The possible responses to this item were A) Not at all true; B) A little true;
C) Pretty much true; D) Very much true; E) Don’t know.

Analytical Plan
The aim of this study is to describe the school climate perceptions of students in military-
connected schools and to compare student perceptions of school climate by military connection.
Descriptive analyses were conducted to compare the school climate perceptions of non-military
students, students with a military parent, and students with a military sibling. As seen in Tables
4-9, bivariate analyses (i.e cross-tabulations) were conducted to determine differences in
percentages among non-military students, students with a military parent, and students with a
military sibling. Bivariate statistical analyses (i.e. chi-square tests of association) were then
conducted to determine associations between each school climate item and military connection.
In addition, Tables 4-9 show that multiple items comprise each construct of school climate.
Results
Sample Characteristics
As seen in Table 3, about 14% of respondents had a connection to the military. 9.3% of
respondents reported having a military parent, while 4.3% reported having a sibling in the
military. As shown in Table 1, approximately half of the respondents were female (51.7%) and
Hispanic (50.3%). White students represented 27.3% of students surveyed. Black students

31
comprised the smallest racial group (3.0%). The proportions of 7
th
, 9
th
, and 11
th
graders were
approximately equal. 32.9% of respondents reported being in the 7
th
grade, 35.2% reported being
in the 9
th
grade, and 31.9% reported being in the 11
th
grade. There was an overrepresentation of
Asian students (12.1%) among respondents with a military parent and an overrepresentation of
Black students among respondents with a military parent (7.2%), while there was an
underrepresentation of Hispanic students with a military parent (44.8%).
Table 3. Sample Characteristics for Study #1 (n= 14,943)
Total
N(%)
No one in
the military
(n=12,990)
Parent
(N=1396)
Sibling
(N=649)
Gender
Female
Male

7606 (51.7%)
7181 (48.6%)

51.5
48.5

51.1
48.9

50.9
49.1
Grade
7
th

9
th

11
th

4588 (32.9%)
4908 (35.2%)
4446 (31.9%)

32.3
35.4
32.3

40.6
33.6
25.8

25.6
33.9
37.5
Race/Ethnicity
Asian/AI/HI/AN
Black
White
Mixed Race
Hispanic

1189 (8.2%)
432 (3.0%)
3948 (27.3%)
1606 (11.1%)
7261 (50.3%)

7.9
2.5
27.3
10.3
52.0

12.1
7.2
25.6
18.1
36.9

7.1
4.0
31.7
12.3
44.8

Military Connection and School Climate
Overall, there were several bivariate associations between military connection and
several items across the six school climate dimensions. These associations showed that students
with a military parent and students with a military sibling had more negative school climate
perceptions. In particular, compared to the non-military students and students with a military
sibling, students with a military parent have a lower sense of belonging, lack caring
relationships, and feel less safe. Also, students with a military sibling have more ambivalent
perceptions of risky behaviors than non-military students and reported more and slightly more

32
negative perceptions of school awareness than students who had no connection to the military
and students with a military parent. No significant differences were identified in meaningful
participation by military connection.
Belonging. As seen in Table 4, students who had a military parent and students with a
military sibling reported having slightly more negative perceptions of belonging than students
with no military connection. Significant bivariate associations were found between military
connection and feeling close to people at their school (X
2
=21.867, df=8, p<0.05). In response to
the item, “I feel close to people at this school,” 6.6% of students with a military parent and 6.2%
of students with a military sibling strongly disagreed, compared to 5.6% of non-military students.
A significant bivariate association was also found between military connection and the item “I
am happy to be at this school” (X
2
= 65.021, df=8, p<0.0001). In response to this item, 10.7% of
students with military parents and 8.9% of students with a military sibling strongly disagreed,
compared to 7.2% of non-military students. A significant bivariate association was also found
between military connection and and students’ feeling part of their school (X
2
= 52,086, df=8,
p<0.0001). 11.2% of students with a military parent and 10.3% of students with a military sibling
strongly disagreed, compared to the 7.6% of non-military students who strongly disagreed.
Table 4. Perceptions of Belonging by Military Connection Status (n=14,943)
Total

Non
Military
(n=12,990)
Military
Parent
(n=1396)
Military
Sibling
(n=649)
I feel close to people at this school*
Strongly disagree
Disagree
Neither disagree/agree
Agree
Strongly Agree

5.7
7.7
23.0
42.6
21.0

5.6
7.4
22.8
43.1
21.1

6.6
9.1
24.4
39.7
20.1

6.2
10.3
24.5
38.5
20.5
(Table 4 continued)

33

(Table 4 continued)
I am happy to be at this school**
Strongly disagree
Disagree
Neither disagree/agree
Agree
Strongly Agree

7.6
7.7
22.4
40.0
22.3

7.2
7.4
22.1
40.4
22.8

10.7
10.6
24.3
37.5
16.8

8.9
7.8
23.4
37.4
22.4
I feel like I am part of this school**
Strongly disagree
Disagree
Neither disagree/agree
Agree
Strongly Agree

8.0
10.8
27.7
37.2
16.4

7.6
10.8
27.7
37.2
16.4

11.2
13.1
27.6
34.7
13.4

10.3
12.2
24.4
34.1
19.1
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001
Safety. Table 5 also shows that students with a military parent feel slightly less safe at
school than students with no military connection. Significant bivariate associations were found
between military connection and the item “I feel safe at my school” (X
2
= 51.467, df=8,
p<0.0001). In response to the item, “I feel safe at my school,” 7.1% of non-military students
strongly disagreed, compared to 10.5% of students with a military parent. Military connection
was also significantly associated with a second safety item, “How safe do you feel when you are
at school?” (X
2
= 41.678, df=8, p<0.0001). About 4.8% of students with a military parent
responded with “very unsafe,” which is slightly higher than the 3.2% of non-military students
who marked the same response. About 2.8% of students with a military sibling reported feeling
very unsafe, slightly lower than both non-military students and students with a military parent.

34

Table 5. Perceptions of Safety by Military Connection Status (n=14,943)
School Climate Dimensions and Items Total

Non
Military
(n=12,990)
Military
Parent
(n=1,318)
Military
Sibling
(n=635)
I feel safe in my school**
Strongly disagree
Disagree
Neither disagree/agree
Agree
Strongly Agree

7.4
8.7
25.6
40.2
18.1

7.1
8.3
25.7
40.6
18.3

10.5
11.8
25.7
36.6
15.4

8.0
10.0
23.5
39.0
19.6
How safe do you feel when you are at
school?**
Very safe
Safe
Neither safe nor unsafe
Unsafe
Very Unsafe

18.9
42.3
30.4
5.1
3.3

19.0
42.6
30.4
4.8
3.2

17.7
39.9
29.5
8.1
4.8

19.7
40.3
31.4
5.7
2.8
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001

Caring Relationships. As seen in Table 6, students with military parents reported
slightly less caring relationships with teachers and other school adults than non-military students
in all items. Significant bivariate associations were found between military connection and
feeling that a teacher really cares (X
2
= 22.150, df=6, p<0.05), having a teacher or adult who
listens to what they have to say (X
2
=14.682, df=6, p<0.05), feeling that there is a teacher who
believes that they will be a success (X
2
= 23.696, df=6, p<0.05), and feeling that there is teacher
or adult who notices when they are not there (X
2
= 18.922, df=6, p<0.05). Slightly higher rates of
negative relationships were found among students with a military parent. For example, in
response to the item, “A teacher or adult who really cares about me,” 14.1% of students with a
military parent responded with “not at all true,” compared to 10.9% of non-military students. In
response to the item, “A teacher or adult tells me when I do a good job,” about 9.6% of students
with a military parent responded with “not at all true,” which is slightly higher than non-military
students (7.7%). In addition, about 11.7% of students with a military parent reported “not at all

35
true” to the item “There is a teacher who listens to me when I have something to say,” compared
to 9.7% of non-military students. In response to the item “There is a teacher or adult who notices
when I’m not there,” about 15.7% of students with a military parent reported “not at all true,”
while a slightly lower proportion of non-military students (12.6%) reported “not at all true.”
Table 6. Perceptions of Caring Relationships by Military Connection Status (n=14,943)

A teacher or adult who…
Total

Non
Military
(n=12,990)
Military
Parent
(n=1,318)
Military
Sibling
(n=635)
Really cares about me*
Not at all true
A little true
Pretty much true
Very much true

11.2
29.6
33.0
26.1

10.9
29.7
33.3
26.1

14.1
30.1
31.2
24.6

10.6
26.9
31.5
31.0
Tells me when I do a good job
Not at all true
A little true
Pretty much true
Very much true

7.9
20.3
35.0
36.8

7.7
20.3
35.2
36.8

9.6
21.1
33.9
35.4

8.6
18.5
32.8
40.1
Wants me to do my best
Not at all true
A little true
Pretty much true
Very much true

5.5
15.0
30.4
49.1

5.4
15.1
30.5
49.0

6.6
15.0
30.1
48.3

5.8
13.3
27.4
53.5
Listens to me when I have something to say*
Not at all true
A little true
Pretty much true
Very much true

9.9
22.0
34.9
33.2

9.7
21.7
35.1
33.4

11.7
24.1
34.4
29.9

9.7
22.9
32.4
35.0
Who believes I will be a success**
Not at all true
A little true
Pretty much true
Very much true

9.5
18.7
31.0
40.8

9.1
18.7
31.3
40.9

11.8
19.9
29.5
38.9

11.8
15.4
28.5
44.3
Who notices when I’m not there *
Not at all true
A little true
Pretty much true
Very much true

12.9
24.9
32.5
29.8

12.6
24.9
32.9
29.5

15.7
24.2
29.5
30.5

13.3
24.3
29.4
33.1
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001

36
Meaningful Participation. No significant bivariate associations were found between
military connection and each of the three meaningful participation items. As seen in Table 7, the
data show that students with a military parent have lower rates of meaningful participation than
non-military students. Significantly different rates by military connection were found in the item,
“I help decide things like class activities or rules.” About 54.3% of students with a military
parent reported “not at all true,” compared to 51.5% of non-military students. Similar proportions
were found in the items, “I do interesting activities at school” and “I do things that make a
difference.”
Table 7. Perceptions of Meaningful Participation by Military Connection Status (n=14,943)
Total

Non
Military
(N=12,990)
Military
Parent
(N=1,318)
Military
Sibling
(N=635)
I do interesting activities at school
Not at all true
A little true
Pretty much true
Very much true

19.0
25.9
25.9
29.1

19.4
26.0
25.8
28.9

19.0
25.8
25.8
29.3

18.7
24.9
24.9
31.5
I help decide things like class activities or
rules*
Not at all true
A little true
Pretty much true
Very much true

51.7
26.4
13.8
8.1

51.5
26.8
13.8
7.9

54.3
23.8
13.3
8.7

52.3
23.5
13.1
11.2
I do things that make a difference
Not at all true
A little true
Pretty much true
Very much true

28.1
34.1
22.7
15.0

28.3
34.0
22.7
14.9

28.5
34.0
22.3
15.2

27.6
32.0
22.7
17.7
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001

Risky Behavior Disapproval. Analyses of military connection and risky behavior
approval suggest that students with a military sibling are slightly more ambivalent of risky
behaviors than students with a military parent and non-military students. Table 8 shows that a

37
significant bivariate association was found between military connection and approval of peers
having one or more drinks ever day (X
2
= 103.826, df=4, p<0.05). Compared to 23.9% of non-
military students and 22.7% of students with a military parent, about 26.1% of students with a
military sibling responded “neither approve or disapprove” to having one or more drinks of
alcohol every day. There were also significant associations between military connection and
approval of peers trying marijuana or hashish (X
2
= 23.516, df=4, p<0.0001) and using marijuana
once a month or more (X
2
= 21.782, df=4, p<0.0001). Students with a military sibling also
responded slightly higher rates of “neither approve or disapprove” to trying marijuana or hashish
(37.5%) and using marijuana once a month or more (35.4%). However, the association between
military connection and carrying a weapon to school was not significant.
Table 8. Perceptions of Risky Behavior Approval by Military Connection Status (n=14,943)
Total

Non
Military
(n=12,990)
Military
Parent
(n=1,318)
Military
Sibling
(n=635)
Smoking one or more packs of cigarettes
Neither approve or disapprove
Somewhat disapprove
Strongly disapprove

20.5
14.3
65.1

21.1
14.5
64.5

18.9
14.1
67.0

22.3
14.2
63.5
Having one or more drinks of alcohol every
day*
Neither approve or disapprove
Somewhat disapprove
Strongly disapprove

22.7
19.6
56.9

23.9
19.9
56.2

22.7
17.1
60.2

26.1
18.6
55.4
Trying marijuana or hashish**
Neither approve or disapprove
Somewhat disapprove
Strongly disapprove

36.2
19.1
44.8

36.8
19.2
44.0

33.5
15.9
50.6

37.5
19.5
43.1
Using marijuana once a month or more**
Neither approve or disapprove
Somewhat disapprove
Strongly disapprove

33.9
17.0
49.0

34.6
17.3
48.1

30.3
15.0
54.8

35.4
16.5
48.1
(Table 8 continued)

38

(Table 8 continued)
Carrying a weapon to school
Neither approve or disapprove
Somewhat disapprove
Strongly disapprove

16.3
9.9
73.8

16.9
9.9
73.2

15.0
9.7
75.3

15.7
10.9
73.4
How would your friends feel about you
smoking one or more packs of cigarettes a day?
Neither approve or disapprove
Somewhat disapprove
Strongly disapprove

15.2
17.0
67.9

15.7
16.8
67.5

14.6
18.5
66.9

16.7
16.1
67.3
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001
Respect for Family. As seen in Table 9, the data suggests that students with a military
sibling have slightly more negative perceptions of peer relationships than students with a military
parent and non-military students. There was one significant bivariate association—military
connection and the item “Other students in school do not really understand me ” (X
2
= 46.290,
df=6, p<0.0001). In response to the item, “Adults in this school respect my family,” about 26.4%
of students with a military sibling responded with “very much true,” which is slightly higher than
the proportion of non-military students (22.1%) and students with a military parent (25.1%).
Table 9. Perceptions of Respect for Family by Military Connection Status (n=14,943)
Total

Non
Military
(N=12,990)
Military
Parent
(N=1,318)
Military
Sibling
(N=635)
Adults in this school respect my family
Not at all true
A little true
Pretty much true
Very much true

15.7
22.7
31.7
30.7

15.5
21.8
31.7
31.0

17.7
24.5
30.2
27.5

14.3
23.5
30.1
32.1
Other students do not really understand me**
Not at all true
A little true
Pretty much true
Very much true

32.6
26.7
18.7
22.1

32.6
26.7
18.7
22.1

26.5
26.5
21.8
25.1

28.0
23.1
22.5
26.4
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001

39
Discussion
Studies have examined the social, emotional, and psychological outcomes of military
children during the Iraq and Afghanistan wars (Chartrand & Seigel, 2007; Gorman et al., 2009;
Hoshmand & Hoshmand, 2007; Huebner et al., 2009). These studies suggest that many teachers,
principals, and other staff in civilian public schools lack awareness and strategies for responding
to the social, emotional, and academic issues of students from military families. In addition,
these studies suggest that a large proportion of military students have more negative school
climate perceptions than civilian students. However, these studies were qualitative, had
exploratory objectives, and did not utilize comprehensive theoretical frameworks of school
climate.
The purpose of this study was to describe the school climate perceptions of military and
non-military students in civilian public schools. This study utilized data from a large population
survey of students in the 7
th
, 9
th
, and 11
th
grades from schools in military-connected school
districts. Bivariate statistical analyses were conducted to compare rates of school climate—
belonging, caring relationships, safety, meaningful participation, respect for family, and risky
behavior disapproval—among non-military students, students with a military parent, and
students with a military sibling. Bivariate statistical analyses were conducted to evaluate
associations between military connection and each school climate construct.
Findings from this study provide descriptive and quantitative empirical evidence that
military-connected students have more negative perceptions of their school environments than
non-military students. The results show that military-connected students consistently reported
significantly lower perceptions of belonging, less supportive relationships with teachers and
other school adults, and lower meaningful participation rates than non military-connected

40
students on all items. This general finding supports recent qualitative studies on the negative
schooling experiences of students from military families.
In addition, this investigation provides rates of school climate perceptions for two
different groups of military-connected students—students with a military parent and students
with a military sibling. The results show that the social and emotional experiences of a school
environment differ by type of military connection. In particular, bivariate statistical analyses
indicated that students with a military parent reported slightly lower rates of belonging, caring
relationships, meaningful participation, and safety than students with a military sibling and non-
military students. Students with a military sibling reported however, more negative perceptions
of respect for family and lower rates of risky behavior disapproval than students with a military
parent and non-military students. However, the results indicated that there were very small rate
differences in school climate perceptions between the military-connected groups— students with
a military parent and students with a military sibling. Moreover, on all indicators, either military-
connected group reported more negative school climate perceptions than non-military students
on all items. This pattern suggests that the psychological impact of having a military sibling may
be similar to the impact of having a military parent.
Studies on children and/or adolescents with a military sibling are scarce. Almost all
studies on military children and adolescents have investigated stressors stemming from having a
parent in the military. However, there is little research that has investigated the psychological
stressors and experiences of adolescents with a military sibling (Rodriguez & Margolin, 2011). A
recent qualitative study found that adolescents with military siblings experienced stress and
anxiety stemming from the grief of a parent, the temporary breakdown of a sibling bond, and
alienation from social supports (Rodriguez & Margolin, 2011). Moreover, psychological

41
theorists have posited that the sibling bond is the most enduring familial bond and heavily
influences an adolescent’s attitudes toward norms regarding risky behaviors (Rodriguez &
Margolin, 2011). More research is needed to understand the schooling experiences of students
with siblings serving in the military.
This study’s findings on school climate perceptions also suggest that staff at military-
connected schools need more awareness and strategies for accommodating the needs and
challenges of military students. Studies have found that school-based extra-curricular activities,
sports and other pro-social school activities with peers facilitate belongingness and academic
achievement for students (Camp, 2000; Perkins & Jones, 2003). Military-connected secondary
students who allocate increased time and energy to household chores and raising younger
siblings are likely to have lower participation rates in school activities than non military-
connected students. Hence, military-connected schools may need to incorporate more flexibility
in providing meaningful participation opportunities to military-connected students (i.e. clubs and
other extracurricular activities specifically for students experiencing the deployment of a family
member.
In addition, recent studies have found military-connected students on average have nine
school transitions from kindergarten to the 12
th
grade (Bradshaw et al., 2010; U.S. Department of
Defense, 2008). While some military-connected school districts have established transition
planning centers and identification procedures, recent studies and technical reports have found
that public schools do not have established procedures for facilitating school transitions for
military-connected students or students in general (Bradshaw et al., 2010). The lack of
identification presents a challenge for military-connected school principals, teachers and other
school staff in assessing the proportion of students who are military-connected and determining

42
if a significant proportion of a student body is experiencing significant stress from deployment
and other military life events. In addition, the lack of transition procedures prevents school staff
from providing military-connected students with appropriate academic, social and emotional
supports. These include providing school accommodations (i.e. flexible attendance policies when
parents return from deployment) that help military-connected students succeed academically
while living daily with military life events. Overall, identification and careful transition planning
could assist teachers in developing a school environment that is culturally responsive to the
schooling experiences of military-connected students.
This study provides a large scale nonclinical perspective of the school climate
perceptions of students in military-connected schools. In addition, comparisons were made
between military and non-military students enrolled in the same school and living in the same
neighborhoods and communities. The scope of this study provided a detailed description of
school climate. Future research needs to examine the impact of a supportive school climate on
social and emotional outcomes known to impact academic achievement such as mental health
and victimization among military-connected students and non-military students. In addition,
military-connected schools widely vary in contextual factors such as school level academic
achievement, socioeconomic status, and concentration of military-connected students. Future
studies need to investigate associations between school-level factors and mental health and
victimization.
Limitations
There are several limitations in this study that are worth noting. First, this study excluded
students from continuation/community day schools. This may have affected the statistical
analyses, since it is feasible that there is an underrepresentation of high-risk secondary students

43
in this study. Second, military connection was defined in two mutually exclusive categories—
students who currently have a parent serving in the military and students who currently have a
sibling serving in the military. A very small proportion of students reported having a grandparent
or other relative who are currently serving in the military. This categorization of military-
connected and non-military students was driven by the research literature (De Pedro et al., 2011;
Rodriguez & Margolin, 2011). These studies have indicated that having a direct family member
in the military (i.e. parent or sibling) is associated with stress and adverse psychological
outcomes among military adolescents. Third, the survey utilized in this study does include items
specifying military students to report if their parents or siblings serve as full-time active duty
service members, Reservists, or National Guard. Military branch is also not specified. Recent
studies have found that the deployment experiences of Reservists and National Guard families
are qualitatively different from families of full-time, active-duty service members (Bowen,
Mancini, Martin, Ware, & Nelson, 2003; MacDermid, Samer, Schwarz, Nishida, & Nyaronga,
2008; Rohall, Weschler, & Segal, 1999). Reservist families, for instance, typically reside in
civilian communities and have a lack of access to social supports and health care providers who
are aware of military culture and life stressors (Bowen et al., 2003). Last, the cross-sectional
design of this study does not allow cause and effect relationships.
Implications
This study is positioned within a larger research agenda aimed at examining the
schooling experiences of military-connected students. De Pedro and colleagues (2011) provide a
rationale for future studies aimed at examining the schooling experiences of military-connected
students. De Pedro and colleagues (2011) found that past research has focused solely on the
external risk issues (i.e. deployment) and resulting mental health outcomes of military children.

44
The authors call for future studies to focus on the impact of a supportive school climate as well
as peer, teacher, and principal awareness and support on the academic, social and emotional
outcomes of military children. This model builds on decades of research on the significant role of
a supportive, nurturing and responsive school environment in promoting positive outcomes for
high-risk student populations (i.e. students experiencing post-trauma issues in war zones,
students in high violence communities).
In summary, this study fills a gap in the military-connected school and military-
literatures. Specifically, future research on students in military-connected schools needs to
incorporate multivariate analyses. These analyses could examine relationships between multiple
components of school climate and key social and emotional outcomes known to impact the social
and academic functioning of secondary students, primarily mental health and school violence
victimization. Multivariate analyses can be conducted to determine whether school climate still
plays a significant role in the mental health outcomes and victimization rates of students who
have experienced the deployment of family members. Moreover, future multivariate analyses
could also assess the degree to which each component of school climate impacts the mental
health outcomes and victimization rates of both non-military and military student populations.
Future studies are also needed to assess cause and effect relationships between protective factors
in the school (i.e. school climate, peer support, etc.) and social and emotional outcomes known to
impact academic functioning. Overall, future research examining the impact of school climate on
the outcomes of military and non-military students in the same schools can potentially inform
school climate interventions.

45
Chapter Three:
The Role of School Climate in the Mental Health of Secondary Students: A Study of
Students in Military-Connected Schools
Introduction
The wars in Afghanistan and Iraq have played a detrimental role in the psychological,
social, and emotional development of the children of military service members. A retrospective
cohort study found that among military children with a deployed parent, the number of mental
and behavioral health visits increased by 11% during the 2005-2006 fiscal year (Gorman, Eide,
& Hisle-Gorman, 2010). In addition, behavioral disorders increased by 19% and stress disorders
increased by 18% (Gorman et al., 2010). Military life events such as parental separation,
household stress during reintegration, exposure to veteran war trauma, multiple school
transitions, and constantly re-adjusting family dynamics adversely affect the mental health of
military children (Chandra, Martin, Hawkins, & Richardson, 2010; Chartrand and Seigel, 2007;
Hoshmand & Hoshmand, 2007; Huebner, Mancini, Bowen, & Orthner, 2009; Jordan, Marmar,
Fairbank, Schlenger, Kulka, Hough, & Weiss, 1992; Rosenheck & Nathan, 1986; Solomon,
1988). Furthermore, research has found that when compared to children in civilian families,
military children have poorer mental health outcomes (Flake, Davis, Johnson, & Middleton,
2009; Galovski & Lyons, 2004).
Stress from the Deployment Cycle
A large body of literature has found that the bulk of military stressors stem from
experiencing the deployment of a family member, specifically a parent or sibling. Amen, Jellen,
Merves, and Lee (1988) have theorized that military families’ experience of deployment can be
divided into three phases—pre-deployment, deployment, and post-deployment. The pre-

46
deployment phase involves planning and preparing for and the eventual departure of a military
parent. The deployment phase includes the challenges associated with maintaining a household
with one parent gone, such as managing household finances and accessing social supports when
needed. The post-deployment phase includes the joy of a reunion with a returning parent and
facilitating the returning parent’s reintegration into society. Overwhelmingly, studies have found
that deployment-related stressors have led to an increase in depressive symptoms, high levels of
anxiety, and suicidal ideation among military children (Burrell, Adams, Durand, & Castro, 2006;
Doyle & Peterson, 2005; Levai, Kaplan, Ackermann, & Hammock, 1995; Reed, Bell, &
Edwards, 2011).
During the phases of the deployment context, military children experience tremendous
stress. Stress and anxiety stem from repeated and prolonged separation from a parent, fears and
anxieties about changes in family dynamics (Huebner, Mancini, Wilcox, Grass, & Grass, 2007),
and fear of parental injury and death (Cozza, Chun, & Pollo, 2005; Flake et al., 2009; Mmari,
Roche, Sudhinaraset, & Blum, 2009). Military children also experience the redistribution of
household roles and responsibilities (Faber, Willerton, Clymer, MacDermid, & Weiss, 2008;
Huebner et al., 2007). Often, when a parent is deployed, military children, especially adolescents,
take on household roles and responsibilities, previously assumed by parents and other adults. In a
qualitative study of adolescent youth coping during deployment, Mmari and colleagues (2009)
found that adolescents’ family roles were enhanced and they performed household duties
normally reserved for adults. Deployment-period responsibilities included increased household
work such as more chores and taking care of younger siblings. Even after a deployed parent is
reunited with his or her family, the experience of reintegration can be stressful. Studies have
found that reintegrated military families must deal with shifting household responsibilities and

47
roles, the stress of a veterans’ re-establishment of employment, and the veterans’ process of re-
establishing a relationship with the military child after months or even years of separation
(Burrell et al., 2006; Doyle & Peterson, 2005; Peebles-Kleiger and Kleiger, 1994).
The bulk of research on military children and mental health has focused exclusively on
children with a military parent. Research on the potential adverse impacts on mental health as a
result of having a sibling in the military is still in its nascent stages. A recent qualitative study
examined in detail that the deployment of a sibling creates psychological strain for a military
child. Rodriguez & Margolin (2011) found that a sense of loss and anxiety permeates the family
system, which, in turn, led to a families’ isolating themselves from support systems and
retreating from previously enjoyable activities. Military children are also aware of their parents’
pain, sadness and worry about the sibling’s safety. Similar to parental deployment, Rodriguez
and Margolin (2011) found that parents would become more protective of a left-behind sibling
with the goal of keeping them from being lonely or sad. However, positive outcomes were also
uncovered, specifically that a military sibling served as important role model of hard-work, self-
sacrifice, and discipline toward personal goals.
Like the existing research literature assessing the relationship between the outcomes of
military children and military-specific stressors, studies have also documented the negative
impact of external risk issues on the academic, social-emotional, and psychological outcomes of
students from historically oppressed demographic groups (by race, ethnicity, gender, SES, and
sexual orientation) (Finn & Rock, 1997; Gutman, Sameroff, & Eccles, 2002; Kagan, 1990). For
instance, research on low-SES, African-American and Latino students residing in high poverty
neighborhoods, has found that these students experience external risk issues such as
discrimination, community violence, household financial stress, war trauma, and high mobility

48
(Finn & Rock, 1997; Gutman et al., 2002; Kagan, 1990; Marachi, Astor, & Benbenishty, 2006).
These factors adversely impact long-term psychological and social outcomes, as well as
academic achievement (Anyon, 1997; Lynch & Cicchetti, 1997; Townsend, 2000).
The Role of School Climate
Studies conducted in various academic disciplines have indicated that a supportive school
climate can reduce students’ risky behavior outcomes (Eisenberg, Neumark-Sztainer, & Perry,
2003; Elliot, Gregory, & Fan, 2010; Hoy, Smith, & Sweetland, 2002; Hoy, Tarter, & Bliss,
1990). Recent reviews of school climate have concluded that there are multiple components of
school climate (Cohen, McCabe, Michelli, & Pickerall; 2009; Zullig, Koopman, Patton, &
Ubbes, 2010). These components include caring relationships with adults and peers, respect for
family issues, approval of risky behavior, sense of safety and order, meaningful participation,
and belonging. Caring relationships is defined by student’s perception of the extent of social and
emotional support they receive from teachers and other school adults, and respect for student’s
family is defined by a student’s perception that teachers and peers are aware of his or her family
issues and background. The dimension of risky behavior disapproval is defined as a student’s
view of risky behaviors among peers. These include substance use, bullying, weapons
possession, and other risky behaviors in the school context. Safety is defined as a student’s sense
of physical security and social-emotional security,is defined as a student’s perception of his or
her participation in school-wide and extracurricular activities. Belonging is defined as the degree
to which students feel they belong to a school community and have a positive connection with
peers and adults in a school community.
Supportive and nurturing school climates can potentially promote positive mental health
among students in different geographical and cultural contexts (Marachi, Astor, & Benbenishty,

49
2006; Pallas, 1988; Wang, 2009). Some studies have found that supportive relationships with
peers and teachers help students adjust behaviorally and emotionally in school contexts, persist
when encountering challenging social and academic situations, develop high self-esteem, and
seek help from adults (Bond et al., 2007; Wilson, 2004; Bonny, Britto, Klostermann, Hornung, &
Slap, 2000). In some cases, supportive relationships with teachers have contributed to lower
levels of anxiety, depression, and obsessive-compulsive personality disorder symptoms. In
schools where students have a strong sense of safety and order and understanding of school
violence policies and procedures, students exhibit less anxiety and depression symptoms
(Marachi, Astor, & Benbenishty, 2006; Wang, 2009). Not surprisingly, students who feel
socially and emotionally connected to a school community have reported fewer anxiety and
depressive symptoms and conduct problems (Bond et al., 2007; Loukas, Suzuki, & Horton,
2006).
In particular, a supportive school climate can curb the negative mental health outcomes of
psychologically strained minority populations. In the United States, for instance, two decades of
research conducted in high poverty urban schools have indicated that different elements of
school climate (i.e. supportive relationships with teachers and peers, connectedness to school
staff, an active role in creating school discipline policies and procedures, and awareness of
students’ family background and issues) significantly reduce mental health problems and
victimization rates among low SES African-American and Latino students (Ladson-Billings,
1994; Lynch & Cicchetti, 1997; McGuire, Anderson, Toomey, & Russell, 2010;).
Given research on other psychologically strained student populations, it would be
expected that a supportive school climate, where teachers and peers are aware of students’
unique family issues and background, would facilitate positive mental health outcomes among

50
military children. Recent studies have suggested that there are negative perceptions of school
climate (i.e. belonging, relationships, and school safety) among military students (Bradshaw,
Sudhinareset, Mmari, & Blum, 2010; Chandra et al., 2010; Mmari, Bradshaw, Sudhinaraset, &
Blum, 2010; Mmari, Roche, Sudhinaraset, & Blum, 2009). Chandra and colleagues (2010) found
that military children reported being bullying victims of anti-war student perpetrators, a sense of
alienation from civilian peers, and a feeling of being misunderstood by school staff members. In
addition, Mmari and colleauges (2010) found that due to multiple transitions, military students
had difficulty with making friends, adjusting to new physical school environments, addressing
learning gaps, and proactively seeking academic and emotional support from school staff and
peers.
While the preceding studies provide some indication of negative social and emotional
climate for military children in civilian schools, some investigations have uncovered attempts by
educators to provide supportive school environments for military children. A few studies have
identified the use of some “homegrown” strategies and practices, primarily aimed toward social
and school connectedness and supportive teacher-student relationships (Bradshaw et al., 2010;
Chandra et al., 2010). This included encouraging military students to participate in
extracurricular activities, connecting military students and families, increased communication
between schools and military families, professional development on military culture, and
assigning a school staff member to act as a supportive adult for military students (Bradshaw et
al., 2010; Chandra et al., 2010; Mmari et al., 2010; Mmari et al., 2009). However, none of these
studies have measured the impact of a supportive school climate on the mental health outcomes
of military children.

51
The preceding studies on military children in public schools have assisted us in exploring
their social and emotional experiences in civilian public schools. Future research needs to
systematically examine how they experience the social and emotional qualities of their school
environments and the potential impact of school climate on their mental health outcomes. In
addition, research so far has focused predominantly on the emotional and psychological
experiences of children whose parents are military-service members while paying little attention
to the experiences of children whose siblings serve in the military. Given the findings from
recent research on military siblings, future research would also to need expand the term military
children to include children with military siblings.
This study serves to address these gaps. In this study, I evaluate and compare the mental
health outcomes of military children (with a parent or sibling in the military) and non-military
children as well as examine relationships between school climate and mental health. Drawing
from existing research on the mental health of military children, it is expected that military
children have more adverse mental health outcomes than non-military children. There has also
been a wide breadth of research on school climate and mental health; it is expected that school
climate promotes positive mental health outcomes and curbs negative mental health outcomes
among military children. Moreover, considering research on the protective effects of school
climate on the mental health outcomes of psychologically strained student populations, it is also
expected that a supportive school climate would promote positive mental health outcomes even
if students have experienced military stressors such as a deployment.
Considering the gaps in school climate research and studies on military children and
mental health, multiple objectives guide this study. First, this study aims to compare mental
health outcomes, specifically the well-being, depression, and suicidal ideation rates of military

52
and non-military students in a normative sample of students in military-connected schools.
Second, this study evaluates associations between school climate and mental health among
students. Third, this study seeks to examine associations between school climate and mental
health, accounting for deployment. Last, this study seeks to compare the effects of school climate
on mental health within separate subsamples of military and non-military students.
Methods
Data Sources
The main source of data in this study is the 2011 California Healthy Kids survey
(CHKS). The data from the California Healthy Kids Survey (CHKS) is comprised of individual
level data where individuals are also identified by school, county, and local educational agency
(i.e. public school district). The CHKS is the largest statewide survey of resiliency, protective
factors, and risk behaviors in the United States and is collected from students at the elementary,
middle, and high school levels. The CHKS is comprised of several modules, including the core
module. The core module includes data on demographics (e.g. age, grade, gender, race and
ethnicity), health-related behaviors, tobacco use, drug use, violence behaviors, bullying,
victimization, and school climate (Austin, Bates & Duerr, 2011).
The CHKS also includes the military-connected module. In 2010, a research team at the
University of Southern California and West Ed collaborated to create the CHKS military-
connected module in response to a growing awareness among researchers and educators on the
social, emotional, psychological and academic needs of military children. Congress directed the
U.S. Department of Education (DoE) and U.S. Department of Defense (DoD) to address these
gaps (U.S. Department of Defense, 2008). The military-connected module is administered to
parents, school staff, and students to gather their views of their schools and the educational,

53
developmental, and health-related needs of their students. The military-connected module
includes military demographic items as well as items on exposure to deployment, school
transitions, family dynamics, parent-school relationships, mental health, and positive well-being.
The student military-connected module is administered to both military and non-military
students in military-connected schools.
The present study utilizes data from both the core module and military-connected
modules collected in Spring 2011. The CHKS core module included 115 items in the middle
school version and 126 items in the high school version.
CHKS Data collection procedure. The CHKS core module and the military-connected
module were administered in the Spring of 2011 in eight military-connected school districts
serving elementary, middle, and high school students, surrounding military bases in the San
Diego metropolitan area. The participating schools were required to survey all students in grades
5,7, 9, and 11. Since the focus of this dissertation is on secondary students, data collected from
7
th
, 9
th
, and 11
th
graders will be analyzed. Prior to the survey administered at a secondary school
site, parental consent was gathered for each participating student. Surveys were also sealed in
envelopes for classrooms in schools. The envelopes reported the number of students absent in
these classes at the day of the survey. There were 599 envelopes. The surveys were administered
by school staff members or by USC team members. Proctoring instructions and trainings were
given to all survey administrators and an introductory script was read to the participating
students. The USC research team provided survey administrators with gift cards as incentives, in
order to facilitate high response rates. Participants were encouraged to answer questions honestly
and assured their responses would remain anonymous. Participants were allowed to withdraw

54
from the survey at any time. The core and military-connected modules took approximately 2
hours to complete.
The CHKS data used in this study is a subsample of 14,943 7
th
, 9
th
, and 11
th
graders. The
overall number of enrolled students eligible to participate in the survey was 18,701. Absent
students and students whose parents refused permission to participate in the study were excluded
from the study sample. The final response rate was 86.73%.
California Department of Education (CDE) Dataquest. The CHKS dataset used for
this dissertation was also merged with two school-level items from the 2010 Accountability
Progress data, Base API and proportion of school receiving free and reduced price meals. A Base
API can be used as a measure of a school’s academic achievement level. A base API score is
calculated from the following indicators: Standardized Testing and Reporting (STAR) test
indicators in core academic subjects (English language arts, math, science, and history-social
science), California Modified Assessment (CMA) test results, California Alternate Performance
Assessment (CAPA) results, and California High School Exit Examination (CAHSEE) results
(California Department of Education, 2011). The API scores fall into a numeric index ranging
from 200 to 1000. After calculation of API, schools receive accountability information in API
reports.
Data related to the poverty level of students in a school is also provided, specifically the
proportion of students qualifying for free and reduced priced meals. The USDA determines
federal eligibility requirements for students to receive free and reduced price lunch (California
Department of Education, 2011). The USDA Department’s guidelines for free and reduced price
meals were obtained by multiplying the 2011 Federal Income poverty guidelines by 1.30 and
1.85 respectively, and by rounding the result upward to the next whole dollar. The data in this

55
dissertation includes percent of students who qualify for free and reduced priced meals by
school.
Measures
Dependent Variables
Well-being. As seen in Table 10, participants were asked about their well-being (α =
.894). Five items were used for this construct: In the last 30 days, how often did you feel full of
energy; happy; proud; good about life; excited; strong. The responses to these items were on a
five-point Likert scale and ranged from 1 = None of the time, 2= A little of the time, 3 = Some of
the time, 4 = Most of the time, 5 = All of the time. First, a well-being scale was created. The
composite score of each participant was derived by summing up the well-being items. Second,
well-being scale scores were divided by two at a cutpoint of 75% to create a variable with a
dichotomous outcome, where 1 = high well-being and 0 = low well-being. A cut off of the 75
th

percentile was utilized to identify those with high levels of well-being.
Depression. As seen in Table 10, participants were asked questions related to depression
(α = .812). Five items were used for this construct: In the last 30 days, how often did you feel
like everything feels hard to do; nothing can cheer you up; restless, cannot stay still in one place;
nervous; do not have much hope. The responses to these items were on a five-point Likert scale
and ranged from 1 = None of the time, 2= A little of the time, 3 = Some of the time, 4 = Most of
the time, 5 = All of the time. First, a depression scale was created. The composite score of each
participant was derived by summing up the depression items. Second, depression scale scores
were divided by two at a cutpoint of 75% to create a variable with a dichotomous outcome,
where 1 = depression and 0 = no depression. A cut off of the 75
th
percentile was utilized to
identify those with high levels of depression.

56
Suicidal Ideation. As seen in Table 10, participants were asked about suicidal ideation in
the past year, “During the past 12 months, did you ever seriously consider suicide?” The
responses to this item were 1 = No and 2 = Yes.
Table 10. Domains, Scales and Items for the Mental Health
Variable Items
Depression “In the last 30 days, how often did you feel…
…like everything feels hard to do.”
…like nothing can cheer you up.”
... restless, cannot stay still in one place.”
…nervous.”
…that you do not have much hope.”
…like you are not important.”
The possible responses to this item were A) None of the time; B) A
little of the time; C) Some of the time; D) Most of the time; E) All of
the time.
Well-Being “In the last 30 days, how often did you feel…
…feel full of energy.”
…happy.”
... proud.”
…nervous.”
…good about life.”
…excited.”
…strong.”
The possible responses to this item were A) None of the time; B) A
little of the time; C) Some of the time; D) Most of the time; E) All of
the time.
Suicidal Ideation

“During the past 12 months, did you ever seriously consider suicide?”
The possible responses to this item were A) No B) Yes

Independent Variables
School Climate Perceptions. According to the school climate literature, there are
multiple dimensions of school climate in the school context. As seen in Table 11, The CHKS
core module provides items covering multiple dimensions of school climate. These include sense
of safety, relationship with adults, belonging, and peer risky climate, meaningful participation,
and school’s respect for student’s family issues.

57
Safety. Two items asked respondents about their perceptions of safety in their schools (α
= .812). One item is “I feel safe in my school.” The responses to this item were on a five point
Likert scale and ranged from 1 = Strongly disagree, 2 = Disagree, 3 = Neither disagree or agree,
4 = Agree, 5 = Strongly agree. The second item is “How safe do you feel when you are at
school?” The responses to this item were 1 = Very safe, 2 = Safe, 3 = Neither safe nor unsafe, 4
= Unsafe, 5 = Very unsafe. The item responses were not congruent. The second item was
recoded as 1 = Very Unsafe, 2 = Unsafe, 3 = Neither safe nor unsafe, 4 = Safe, 5= Very Safe.
The composite score of each participant was derived by summing up the two items and dividing
by two. Therefore, the safety scale scores range from 1-5.
Relationship with Adults. In this scale, participants were asked about their relationships
with adults in the school (α = .880). Six items were used for this construct: At my school, there is
a teacher or some other adult who tells me I do a good job; who tells me when I do good; who
notices when I’m not there; who always wants me to do my best; who listens to me when I have
something to say; who believes I will be a success. The responses to these items were on a four-
point Likert scale and ranged from 1 = Not at all true, 2= A little true, 3 = Pretty much true, 4 =
Very much true. The composite score of each participant was derived by summing up the six
items and dividing by six. Therefore, the relationships scale scores range from 1-4.
Belonging. In this scale, participants were asked about their sense of belonging (α =
.797). Three items were used for this construct: Strongly agree or disagree with, I feel close to
people at this school; I am happy to be at this school; I feel like I am part of this school The
possible responses to these items were on a five-point Likert scale and ranged from 1 = Strongly
disagree, 2= Disagree, 3 = Neither disagree or agree, 4 = Agree, 5 = Strongly agree. The

58
composite score of each participant was derived by summing up the three items and dividing by
three. Therefore, the safety scale scores range from 1-5.
Risky Behavior Approval. In this scale, participants were asked about their perceptions of
risky behaviors (α = .900). Six items were used for this construct: How do you feel about
someone your own age smoking one or more packs of cigarettes a day; having one or more
drinks of alcohol every day; trying marijuana or hashish; using marijuana once a month or more,
carrying a weapon to school; how would your friends feel about you smoking one or more packs
of cigarettes a day? The possible responses to these items were on a three-point Likert scale and
ranged from 1 = Neither approve or disapprove, 2= Somewhat disapprove, 3 = Strongly
disapprove. The composite score of each participant was derived by summing up the six items
and dividing by six. Therefore, the safety scale scores range from 1-3.
Meaningful Participation. In this scale, participants were asked about their meaningful
participation at school (α = .742). Three items were used for this construct: I do interesting
activities at school; At school, I help decide things like class activities; I do things that make a
difference. The possible responses to these items were on a four-point Likert scale and ranged 1
= Not at all true, 2= A little true, 3 = Pretty much true, 4 = Very much true. The composite score
of each participant was derived by summing up the three items and dividing by 3. Therefore, the
safety scale scores range from 1-4.
Respect for Family. Two items asked participants to report their school’s respect for their
family life (α = .85). The first item was “Adults in this school respect my family.” The second
item asked participants to report school adult’s understanding for their family life. The item was
“Other students in school do not really understand my family life.” The possible responses to this
item were 1 = Not at all true, 2= A little true, 3 = Pretty much true, 4 = Very much true. The

59
items are not congruent. The second item was reverse coded. The composite score of each
participant was derived by summing up the two items and dividing by two. Therefore, the safety
scale scores range from 1-4.
Table 11. Domains, Scales and Items for the School Climate Variables
Variable Items
Belonging “Strongly agree or disagree with…
…I feel close to people at this school.”
…I am happy to be at this school.”
…I feel like I am part of this school.”
The possible responses to these items were A) Strongly disagree; 2)
Disagree; 3) Neither disagree or agree; 4) Agree; 5) Strongly agree.
Relationships
with Adults
“At my school there is a teacher or some other adult who…
…tells me I do a good job.”
…tells me when I do good.”
…who notices me when I’m not there.”
…who always wants me to do my best.”
…who notices when I’m not there.”
…who always wants me to do my best.”
…who listens to me when I have something to say.”
…who believes I will be a success.”
The possible responses to these items were A) Not at all true; 2) A little
true; 3) Pretty much true; 4) Very much true;
Safety “I feel safe in my school.”
The possible responses to this item were
A) Strongly Disagree; 2) Disagree; 3) Neither disagree or agree; 4) Agree;
5) Strongly agree.
“How safe do you feel when you are at school?”
The possible responses to this item were
A) Very safe; B) Safe; C) Neither safe nor unsafe; D) Unsafe; E) Very
unsafe.
Meaningful
Participation
“At school…
…I do interesting things at school.”
…I help decide things like class activities.”
…I do things that make a difference.”
The possible responses to this item were A) Not at all true; B) A little true;
C) Pretty much true; D) Very much true; E) Don’t know.
(Table 11 continued)

60
(Table 11 continued)
Approval of
Risky Behaviors
“How do you feel about…
…someone your own age smoking one or more packs of cigarettes a day.”
…having one or more drinks of alcohol every day.”
…trying marijuana or hashish.”
…using marijuana once a month or more.”
…carrying a weapon to school.”
The possible responses to this item were A) Not at all true; B) A little true;
C) Pretty much true; D) Very much true.
Respect for
family
“Adults in this school respect my family.”
“Other students in school do not really understand my family.”
The possible responses to this item were A) Not at all true; B) A little true;
C) Pretty much true; D) Very much true; E) Don’t know.

Deployment (Past Ten Years). As seen in Table 12, one item asked participants to report
their experiences with the deployment of a family member. The item was “About your family
and the military, as far as you can remember, how many times in the last 10 years did any
member of your family leave home and serve (deploy) outside the USA?” The possible
responses to this item were 1 = Never, 2 = Once, and 3 = Twice or more. Two dummy variables
were created: Deployment (1 time) and deployment (2 or more times).
Table 12. Deployment Item
Variable Item
Deployment “As far as you can remember, how many times in the past 10 years
did any member of your family leave home and serve (deploy) outside
the USA?”

The possible responses to this item were A) Never; B) Once; C) Twice
or more; D) Don’t know.

Race and Ethnicity. As seen in Table 13, two items were used to determine the racial and
ethnic backgrounds of the participants. Participants were asked, “What is your race?” the
possible responses were 1 = American Indian or Alaska Native, 2 = Asian, 3= Black or African-
American, 5 = Native Hawaiian or Pacific Islander, and 5 = White. Another item asked, “Are

61
you of Hispanic or Latino origin?” The possible responses were 1) Yes and 2) No. A variable
was computed for race and ethnicity and the categories were 1) Asian/AI/NH/PI 2) Black 3)
White 4) Mixed 5) Hispanic/Latino. Then, race and ethnicity was represented by White as the
reference and four dummy variables—Asian/AI/NH/PI, Black, Mixed, and Hispanic/Latino.
Grade level. One item was used to determine the grade of the participants. Participants
were asked, “What grade are you in? The possible responses were 1) 6
th
grade, 2) 7
th
grade, 3) 8
th

grade, 4) 9
th
grade, 5) 10
th
grade, 6) 11
th
grade, 7) 12
th
grade, 8) Other grade. Grade level was
represented by a series of dummy variables: 7
th
grade, 9
th
grade, and 11
th
grade.
Gender. Participants were asked to report their gender. They were asked, “What is your
sex?” and the possible responses were 1 = male and 2 = female. The items were recoded as 0 =
female and 1 = male. Gender was represented by a dummy variable coded as 1= female and 0 =
male.
Military Connection. One item was used to determine a respondent’s military
connection. Participants were asked to report their connection to the military in one item. The
item asks participants, “Who in your family is in the military (Army, Navy, Marines, Air Force,
national Guard or Reserves)?” the possible responses were 1 = No one in my family is in the
military, 2 = Father, 3 = Mother, 4 = Brother or sister, and 5 = Grandparent or other relative.
Based on the literature, military connection is defined as having a parent or sibling serving in the
military (De Pedro et al., 2011; Rodriguez & Margolin, 2011). Thus, the variable was recoded
with the following categories: 0) Not in the military 2) Having a military parent and 3) Having a
military sibling. Military connection was represented by two dummy variables: Having a military
parent and having a military sibling.

62
Table 13. Domains, Scales and Items for the Demographic Variables
Variable Items
Gender “What is your sex?”
Respondents chose one of the following: A) Male and B) Female.
Grade “What grade are you in?”
Respondents chose one of the following: A) 6
th
grade; B) 7
th
grade; C)
8
th
grade; D) 9
th
grade; E) 10
th
grade; F) 11
th
grade; G) 12
th
grade; H)
Other grade; I) Ungraded
Race and
Ethnicity
“What is your race?”
Possible responses included A) American Indian or Alaska Native; B)
Asian; C) Black or African American; D) Native Hawaiian or Pacific
Islander; E) White; F) Mixed (two or more) races.
“Are you of Hispanic or Latino origin?”
Respondents chose one of the following: A) No; B) Yes
Military
Connection
“Who in your family is in the military (Army, Navy, Marines, Air Force,
National Guard, or Reserves)?
Respondents could mark more than one of the following answers: A) No
one in my family is in the military; B) Father; C) Mother; D) Brother or
sister; E) Grandparent or other relative; F) Don’t know

School Level Variables
Taken from publically available CDE data, two school-level variables were also used to
represent two school contextual factors surrounding school climate effects on mental health and
victimization—academic performance index (API) and free and reduced price lunch. A third
school-level variable, military concentration, was generated from student level data from the
CHKS military-connected module.
Academic Performance Index (API). To obtain a school’s base API score, the
following indicators are calculated: Standardized Testing and Reporting (STAR) test indicators
in core academic subjects (English language arts, math, science, and history-social science),
California Modified Assessment (CMA) test results, California Alternate Performance
Assessment (CAPA) results, and California High School Exit Examination (CAHSEE) results.
The API scores fall into a numeric index ranging from 200 to 1000.

63
Free and Reduced Price Meals. The California Department of Education also publishes
the percentage of students who qualify for free and reduced price meals in each school. The
USDA determines federal eligibility requirements for students to receive free and reduced price
lunch. The USDA Department’s guidelines for free and reduced price meals were obtained by
multiplying the 2011 Federal Income poverty guidelines by 1.30 and 1.85 respectively, and by
rounding the result upward to the next whole dollar.
Military Concentration. The military-connected module in the CHKS includes an item
that asks participants the following: “Who in your family is in the military (Army, Navy,
Marines, Air Force, National Guard, or Reserves?” The possible responses are A) No one in my
family is in the military, B) Father, C) Mother, D) Brother or sister; E) Grandparent or other
relative; F) Don’t know. The item was recoded as 1 - Having a parent in the military and 0 - No
parent in the military. A variable aggregated at the school level was created. This aggregate
variable represents the proportion of students with a military parent in a school.
Analytical Plan
The aim of this study was to examine associations between school climate and mental
health among students in military-connected schools. The data was analyzed using bivariate and
multivariate analyses. Bivariate analyses were conducted to outline the characteristics of the
sample by gender, grade level, race and ethnicity, and military connection. Bivariate analyses
were then performed to analyze mental health outcomes among non-military students, students
with a military parent and students with a military sibling. Analysis of variance (ANOVA) tests
of the well-being and depression scales by military-connection compared the overall well-being
and depression rates of the three groups. Cross-tabulations were then generated to compare the
three groups by each well-being, depression, and suicidal ideation item. Chi-square analyses

64
were conducted to determine significant associations between military-connection and each
mental health item.
Multivariate analyses using logistic regression were then conducted. The logistic
regression technique is used when the dependent variable is limited or constrained to a
dichotomous outcome. The dependent variables in this study were nominal and dichotomous and
coded as 0 or 1. A direct method procedure was used to assess the predictive value of
independent variables of interest, while controlling for other independent variables. In addition,
logistic regression is an appropriate statistical technique for examining differences in outcomes
by demographic group.
In this study, logistic regression analyses were conducted to examine associations
between school climate and mental health. Separate logistic regression analyses were conducted
to assess relationships between school climate and mental health within each group of students
by military-connection—non-military students, students with a military parent, and students with
a military sibling. Logistic regression analyses were also conducted to assess relationships
between school climate and mental health, controlling for deployment.
Prior to conducting the logistic regression analyses, several decisions were made to
evaluate the logistic regression models. First, statistical significance tests were set at p < .05.
Second, it was determined that the strength of each independent variable would be measured by
the odds ratios and confidence intervals of each predictor. Odds ratios and confidence intervals,
instead of individual regression coefficients, provide the researcher easier interpretation of
results and are generally used in studies with large samples. Third, a series of indicators, the
likelihood ratio test, Cox & Snell R
2
, Nagelkerke R
2
,

were used to determine if each logistic
model was a model fit. A logistic model is said to provide a better fit to the data if it

65
demonstrates an improvement over the intercept model, which contains no predictors. Fourth,
goodness-of-fit statistics were used to assess the fit of a logistic model against observed
outcomes. The inferential goodness-of-fit statistic used in this study is the Hosmer-Lemeshow
(H-L) test. The H-L statistic is a Pearson chi-square statistic calculated from a 2 x g table of
observed and estimated expected frequencies, where g is the number of groups formed from the
estimated probabilities. A non-significant H-L statistic indicates that there is no significant
difference between the observed frequencies and the expected frequencies derived from the
model, and hence, the model fits the data well. Fifth, the percent correct prediction statistic was
evaluated to determine the percent of all cases correctly predicted in each logistic regression
model. This statistic is an indicator of predictive power of each logistic regression model. . In
this study, overall percent correct prediction statistics for all logistic regression models ranged
from 75% to 95%. Also, the number of cases observed to be 0 that were correctly predicted to be
0 ranged from 682 to 9543. The number of cases observed to be 1 that were correctly predicted
to be 1 ranged from 88 to 1551.
Sixth, listwise case deletion was used for the logistic regression models. Last, additional
analyses were also conducted to account for school-level factors, and statistical interactions
incorporating school-level factors and military connection were also conducted; however, no
significant statistical interactions. SPSS 19 was utilized in all univariate, bivariate, and
regression analyses.
Two other multivariate statistical techniques were originally considered for this study.
First, a multivariate linear regression (MLR) approach was considered. MLR was deemed
inappropriate since results would not be meaningful for students at clinical levels of depression,
high levels of well-being, and high levels of suicidal ideation. MLR was also deemed

66
inappropriate since linear regression analyses assume that dependent variables are interval level
and continuous
1
. Second, a two-level hierarchical linear model (HLM) was prepared to assess the
effects of school-level and individual-level variables on mental health; however, preliminary
analyses indicated an intra-class correlation coefficient of .038. Due to the lack of practical
significance for explaining 3.8% of between school variation, logistic regression analyses were
conducted to address the study’s research questions.
Results
Descriptive Statistics of the Sample
In the first step of analysis, descriptive statistics were conducted to determine the
demographic characteristics of the overall sample and demographic differences between military
and non-military students. Table 14 shows that the sample is almost evenly split by male and
female and by grade level (7
th
, 9
th
, and 11
th
grades). The results indicate a diverse sample of
students by race and ethnicity. Almost three-fourths of the sample (72.7%) is non-white.
Hispanic students comprise the largest racial/ethnic group in the sample (50.3%), while Black
students comprise the smallest proportion of students (3.0%). Table 14 also shows that military-
connected students comprised almost 15% of the sample. About 4.2% of the sample is comprised
of students with a military sibling, while 8.8% of the students have a parent in the military.
86.9% of the sample is classified as non-military.
Bivariate analyses were also conducted within three subsamples—non-military students,
students with a parent serving in the military, and students with siblings serving in the military.
When compared to the proportions of students by race and ethnicity in the non-military sample,

1
Multivariate linear regression model were conducted to determine relationships between school
climate and mental health. An aim of this study was to assess differences in mental health
outcomes among demographic groups. Hence, odds ratios from logistic regression models
provided more meaningful interpretations.

67
Table 14 indicates that among students with a military parent, Asian, Black, and Mixed students
represent a larger proportion of total students (12.1%, 7.2%, and 18.1%, respectively), while
Hispanic students represent a smaller proportion (36.9%). Among students with a sibling, white
students are slightly overrepresented (31.7%). When compared to the proportions of students by
grade in the non-military group, Table 5 indicates that among students with a military parent,
students are slightly younger overall. 40.6% of students with a military parent are 7
th
graders,
while 25.8% are 11
th
graders. In contrast, the results suggest that students with siblings serving in
the military are a slightly younger group when compared to non-military students and students
with military parents. Table 14 shows that 25.6% of students with military siblings are 7
th

graders, while 37.5% are in the 11
th
grade.
Table 14. Sample Characteristics for Study #2 (n= 14,943)
Total
N(%)
No one in the
military
(N=12,990)
%
Parent
(N=1396)
%
Sibling
(N=649)
%
Gender
Female
Male

7606 (51.7%)
7181 (48.6%)

51.5
48.5

51.1
48.9

50.9
49.1
Grade
7
th

9
th

11
th

4588 (32.9%)
4908 (35.2%)
4446 (31.9%)

32.3
35.4
32.3

40.6
33.6
25.8

25.6
33.9
37.5
Race/Ethnicity
Asian/AI/HI/AN
Black
White
Mixed Race
Hispanic

1189 (8.2%)
432 (3.0%)
3948 (27.3%)
1606 (11.1%)
7261 (50.3%)

7.9
2.5
27.3
10.3
52.0

12.1
7.2
25.6
18.1
36.9

7.1
4.0
31.7
12.3
44.8

68
Military Connection and Mental Health
The next series of analyses consisted of detailed bivariate analyses. These analyses were
conducted to compare well-being, depression, and suicidal ideation rates among non-military
students, students with a military parent, and students with a military sibling. Overall, the results
show that students with a military sibling reported slightly higher rates of depression and suicidal
ideation than non-military students and students with a military parent. Also, non-military
students have slightly higher rates of well-being than students’ with military parents and students
with military siblings. In general, results from the bivariate analyses suggest the current war
context may affect the mental health outcomes of students with a military sibling to a greater
extent than students with a military parent and non-military students. However, discrepancies in
mental health outcomes among the three groups are unexpectedly small, given the unique
stressors of the present war context.
Overall Well-being and Depression by Military-Connection. Table 15 shows the first
set of bivariate analyses were conducted to compare the overall well-being and depression scores
of the three groups of students. ANOVA results indicate significant mean differences in overall
well-being among the three groups of students are significant, F (1, 11496) = 3.77, p=.01.
Students with a military sibling had slightly lower composite well-being scores than students
with a military parent and non-military students. The average well-being score for non-military
students is 21.02 (SD=5.74). The well-being scores for students with a military parent and non-
military students are slightly lower (M= 20.45, SD= 5.97 and M=20.76, SD=5.76, respectively).
ANOVA results indicate that there are significant mean differences in overall depression among
the three groups of students F (1, 11496) =3.13, p=.04). Students with a military sibling reported
slightly higher composite depression scores than students with a military parent and non-military

69
students. The average depression score for students with a military sibling is 14.26 (SD=5.30),
while the depression score for students with a military parent and non-military students are
slightly lower (M= 13.96, SD= 5.50 and M=13.53, SD=5.16, respectively).
Table 15. Means and Standard Deviations of Mental Health Scales
Total

Non
Military
Parent

Sibling

Well-Being*
Depression*
20.95 (5.77)
13.60 (5.21)
21.02 (5.74)
13.53 (5.16)
20.45 (5.97)
13.96 (5.50)
20.76 (5.76)
14.26 (5.30)
*Significant for differences by military-connected status, p < .05
Composite depression and well-being scores were then dichotomized to compare the
rates of military and non-military students who have a high level of depression and have high
well-being. Table 16 indicates no significant differences among the three groups in terms of
well-being (X
2
=.052, df=2, p=.97). However, significant differences among the three groups
were detected in depression rates (X
2
=9.85, df=2, p<.05). About 7.9% of students with a military
parent reported high levels of depression, slightly higher than students with a military sibling
(6.9%) and non-military students (5.7%).
Table 16. Depression, Well-Being, and Suicidal Ideation by Military Connection (n=14,943)
Total

Non Military
(n= 12,990)
Military Parent
(n= 1,318)
Military
Sibling
(n= 635)
Well-Being
High
Low
Depression*
High
Low
Suicidal Ideation**
Yes
No

1318 (10.7%)
12,265 (89.3%)

681 (5.7%)
10,818 (94.1%)

1,947 (20.5%)
7,573 (79.5%)

1,133 (10.7%)
9,417 (89.3%)

559 (5.7%)
9,329 (94.3%)

1,622 (19.6%)
6,671 (80.4%)

125 (10.7%)
1,046 (89.3%)

86 (7.9%)
1001 (92.1%)

204 (26.1%)
578 (73.9%)

60 (11.0%)
484 (89.0%)

36 (6.9%)
488 (93.1%)

121 (27.2%)
324 (72.8%)
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001

70
Well-being (Past 30 days). More detailed bivariate analyses were conducted to compare
rates of well-being among the three groups of students specifically drawing comparisons within
each well-being item (see Table 17). Chi-square tests of associations were conducted to assess
significant association between military-connection and each of the six well-being items. Table
17 shows that there were significant differences among the three groups of students within five
of the six well-being indicators. Significant associations were detected between military-
connection and feeling happy (X
2
= 41.549, df=8, p<0.05), feeling proud (X
2
= 16.158, df=8,
p<0.05), feeling good about life (X
2
= 23.968, df=8, p<0.05), feeling excited (X
2
= 15.649, df=8,
p<0.05), and feeling strong (X
2
= 16.481, df=8, p<0.05).
Table 17 indicates that non-military students have slightly higher rates of well-being than
students with a military parent and students with a military sibling. About 63.7% of non-military
students reported feeling happy most or all of the time, which is slightly higher than students
with a military sibling (58.6%) and students with a military parent (56.8%). A slightly higher
proportion of non-military students reported feeling proud most or all of the time (44.9%), when
compared to students with a military sibling (44.9%) and students with a military parent (43.0%).
Non-military students also had slightly higher rates of feeling good about life most or all of the
time (57.3%), than students with a military sibling (55.8%) and students with a military parent
(53.1%). Slightly higher rates of feeling excited and feeling strong most or all of the time were
found among non-military students (44.5%, 51.1%, respectively) than rates among students with
a military sibling (41.5%, 48.7%, respectively) and students with a military parent (43.8%,
50.9%, respectively).

71
Table 17. Well-Being by Military Connection Status (n=14,943)
Total

Non
Military
(N=12,990)
Military
Parent
(N=1,318)
Military
Sibling
(N=635)
Feel full of energy
None of the time
A little of the time
Some of the time
Most of the time
All of the time

6.3
12.7
29.2
34.4
17.3

6.3
12.4
29.4
34.5
17.4

7.3
14.7
28.0
34.4
15.5

5.2
14.3
29.2
32.0
19.3
Feel happy**
None of the time
A little of the time
Some of the time
Most of the time
All of the time

4.3
10.2
22.7
42.0
20.8

4.2
9.7
22.4
42.7
21.0

5.9
12.9
24.4
38.7
18.1

3.8
12.8
24.8
36.5
22.1
Feel proud*
None of the time
A little of the time
Some of the time
Most of the time
All of the time

9.2
15.4
29.4
28.6
17.4

9.0
15.2
29.4
29.2
17.2

10.1
16.8
30.2
25.2
17.8

10.0
16.1
29.1
24.9
20.0
Feel good about life**
None of the time
A little of the time
Some of the time
Most of the time
All of the time

7.4
13.3
22.5
32.5
24.4

7.0
13.2
22.5
32.7
24.6

10.4
14.3
22.2
30.0
23.1

7.7
12.8
23.7
32.6
23.2
Feel excited*
None of the time
A little of the time
Some of the time
Most of the time
All of the time

7.6
15.3
32.9
28.1
16.0

7.4
15.1
33.0
28.4
16.1

9.8
16.9
31.9
26.9
14.5

7.7
15.5
32.9
26.2
17.6
Feel strong*
None of the time
A little of the time
Some of the time
Most of the time
All of the time

9.0
14.1
26.1
28.9
21.9

8.9
13.7
26.3
29.3
21.8

10.2
16.3
24.7
27.5
21.2

9.1
15.5
24.5
25.9
25.0
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001

72
Depression (Past 30 days). Table 18 shows that more detailed analyses were also
conducted to compare the depression rates of the three groups of students, specifically examining
each group’s rates within each depression item. Results from this set of analyses suggest that
students with a military sibling had slightly higher rates than students with a military parent and
non-military student within each depression indicator.
Chi-square tests of association were conducted to determine if there were significant
associations between military-connection and each of the six depression indicators. The results
indicate that there are significant associations between military-connection and four of the six
depression indicators, including feeling like everything is hard to do (X
2
= 21.521, df=8, p<0.05),
feeling like nothing makes you happy (X
2
= 37.687, df=8, p<0.05), feeling restless, cannot stay in
one place (X
2
= 21.115, df=8, p<0.05), and feeling like you are not important (X
2
= 24.624, df=8,
p<0.05). About 21.6% of students with a military sibling reported feeling like everything is hard
to do most or all of the time, slightly higher than students with a military parent (19.0%) and
non-military students (16.8%). In addition, 17.3% of students with a military sibling reported
feeling like nothing makes you happy most or all of the time, higher than students with a military
parent (16.4%) and non-military students (12.5%). A higher percentage of students with a
military sibling (30.2%) reported feeling restless than students with a military parent (28.8%)
and non-military students (25.8%). In addition, 21.2% of students with a military sibling reported
most or all of the time feeling like you are not important, higher than students with a military
parent (18.5%) and non-military students (16.2%).

73
Table 18. Depression by Military-Connection Status (n=14,493)
Total

Non
Military
(N=12,990)
Military
Parent
(N=1,318)
Military
Sibling
(N=635)
Feel like everything is hard to do*
None of the time
A little of the time
Some of the time
Most of the time
All of the time

20.6
37.3
24.8
11.0
6.2

20.5
37.9
24.9
10.9
5.9

21.5
35.6
23.9
11.6
7.4

21.1
31.2
26.1
13.4
8.2
Feel like nothing makes you happy**
None of the time
A little of the time
Some of the time
Most of the time
All of the time

47.2
24.4
15.2
8.2
4.9

48.2
24.2
15.1
7.8
4.7

42.4
25.5
15.7
10.7
5.7

40.9
25.9
15.8
10.6
6.7
Feel restless, cannot stay in one place*
None of the time
A little of the time
Some of the time
Most of the time
All of the time

24.6
24.8
24.3
15.5
10.8

24.9
25.2
24.2
15.3
10.5

23.3
23.6
24.3
16.7
12.1

21.7
20.7
27.3
16.2
14.0
Feel nervous
None of the time
A little of the time
Some of the time
Most of the time
All of the time

19.7
36.6
27.5
10.8
5.4

19.8
36.6
27.7
10.7
5.2

19.7
36.3
26.0
11.5
6.5

18.6
36.6
26.7
11.1
7.0
Feel you do not have much hope
None of the time
A little of the time
Some of the time
Most of the time
All of the time

49.1
21.8
15.4
8.3
5.4

49.4
21.8
15.2
8.2
5.3

47.6
22.4
15.7
8.8
5.6

45.8
20.4
17.7
9.0
7.1
Feel like you are not important*
None of the time
A little of the time
Some of the time
Most of the time
All of the time

44.3
22.9
16.1
9.6
7.1

45.0
22.9
15.9
9.4
6.8

40.5
24.0
17.0
10.0
8.5

39.8
21.1
18.0
12.1
9.1
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001

74
Suicidal Ideation (Past Year). A bivariate analysis was then conducted to compare the
rates of suicidal ideation of each group of students (see Table 16). A significant bivariate
association was found between military-connection and suicidal ideation (X
2
= 31.753, df=8,
p=.000). Students with a military sibling (27.6%) reported higher rates of suicidal ideation than
students with a military parent (26.2%) and non-military students (19.6%).
In general, the small differences in mental health outcomes among the three groups stand
in stark contrast to recent studies comparing the mental health outcomes of military-connected
and non-military youth. Results from these studies found larger differences in depression and
anxiety between non-military and military-connected youth. In addition, students with a military
sibling had the most adverse mental health outcomes. This finding expands on recent findings
from a qualitative study on the psychological and emotional challenges that siblings of military
service members experience in the household context.
School Climate and Mental Health
Correlational Analyses. As seen in Table 19, analyses were conducted to examine the
relationship between multiple components of school climate and mental health. Bivariate
analyses were first conducted to determine relationships between multiple components of school
climate and mental health (i.e. well-being, depression, suicidal ideation). As expected, results
from the correlational analyses indicate significant relationships between school climate and
mental health. However, correlations between multiple components of school climate and mental
health were smaller in magnitude than expected.
In the realm of belonging, the results suggest that belonging and caring relationships are
associated with higher levels of well-being and lower levels of depression and suicidal ideation.
As seen in Table 19, there is a significant and moderate negative correlation between belonging

75
and depression (r= -.21) and between belonging and suicidal ideation (r=-.17). As expected, there
is a significant positive correlation between belonging and well-being (r=.30). Table 19 also
shows significant negative correlations between caring relationships and two negative mental
health indicators, depression (r=-.17) and suicidal ideation (r=-.13). As expected, there is a
significant positive correlation between caring relationships and well-being (r=.30).
In the case of meaningful participation, the results show that higher levels of meaningful
participation are associated with lower levels of depression and suicidal ideation and a higher
degree of well-being. Table 19 indicates significant negative correlations between meaningful
participation and depression (r=-.11) and suicidal ideation (r=-.10). A significant moderate
positive correlation was found between meaningful participation and well-being (r=.28).
Also, as expected, the results show that more risky behavior disapproval is associated
with lower levels of depression and suicidal ideation as well as a higher degree of well-being. A
significant negative relationship between risky behavior disapproval and depression (r=-.16) was
found, while a significant positive relationship between risky behavior disapproval and well-
being was detected (r=.17). Table 19 shows significant associations between lack of family
respect and each mental health category. Significant positive associations were detected between
lack of respect for family and depression (r=.14) and suicidal ideation (r=.03). Surprisingly, no
significant correlations were found between safety and each of the mental health variables.
Results from the correlation analyses suggest that a supportive school climate can
potentially facilitate positive mental health outcomes among students. However, contrary to
expectations, the results indicate that safety does not play a significant role in mental health. In
addition, the smaller than expected correlation coefficients between school climate and mental

76
health contradict decades of theoretical work outlining the significance of school climate in the
emotional and psychological well-being of adolescents.
Table 19. Correlational Analyses for Study #2
1 2 3 4 5 6 7 8 9 10
1. Belonging -- .41**

.24** .36** .07** .22** -.03** -.21** .30** -.17**
2. Relationships .41** -- .08** .43** .19** .26** -.02* -.17** .29** -.13**
3. Safety .24** .08** -- .01 .03** .08** .00 .00 -.01 .01
4. Meaningful
Participation
.36** .43** .01 -- .14** .19** .02 -.11** .28** -.10**
5. Respect for
Family
.07** .19** .03** .14** -- .13** .04** .14** .10** .03*
6. Risky
Behavior
.22** .26** .08** .19** .13** -- .01 -.16** .17** -.11
7. Deployment -.03** -.02* -.01 .02 .04** .01 -- .08** -.01 .07**
8. Depression -21** .17** .00 -.11** .14** -.16** .08** -- -.19** .31**
9. Well-Being .30** .29** 0.01 .28** .10** .17** -.01 -.19** -- -.29**
10 Suicidal
Ideation
-.17** -.13** .01 -.10** .03* -.11** .07** .31** -.29** --
*p < .05
**p <.001

Multivariate Analyses. As seen in Tables 20-24, multivariate analyses were conducted
to further evaluate relationships between school climate and mental health, while accounting for
other covariates present in a military-connected school such as student demographics (gender,
grade level, race and ethnicity, and military connection) and the experience of deployment.
Table 20 shows results for the logistic regression model of well-being, depression, and
suicidal ideation by school climate. Overall, the results indicate that supportive school climates
are associated with an increased probability of well-being and a reduced likelihood of depression
and suicidal ideation. However, similar to results in the correlational analyses, the magnitude of
odds ratios in the following logistic regression is smaller than expected.

77
As seen in Table 20, belonging and caring relationships significantly predicted all three
mental health outcomes. A higher degree of belonging is significantly associated with an
increased likelihood of well-being (OR = 1.26; 95% CI = 1.15 – 1.37), a decreased likelihood of
depression (OR = .70; 95% = .63-.77), and a decreased likelihood of having suicidal ideation
(OR = .71; 95% CI = .66-.76). A higher degree of caring relationships with teachers and adults at
school is significantly associated with well-being (OR = 1.25; 95% CI
= 1.12 – 1.40), a
decreased likelihood of negative mental health (OR = .76; 95% CI = .67 - .85) and a decreased
likelihood of suicidal ideation (OR = .79; 95% CI = .72-.86).
As seen in Table 20, the results also show that risky behavior disapproval significantly
predicts both depression and suicidal ideation. An increasing level of risky behavior disapproval
is associated with a 43% decreased odds of having depression (OR = .57; 95% CI = .50 - .65)
and a 31% decreased odds of suicidal ideation (OR = .69; 95% CI = .63-.76). The results also
suggest that lack of respect for family is related to higher rates of depression and suicidal
ideation. The results indicate that a low degree of respect for family is associated with 50%
increased odds of depression (OR = 1.50; 95% CI – 1.36-1.66) and a 21% increased likelihood of
suicidal ideation (OR = 1.21; 95% CI = 1.13-1.29).
Meaningful participation predicted both well-being and depression. Unexpectedly, lack
of meaningful participation is associated with a 55% increased likelihood of well-being
(OR=1.55; 95% CI=1.42-1.69) and a 15% increased odds of depression (OR=1.15: 95%
CI=1.02-1.30). In the case of safety, the results show that there are associations between safety
and mental health, though small in magnitude. A lower degree of safety is associated with a
slightly decreased likelihood of well-being (OR = .84; 95% CI = .78- .90) and only a 9%
increased odds of having suicidal ideation (OR = 1.09; 95% = 1.03-1.16). In addition, Table 20

78
shows that disapproval of risky behaviors significantly predicted depression and suicidal
ideation. A higher degree of disapproval of risky behaviors is significantly associated with a 43%
decreased likelihood of depression (OR = .57; 95% CI=.50-.65) and a 31% decreased odds of
suicidal ideation (OR = .69; 95% CI= .63-.76).
Table 20 also shows associations between demographic variables and mental health. No
significant associations between military connection and well-being were found. In the case of
negative mental health outcomes, when compared to non-military students, having a military
parent is associated with an increased likelihood of depression (OR = 1.34; 95% CI =1.02-1.75)
and having suicidal ideation (OR = 1.31; 95% = 1.08-1.59). Also, having a military sibling is
associated with an increased likelihood of having suicidal ideation (OR = 1.50; 95% CI = 1.18 –
1.91). In the case of race and ethnicity, the most concerning finding involved analyses on Asian
students and mental health. The results suggest that Asian students have higher rates of
depression in comparison to White students. When compared to being white, being Asian is
associated with a decreased likelihood of well-being (OR = .72; 95% CI = .54-.96) and increased
likelihood of depression (OR = 1.43; CI = 1.05 - 1.95) and having suicidal ideation (OR = 1.26;
CI = 1.02-1.56). Also, while the odds ratios are not significant, Black students are more likely
than white students to have well-being (OR=1.46; 95% CI=.99-2.14) and are also more likely to
have depression than white students (OR=1.33; 95% CI=.82-2.19). Gender differences indicate
that male students have better mental health outcomes than female students. Male students are
almost twice as likely to have well-being (OR = 1.95; 95% CI = 1.70 - 2.23) than female
students.

79
Table 20. Logistic Regressions of Mental Health by Demographic and School Climate Variables
among all students (n=14,943)
Predictors Well-Being
OR (95% CI)
Depression
OR (95% CI)
Suicidal Ideation
OR (95% CI)
Constant
Grade
.04**
.82** (.78-.85)
.37**
.95 (.90-1.01)
3.36**
.90** (.85-.95)
Gender
Female (reference)
Male

1.00
1.94** (1.69-2.22)

1.00
1.03 (.86-1.24)

1.00
.57** (.51-.64)
Race/Ethnicity
White (reference)
Asian/AI/AN/PI/HI
Black
Mixed
Hispanic

1.00
.72* (.54-.96)
1.46 (.99-2.14)
1.05 (.83-1.32)
1.19* (1.02-1.39)

1.00
1.43* (1.05-1.95)
1.33 (.82-2.19)
1.28 (.95-1.72)
.99 (.79-1.23)

1.00
1.26* (1.02-1.56)
1.02 (.72-1.44)
1.28* (1.05-1.56)
.90 (.79-1.04)
Military-
Connection
None (reference)
Parent
Sibling

1.00
1.02 (.81-1.28)
1.13 (.83-1.55)

1.00
1.34* (1.02-1.75)
1.28 (.88-1.88)

1.00
1.31* (1.08-1.59)
1.50** (1.18-1.91)
School Climate
Belonging
Relationships
Safety
Meaningful Part.
Respect for Family
Risky Behavior

1.26** (1.15-1.37)
1.25** (1.12-1.40)
.84** (.78-.90)
1.55** (1.42-1.69)
1.01 (.93-1.09)
.96 (.86-1.07)

.70** (.63-.77)
.76** (.67-.85)
1.03 (.95-1.12)
1.15* (1.02-1.30)
1.50**(1.36-1.66)
.57**(.50-.65)

.71** (.66-.76)
.79** (.72-.86)
1.09* (1.03-1.16)
.95 (.88-1.02)
1.21** (1.13-1.29)
.69** (.63-.76)
Model Evaluation
% Correctly
Predicted
-2 Log Likelihood
Cox & Snell R
2
Nagelkerke R
2
Block X
2
(df)
Model X
2
(df)
Goodness of fit
Homer/Lemeshow

90.2%
6268.78
.05
.10
488.22 (14)
488.22 (14)

8.35

94.5%
3961.58
.03
.08
271.03 (14)
271.03 (14)

5.5

79.3%
7458.60
.06
.10
486.46 (14)
486.46 (14)

12.40

In general, the results show that school climate can influence the mental health outcomes
of students. In light of the breadth of theoretical literature advocating the protective effects of
school climate on mental health, odds ratios for school climate are smaller in magnitude than

80
expected. This unexpected finding suggests that other factors, in addition to school climate and
student demographics, could account for the mental health outcomes of students.
Deployment and Mental Health
Since this study takes place in the context of military-connected schools, it is expected
that military-related stressors, especially the deployment of a family member, may also impact
mental health outcomes. As seen in Table 21, the next step of analyses examined the role of
deployment and school climate in the mental health outcomes of students. Analyses were
conducted to determine the associations between deployments (single or multiple) and mental
health. Overall, deployment was a significant predictor of depression and suicidal ideation,
however, did not significantly predict well-being. Experiencing a multiple deployment is
associated with depression outcomes (OR =1.42; 95% CI = 1.27-1.59). Single and multiple
deployments were each associated with an increased likelihood of having suicidal ideation. In
comparison to the effect of single deployments, the results show that multiple deployments have
a greater effect on suicidal ideation (OR = 1.41; 95% CI = 1.20-1.65).
Despite the inclusion of deployment in the logistic regression model, multiple
components of school climate are still significantly associated with well-being, depression, and
suicidal ideation. As seen in Table 21, belonging and caring relationships continue to
significantly predict all three mental health outcomes, and the odds ratios are similar in
magnitude to the non-deployment logistic regression model seen in Table 21. A sense of
belonging is associated with an increase in likelihood of well-being (OR = 1.23; 95% CI = 1.12 -
1.36) and a decreased likelihood of depression outcomes (OR = .68; 95% CI = .61-.76) and
suicidal ideation (OR = .71; 95% CI = .66-.76). Also, caring relationships are associated with an

81
increased odds of having well-being (OR = 1.25; 95% CI = 1.11-1.41) and a decreased odds of
depression (OR = .75; 95% CI = .65-.86) and suicidal ideation (OR = .81; 95% CI = .73-.88).
In the deployment model (see Table 21), the effects of level of safety and meaningful
participation on mental health outcomes are similar in magnitude when compared to the non-
deployment model. Table 21 indicates that an unsafe school environment decreases the
likelihood of well-being (OR = .88; 95% CI = .81-.94) and slightly increases the likelihood of
suicidal ideation (OR=1.08; 95% CI=1.01-1.15). In addition, meaningful participation is
associated with an increased likelihood of well-being (OR = 1.60; 95% CI = 1.45-1.76) and
depression (OR= 1.15; 95% CI=1.02-1.31). Table 21 also indicates that respect for family and
risky behavior perceptions continue to be significant predictors of depression and suicidal
ideation, while the odds ratios are similar in magnitude when compared to those in the non-
deployment model. A lack of respect for family is associated with an increased likelihood of
depression (OR = 1.42; 95% CI = 1.27-1.59) and suicidal ideation (OR = 1.22; 95% CI = 1.14-
1.31). Also, a higher disapproval of risky behavior is associated with a decreased likelihood of
depression (OR = .59; 95% CI = .51-.69) and suicidal ideation (OR = .67; 95% CI = .61-.74).
Overall, deployment plays a significant role in depression and suicidal ideation among
students in military-connected schools. Moreover, as seen in Table 21, odds ratios for school
climate in these analyses were similar in magnitude to odds ratios seen in the non-deployment
model (see Table 20). This similarity suggests that school climate still plays a significant role in
promoting well-being and curbing rates depression and suicidal ideation, even though students
are experiencing single and multiple deployments of a family member.

82
Table 21. Logistic Regressions of Mental Health by Demographics, School Climate and
Deployment (n=14,943)
Predictors Well-Being
OR (95% CI)
Depression
OR (95% CI)
Suicidal Ideation
OR (95% CI)
Constant
Grade
.03**
.81** (.77-.85)
.45*
.94* (.88-.99)
3.29**
.91* (.86-.97)
Gender
Female (reference)
Male

1.00
1.93** (1.67-2.23)

1.00
1.06 (.87-1.28)

1.00
.55**(.49-.63)
Race/Ethnicity
White (reference)
Asian/AI/AN/PI/HI
Black
Mixed
Hispanic

1.00
.79 (.58-1.07)
1.65*(1.10-2.47)
1.11 (.87-1.42)
1.27* (1.07-1.51)

1.00
1.52* (1.09-2.12)
1.20 (.70-2.06)
1.18 (.87-1.63)
.90 (.71-1.41)

1.00
1.26*(1.01-1.59)
1.08 (.75-1.56)
1.21 (.98-1.50)
.90 (.78-1.04)
Deployments
One
Two or more

1.07 (.84-1.36)
1.14 (.94-1.37)

1.02 (.73-1.42)
1.37* (1.08-1.73)

1.33* (1.08-1.63)
1.41** (1.20-1.65)
School Climate
Belonging
Relationships
Safety
Meaningful Part.
Respect for Family
Risky Behavior

1.21** (1.10-1.33)
1.24**(1.10-1.40)
.87** (.81-.94)
1.59** (1.44-1.75)
.99 (.92-1.09)
1.02 (.90-1.15)

.68** (.61-.76)
.76** (.66-.87)
1.04 (.95-1.14)
1.15* (1.02-1.31)
1.43** (1.28-1.59)
.57** (.49-.67)

.71** (.66-.76)
.82**(.75-.90)
1.08* (1.01-1.15)
.92*(.84-.99)
1.21**(1.12-1.30)
.68** (.61-.75)
Model Evaluation
% Correctly Predicted
-2 Log Likelihood
Cox & Snell R
2
Nagelkerke R
2
Block X
2
(df)
Model X
2
(df)
Goodness of fit
Homer/Lemeshow

90.3%
5375.93
.05
.10
422.87 (14)
422.87 (14)

14.12

94.8%
3340.55
.03
.08
230.33 (14)
230.33 (14)

3.04

79.4%
6442.86
.06
.10
439.38 (14)
439.38 (14)

15.15
`*Significant for differences by military-connected status, p < .05
**Significant for differences by military-connected status, p <.00

School Climate and Mental Health among Military and Non-Military Students
A large body of research suggests that school climate has uniquely strong effects on
student populations experiencing tremendous stress in social contexts outside of the school (e.g.
household, community). To test this assumption on military students, the next steps of analyses

83
examined relationships between school climate and mental health separately for military students
(students with a military parent and students with a military sibling) and non-military students.
Overall, the results show that the effects of school climate on mental health are similar among
military and non-military students.
Well-Being. Table 22 shows logistic regressions of well-being by school climate among
non-military and military students. An increasing level of belonging is associated with increased
likelihood of having well-being among non-military students (OR = 1.20; 95% CI= 1.08-1.33).
Similar to non-military students, belonging is associated with a 26% increased likelihood of
having well-being among military students (OR = 1.26; 95% CI = .99-1.61). As expected, among
non-military students, a higher degree of caring relationships is associated with a 24% increase
in having well-being (OR=1.24; 95% CI = 1.09-1.41). Although the results for caring
relationships present a non-significant result, the odds ratio for caring relationships among
military students is similar in magnitude to the odds ratio generated for non-military students
(OR=1.28; 95% CI .95-1.74). This result indicates that an increasing level of caring relationships
is associated with a 28% increased likelihood of having well-being among military students.
Table 22 also indicates that the effects of meaningful participation and safety on well-
being are similar in magnitude for non-military and military students. Among non-military
students, lack of safety is associated with a 11% decreased probability in having well-being (OR
= .89; 95% CI = .82-.97). Similarly, among military students, lack of safety is associated with a
21% decreased odds of having well-being (OR= .79; 95% CI = .65-.95). In the realm of
meaningful participation, the odds ratios for both non-military students and military students are
also similar in magnitude. Among non-military students, an increasing level of meaningful
participation is associated with a 62% increased probability of having well-being (OR = 1.62;

84
95% CI = 1.46-1.81). In the case of military students, increasing level of meaningful
participation is associated with a 38% increased odds of having well-being (OR = 1.38; 95% CI
= 1.06-1.78). The separate multivariate analyses indicate that respect for family and risky
behavior approval do not predict likelihood of having well-being.
Table 22. Logistic Regressions of Well-Being by Demographics and School Climate among
Non-Military Students and Military Students
Predictors Well-Being
OR (95% CI)

Non-Military Military
Constant
Grade
.02**
.81** (.78-.85)
.07**
.77** (.74-3.41)
Gender
Female (reference)
Male

1.00
1.87**(1.60-2.22)

1.00
2.24** (1.52-3.32)
Race/Ethnicity
White (reference)
Asian//AI/AN/PI/HI
Black
Mixed
Hispanic

1.00
.78 (.54-1.07)
1.61 (.99-2.61)
1.20 (.92-1.57)
1.31* (1.09-1.57)

1.00
.88 (.44-1.75)
1.59 (.74-3.41)
.79 (.43-1.44)
1.07 (.67-1.70)
Deployment
None (reference)
One
Two or More

1.00
1.21 (.92-1.57)
1.14 (.89-1.46)

1.00
.67 (.36-1.25)
1.07 (.67-1.71)
Military
Military Parent
Military Sibling

--
--

1.00
1.12 (.73-1.73)
School Climate
Belonging
Relationships
Safety
Meaningful Part.
Respect for Family
Risky Behavior

1.20** (1.08-1.33)
1.24** (1.09-1.41)
.89* (.82-.97)
1.62** (1.46-1.81)
.99 (.90-1.08)
1.04 (.91-1.19)

1.26* (.99-1.61)
1.28 (.95-1.74)
.79* (.65-.95)
1.38* (1.06-1.78)
1.09 (.87-1.36)
.94 (.68-1.29)
(Table 22 continued)

85
(Table 22 continued)
Model Evaluation
% Correctly Predicted
-2 Log Likelihood
Cox & Snell R
2
Nagelkerke R
2
Block X
2
(df)
Model X
2
(df)
Goodness of fit
Homer/Lemeshow

90.4%
4588.41
.05
.10
364.61 (14)
364.61 (14)

10.21

90.1%
776.972
.05
.11
71.59 (15)
71.59 (15)

21.53
`*Significant for differences by military-connected status, p < .05
**Significant for differences by military-connected status, p <.001

Depression. Table 23 shows separate logistic regressions of depression by school climate
among military and non-military students. An increasing level of belonging is associated with a
decreased likelihood of depression for both non-military and military students. Odds ratios were
similar in magnitude for both non-military students (OR=.66; 95% CI = .66-.75) and military
students (OR=.77; 95% CI = .60-.99). In the case of caring relationships, an increasing degree of
caring relationships is associated with a 23% decreased probability of depression among non-
military students (OR=.77; 95% CI=.66-.89). Similarly, an increasing degree of caring
relationships is associated with a 29% decreased likelihood of depression among military
students (OR=.71; 95% CI=.52-.98).
The effects of meaningful participation on depression were also similar in magnitude
among military and non-military students, however, unexpectedly, increasing levels of
meaningful participation are associated with an increased likelihood of depression among non-
military and military students. Among non-military students, the results show a non-significant
result, however, an increasing degree of meaningful participation is associated with a 15%
increased odds of depression (OR=1.15; 95% CI = .99-1.33). Among military students, the
results also show a non-significant result, however, the magnitude of the odds ratio is similar to
the odds ratio for non-military students (OR=1.20; 95% CI = .89-1.60). In the case of risky

86
behavior approval, an increasing degree of disapproval of risky behaviors is associated with a
44% decreased odds of depression among non-military students (OR=.56; 95% CI = .48-.66).
Similarly, among military students, an increasing degree of disapproval of risky behaviors is
associated with a 38% decreased odds of depression (OR=.62; 95% CI=.43-.88).
In the case of respect for family, the magnitude of odds ratios for both non-military and
military students were also similar. Among non-military students, lack of respect for family is
associated with a 40% increased odds of depression (OR=1.40; 95% CI=1.25-1.58). The effect of
respect for family on depression is slightly stronger for military students. Among military
students, lack of respect for family is associated with a 48% increased odds of depression
(OR=1.48; 95% CI=1.14-1.92).
Table 23. Logistic Regressions of Depression by Demographics and School Climate among Non-
Military Students and Military Students
Predictors Depression
OR (95% CI)

Non-Military Military
Constant
Grade
.56
.94 (.88-1.00)
.21*
.94 (.82-1.08)
Gender
Female (reference)
Male

1.00
.99 (.82-1.21)

1.00
1.52 (.97-2.40)
Race/Ethnicity
White (reference)
Asian//AI/AN/PI/HI
Black
Mixed
Hispanic

1.00
1.46* (1.02-2.08)
1.36 (.71-2.58)
1.21 (.83-1.77)
1.02 (.78-1.21)

1.00
1.17 (.59-2.31)
.76 (.28-2.11)
1.01 (.54-1.89)
.51 (.28-.93)
Deployment
None (reference)
One
Two or More

1.00
1.02 (.69-1.51)
1.21 (.88-1.66)

1.00
1.07 (.50-2.33)
1.50 (.82-2.38)
Military
Military Parent
Military Sibling

--
--

1.00
.85 (.50-1.44)
(Table 23 continued)

87
(Table 23 continued)
School Climate
Belonging
Relationships
Safety
Meaningful Part.
Respect for Family
Risky Behavior

.66** (.59-.75)
.77** (.66-.89)
1.02 (.92-1.13)
1.15* (.99-1.33)
1.40** (1.25-1.58)
.56** (.48-.66)

.77* (.60-.99)
.71* (.52-.98)
1.10 (.89-1.35)
1.20 (.89-1.60)
1.48** (1.14-1.92)
.62* (.43-.88)
Model Evaluation
% Correctly Predicted
-2 Log Likelihood
Cox & Snell R
2
Nagelkerke R
2
Block X
2
(df)
Model X
2
(df)
Goodness of fit
Homer/Lemeshow

95%
2748.24
.03
.08
194.18 (14)
194.18 (14)

10.72

93.1%
689.60
.03
.09
44.14 (15)
44.14 (15)

5.14
`*Significant for differences by military-connected status, p < .05
**Significant for differences by military-connected status, p <.001

Suicidal Ideation. Table 24 shows logistic regressions of suicidal ideation by school
climate among military and non-military students. In the case of belonging, odds ratios for non-
military and military students were similar in magnitude. Among non-military students, an
increasing level of belonging is associated with a 29% decreased likelihood of suicidal ideation
(OR=.71; 95% CI=.65-.77). Similarly, among military students, an increasing degree of
belonging is associated with a 26% decreased odds of suicidal ideation (OR=.74; 95% CI=.62-
.89). The effect of caring relationships on suicidal ideation among military students is slightly
stronger than the effect seen among non-military students. An increasing level of caring
relationships is associated with a 17% decreased likelihood of suicidal ideation (OR=.83; 95%
CI=.75-.91). Among non-military students, an increasing degree of caring relationships is
associated with a 26% decreased likelihood of suicidal ideation (OR=.74; 95% CI=.59-.93).
A lack of school safety has slight effects on suicidal ideation for both military and non-
military students. Among non-military students, lack of safety is associated with only an 10%

88
increased likelihood of suicidal ideation (OR=1.10; 95% CI=1.02-1.18). Similarly, among
military students, lack of safety is associated with only a 3% increased likelihood of suicidal
ideation (OR=1.03; 95% CI=.89-1.19). Odds ratios for lack of respect for family in both non-
military and military populations are similar in magnitude. Among non-military students, lack of
respect for family is associated with a 20% increased likelihood of suicidal ideation (OR=1.20;
95% CI=1.11-1.30). Among military students, lack of respect for family is associated with a 24%
increased likelihood of suicidal ideation (OR=1.24; 95% CI= 1.04-1.48).
The effect of risky behavior disapproval on suicidal ideation among non-military students
is also similar to the effect seen in military students. Among non-military students, increasing
risky behavior disapproval is associated with a 33% decreased probability of suicidal ideation
(OR = .67; 95% CI = .60-75). Among military students, the results show that increasing risky
behavior disapproval is associated with a 25% decreased probability of suicidal ideation
(OR=.75; 95% CI=.58-.97). In the case of meaningful participation, the odds ratios for both non-
military and military student populations are small in magnitude and non-significant. A higher
degree of meaningful participation is associated with a slight decreased odds of suicidal ideation
among non-military students (OR = .88; 95% CI=.81-.97), but a slight increased likelihood of
suicidal ideation among military students (OR = 1.09; 95% CI=.90-1.33).
Table 24. Logistic Regressions of Suicidal Ideation by Demographics and School Climate among
Non-Military Students and Military Students (n=14,943)
Predictors Suicidal Ideation
OR (95% CI)

Non-Military Military
Constant
Grade
3.45**
.92* (.86-.98)
2.43
.93 (.81-1.08)
Gender
Female (reference)
Male

1.00
.52** (.48-.62)

1.00
.73 (.54-.99)
(Table 24 continued)

89
(Table 24 continued)
Race/Ethnicity
White (reference)
Asian//AI/AN/PI/HI
Black
Mixed
Hispanic

1.00
1.17 (.93-1.49)
1.07 (.70-1.64)
1.26* (1.01-1.57)
.86 (.74-1.00)

1.00
1.58* (.93-2.66)
1.10 (.59-2.10)
1.34 (.85-2.11)
1.12 (.76-1.62)
Deployment
One
Two or more

1.34* (1.09-1.76)
1.46** (1.18-1.80)

.89 (.56-1.43)
1.09 (.75-1.59)
Military
Military Parent
Military Sibling

--
--

1.00
1.20 (.86-1.66)
School Climate
Belonging
Relationships
Safety
Meaningful Part.
Respect for Family
Risky Behavior

.71** (.65-.77)
.83** (.75-.91)
1.10* (1.02-1.18)
.88* (.81-.97)
1.20** (1.11-1.30)
.67** (.60-.75)

.74** (.62-.89)
.74* (.59-.93)
1.03 (.89-1.19)
1.09 (.90-1.33)
1.24* (1.04-1.48)
.75* (.58-.97)
Model Evaluation
% Correctly Predicted
-2 Log Likelihood
Cox & Snell R
2
Nagelkerke R
2
Block X
2
(df)
Model X
2
(df)
Goodness of fit
Homer/Lemeshow

80.6%
5395.22
.07
.10
392.73 (14)
392.73 (14)

17.73

75%
1042.64
.05
.08
51.17 (15)
51.17 (15)

4.39
`*Significant for differences by military-connected status, p < .05
**Significant for differences by military-connected status, p <.001

School-Level Effects
Analyses were conducted to assess associations between school-level variables and
mental health outcomes. Contrary to expectations, school military concentration and academic
achievement did not significantly predict well-being, depression, and suicidal ideation. However,
school level poverty significantly predicts depression. The results show that a higher degree of
poverty is associated with an increased likelihood of depression (OR = .49; 95% CI = .25-.96).

90
Discussion
Using a population sample of middle and high school students in eight military-connected
school districts, this study examines rates of well-being, depression, and suicidal ideation among
students by military connection (non-military students, students with a military parent, and
students with a military sibling). Relationships between school climate and mental health among
military and non-military students, accounting for student demographics and the experience of
deployment were also examined. The results of this study indicate that multiple components of
school climate promote higher rates of well-being and curbs rates of depression and suicidal
ideation among students in military-connected schools, even when accounting for students’
experiences with the deployment of a family member. In addition, logistic regression analyses
conducted show similar effects of school climate on mental health among non-military students,
students with a military parent and students with a military sibling.
Well-Being, Depression and Suicidal Ideation Rates
The bivariate analyses in this study indicate that students with a military sibling have
slightly higher rates of depression than students with military parents, while non-military
students had the highest rates of well-being. This finding supports previous research on the
mental health of military children. These studies have found that the rates of depressive and
psychiatric symptoms among military children and adolescents are significantly higher than
civilian children in the Iraq and Afghanistan war context (Chartrand & Seigel, 2007; Gorman et
al., 2009; Hoshmand & Hoshmand, 2007; Huebner et al., 2009). The greatest discrepancies
between students with a military parent and non-military students were found in suicidal
ideation. This concerning finding may be attributed to the unique military stressors that military
children have experienced during the Iraq and Afghanistan wars, which includes multiple and

91
prolonged parental deployments. In addition, unlike previous wars, a large proportion of military
service members are from the Reservist force, have less access to mental health services on
military installations and reside in civilian communities where military-specific social supports
are not necessarily available (Bowen, Mancini, Martin, Ware, & Nelson, 2003; MacDermid,
Samer, Schwarz, Nishida, & Nyaronga, 2008; Rohall, Weschler, & Segal, 1999).
Moreover, this finding further develops knowledge about the psychological stressors that
adolescents experience when a sibling is serving in the military (Rodriquez & Margolin, 2011).
A recent qualitative study suggested that the deployment of a sibling was associated with stress
and anxiety among family members, shifts in household roles and responsibilities, alienation
from peers and other social supports, and sadness from temporary separation (Rodriguez &
Margolin, 2011). From a psychological perspective, Rodriguez and Margolin (2011) have
theorized that the sibling relationship is potentially the most enduring family bond that
adolescents experience and that stable and positive sibling relationships are linked to enhanced
cognitive, emotional, moral, and psychosocial outcomes. In addition, during times of family and
household stress, children seek out emotional and social support. In severely stressful cases,
sibling relationships may provide a buffering effect when household and financial stability is at
risk.
School Climate and Mental Health
In this study, multiple components of school climate were significantly associated with
reduced rates of depression and suicidal ideation and higher rates of well-being. This finding
supports a large body of research that a caring and supportive school environment can promote
well-being and curb depression and suicidal ideation. This study is unique since school climate
has multiple constructs—belonging, caring relationships, meaningful participation, risky

92
behavior disapproval, and safety. Logistic regression analyses yielded detailed information about
the relationships between each component of school climate and the three indicators of mental
health.
Both belonging and caring relationships had significant associations with well-being,
depression and suicidal ideation. This finding supports research that has found students who feel
connected to their school communities as well as supported and cared for by teachers and other
school staff have enhanced well-being and motivation and lower likelihood of exhibiting
depressive symptoms and thoughts of suicide. One explanation for this finding is that school-
based approaches that promote belonging such as school assemblies, activities, clubs, cultural
displays (i.e. bulletin boards), and extracurricular activities can help create a sanctuary for
students experiencing psychological strain of household and community stressors (Chandra et
al., 2010). In a recent study, McGuire and colleagues (2010) found that school efforts to make a
classroom or school culture accepting of transgendered students influenced the way
transgendered students felt emotionally and socially connected to other students and school staff.
At the same time, these “transgender friendly” approaches transformed the cultural norms of
students and staff, making them more welcoming of transgendered students.
This study has also found that risky behavior disapproval has significant associations
with negative mental health outcomes (depression, suicidal ideation). These findings support
research identifying associations between substance use and negative mental health problems
(depression, suicidal ideation) among adolescents. School climate intervention programs that aim
to prevent substance use among adolescents are utilized in public schools throughout the country,
and have been shown to reduce substance use among middle and high school students.

93
Some findings were unexpected and make unique contributions to the school climate
literature. First, the weakest associations were found between safety and each mental health
indicator. This challenges several studies that have found significant relationships between
perceptions of safety and mental health outcomes among middle and high school students. This
finding suggests that other components of school climate (e.g. belonging, caring relationships)
may be more effective in promoting positive mental health and reducing rates of depression and
suicidal ideation. School climate interventions that focus exclusively on safety may prove to be
ineffective in addressing the mental health as well as social and emotional needs of middle and
high school students. Second, this study found that meaningful participation had a positive
relationship with depression, though small in magnitude. It is well-established in past studies that
school engagement leads to enhanced self-concept, self-esteem and general well-being.
However, it may be expected that high achieving students in competitive middle and high
schools who have high levels of engagement and participation in a school community may also
experience much stress and anxiety and if left unaddressed by school counselors and other
mental health professionals, depression and suicidal ideation.
Another major significant finding is that single and multiple deployments were
significantly related to each mental health indicator, even when accounting for multiple
components of school climate. As expected, odds ratios for associations between multiple
deployments and each mental health indicator were larger in magnitude than experiencing a
single deployment. This finding supports studies conducted during the Iraq and Afghanistan
wars, indicating that military-connected adolescents experience tremendous stress and anxiety as
a result of coping with parental separation, constantly shifting roles and responsibilities in the
household, fear of death and injury, and the mental health of a left-behind parent or caregiver

94
(Huebner et al., 2007; Rodriguez & Margolin, 2011). Also, unique to the current war context,
military children typically experience multiple deployments of family members. This study was
able to measure multiple deployments, given the survey item asking students to report number of
deployments of family members in the past 10 years. These findings also present a unique
contribution to the current psychological literature on military children and school climate, since
the adverse effects of deployment still persist despite the presence of a supportive school climate.
While deployment remains a strong predictor of mental health outcomes, it should be
noted that relationships between school climate and each mental health category are still
significant. In fact, the magnitude of odds ratios between each school climate component and
mental health category in the deployment model (see Table 21) are still similar to those seen in
the non-deployment model (see Table 20). This finding supports theoretical work and studies
that have found that a supportive school environment still significantly influences mental health
outcomes, even among high-stress populations.
School Climate and Mental Health among Military and Non-Military Students
The results of this study also deepen our knowledge of how the social and emotional
qualities of a school environment influence the mental health outcomes of military children.
Recent qualitative studies on military-connected schools have found that on one hand, some
school staff are struggling to address the social and emotional needs and challenges of military
children, while on the other hand, teachers and other school adults are developing their own
homegrown strategies for making schools more welcoming to military students (Bradshaw et al.,
2010; Mmari et al., 2009). This finding provides empirical support that multiple components of
school climate, especially belonging, caring relationships and family respect, are associated with
lower rates of depression and suicidal ideation as well as enhanced well-being. More detailed

95
analyses can be conducted in schools to explore how teachers facilitate belonging, caring
relationships, and respect for family background among military students. Findings from these
potential studies can uncover promising strategies specifically addressing the school-related
challenges of military students.
Also, unique to this study are analyses that allow us to compare associations between
school climate and mental health among military and non-military student populations.
Associations between each component of school climate and mental health category were similar
in magnitude for both military and non-military populations. To date, there have been few
studies that have compared associations between school climate and mental health between
different demographic groups. Given the distinct stressors that military children experience, it is
surprising that a supportive school environment can influence their mental health outcomes
similarly to non-military children. It is possible that staff and students at the military-connected
schools in our sample are attuned to the challenges of military life due to their proximity to
military installations and exposure to military students. Day-to-day experience with military
culture provides school staff with first-hand knowledge to make a school’s social and emotional
climate to suit the needs of military children. Further analyses can measure a school’s awareness
of military culture and how this influences school climate and mental health outcomes among
military students.
Summary
Overall, the findings in this study indicate that there are significant associations between
school climate and mental health, even when accounting for the experience of deployment. In
addition, students with a military sibling reported slightly more adverse mental health outcomes,
while non-military students reported higher rates of well-being. Also, the findings suggest that

96
associations between school climate and mental health were still significant among military
students; moreover, the magnitude of associations of school climate and mental health are similar
in magnitude for both military and non-military students. The results of this study suggest that
future theories, educational policies and interventions on the mental health of students in
military-connected schools be focused on enhancing the social and emotional climate of these
school environments.
Limitations
Several limitations in this study must be noted before generalizing findings to current
practice. First, this study examined associations between school contextual variables (i.e. school
academic achievement, poverty level, and military concentration) and mental health, finding no
significant associations. Academic achievement and poverty indicators however were school-
level variables, assigned to all sampled students within a school. Future analyses would need to
also include academic achievement and poverty data at the student level to further assess
relationships between these variables and mental health. Third, logistic regression analyses on
mental health by school climate were conducted in separate samples of military and non-military
children. These analyses were conducted to compare the magnitude of associations between the
two groups of students. Future investigations could explore the relationship between school
climate and mental health also among students with a military sibling. Last, causality cannot be
assumed due to the cross-sectional design of this study.
Implications
The findings from this study highlight the significant role of school climate in promoting
well-being and curbing rates of depression and suicidal ideation among students in military-
connected schools. Research has found that when experiencing parental deployment and other

97
military stressors, military children develop healthy coping strategies in supportive family,
household, and community contexts (i.e. easy access to military-specific mental health services)
(Flake et al., 2009; MacDermid et al., 2009); however, this area of research has largely ignored
the role of school climate. Future studies should be aware that the presence of a positive school
climate—caring relationships, belonging, family respect, safety, and risky behavior
disapproval—can influence the mental health outcomes of military children, even accounting for
the stress of deployment. More importantly, the utilization of a theoretically-driven model of
school climate in this study highlights that multiple components of school climate can curb rates
of depression and suicidal ideation and enhance well-being among all students in military-
connected schools.
The findings in this study challenge current trends in federal educational policy (i.e. No
Child Left Behind). Schools are embedded in high stakes accountability for academic
achievement, especially targeting traditionally underserved and low achieving minority groups
(Cochran-Smith, 2005; Greenburg et al., 2003). This policy context contradicts decades of
research and theoretical work highlighting the importance of social and emotional skill building,
mental health and social and emotional climate in improving academic outcomes. Cohen and
colleagues (2010) found that school climate is excluded from general state accountability
systems in almost all fifty states, leaving it up to state educational policy governance structures
to voluntarily pursue school climate improvement. Further, school climate is defined and
measured in various ways, not necessarily defined by the research literature. Findings from this
study suggest that school climate improvement, especially among high risk populations (i.e.
military children) be included in federal and state accountability systems.

98
In summary, this study addressed gaps in the military-connected school and military
children and school climate literatures and has implications for future research. Future studies
should continue to investigate how school climate can influence the mental health of students,
both military and non-military, in military-connected schools. Studies that are longitudinal and
multivariate are needed to examine cause and affect relationships. In addition, detailed
qualitative studies are needed to examine how school staff and students in military-connected
schools improve school climate perceptions among students. These studies can generate
homegrown-practices as well as provide more detailed information as to how each component of
school climate can enhance the well-being and curb depression and suicidal ideation rates among
students in military-connected schools.

99
Chapter Four:
Victimization of Students in Military-Connected Schools: The Role of School
Climate and Deployment
Introduction
The Children of Military Service Personnel
In recent years, a growing number of studies in psychology, public health and medicine
have examined the social, emotional, and psychological outcomes of the children of military
service members (Burrell, Adams, Durand, & Castro, 2006; Doyle and Peterson, 2005; Gorman,
Eide, & Hisle-Gorman, 2010; Reed, Bell, & Edwards, 2011). Since 2002, the Department of
Defense Educational Activity (DoDEA) estimates that about 2 million children (ages 0-18) have
experienced the deployment of a parent, and a large proportion of these children have
experienced multiple parental deployments. DoDEA recently estimated that there are about 1.2
million school-aged children of active-duty military service members. An additional 625,000
children of National Guard, and 705,000 children of Reserve members attend public schools
operated by local educational agencies. 80% of these students comprise a significant minority of
students (at least 400 students or more than 10% of total student enrollment) in 214 school
districts in the United States (Kitmitto et al., 2011). Public schools that serve a significant
number or proportion of military students (about 400 military students or 4% of total student
enrollment) are known as military-connected schools (Kitmitto et al., 2011).
Studies have found that military children have more adverse psychological, social, and
emotional outcomes than their civilian peers in the present Iraq and Afghanistan war context.
Recently, a retrospective cohort study on military children and military deployments found that
during the 2005-2006 fiscal year the number of mental and behavioral health visits increased by

100
11%, behavioral disorders increased by 19%, and stress disorders increased by 18% in military
children when a military parent deployed (Gorman et al., 2010). In addition, rates increased
particularly in adolescents and children of married and male military parents (Gorman et al.,
2010). Other studies have suggested that since the beginning of the Iraq and Afghanistan wars,
there has been a gradual increase in depressive symptoms, high levels of anxiety, and
oppositional behavior in schools (Doyle & Peterson, 2005; Flake, Davis, Johnson, & Middleton,
2009; Galovski & Lyons, 2004; Gorman, Eide, & Hisle-Gorman, 2010; Reed, Bell, & Edwards,
2011).
The Experience of Deployment
A large body of research indicates that deployment related stressors negatively impact the
psychological, social, and emotional issues of military children. During this period (i.e. pre-
deployment, deployment, reunion, and reintegration), military children experience parental
separation (Huebner, Mancini, Wilcox, Grass, & Grass, 2007), geographic relocation and school
transitions (Bradshaw, Sudhinaraset, Mmari, & Blum, 2010; Medway & Merchant, 1987), and
fear of death and injury of a deployed parent (Cozza, Chun, & Pollo, 2005; Flake et al., 2009;
Mmari, Roche, Sudhinaraset, & Blum, 2009). Military children also experience the daily stress
and anxiety of a left-behind parent (Cozza, Chun, & Polo, 2005; Morris & Age, 2009). The stress
and anxiety of the left-behind parent negatively impacts the psychological and behavioral
outcomes of military children (Cozza et al., 2005; Flake et al., 2009). Cozza and colleauges
(2005) found that left-behind parents share with their children developmentally inappropriate
information and communicate their own sense of grief and loss related to the fear of spousal
injury or death. In addition, left-behind family members must also redistribute household roles
and responsibilities (Faber, Willerton, Clymer, MacDermid, & Weiss, 2008; Huebner et al.,

101
2007). In a qualitative study of adolescent youth with deployed parents, Huebner et al. (2007)
found that military adolescents assumed household chores and responsibilities previously
designated to adults, including taking care of siblings and supporting left-behind parents
financially. While adolescents with deployed parents experience anxiety from growing
household responsibilities, Huebner and colleagues (2007) also found that they exhibited pride
for making adult household contributions.
After a deployed parent is reunited with his or her family, the experience of post-
deployment reintegration can be also stressful. Studies have found that reintegrated military
families must deal with re-shifting household responsibilities and roles, the financial stress of a
veterans’ re-establishment of employment, and re-establishing a relationship with a reunited
parent after months or even years of separation (Burrell et al., 2006; Doyle & Peterson, 2005;
Peebles-Kleiger & Kleiger, 1994). These stressors have been found to result in negative social
and psychological outcomes among military children (Burrell et al., 2006; Doyle & Peterson,
2005; Peebles-Kleiger & Kleiger, 1994). Adolescents, in particular, struggle with relinquishing
adult responsibilities attained when a parent was a deployed (Huebner et al., 2007). In more
severe cases, military children experience significant stress and anxiety during the post-
deployment reintegration period when the veteran returns with a war-related trauma illness or
post-traumatic stress disorder (PTSD) (Galovski & Lyons, 2004; Jordan et al., 1992; Rosenheck
& Nathan, 1986; Solomon, 1988). In studies conducted in different war contexts (including the
Iraq and Afghanistan wars), research has found that veteran trauma is related to marital
problems, unstable family systems, poor parenting, secondary traumatization between a veteran
and his or her child, and higher rates of domestic violence (Al-Turkait & Ohaeri, 2008; Goff,
Crow, Reisbig, & Hamilton, 2007; Jordan et al. 1992; Rosenheck & Fontana, 1998; Rosenheck

102
& Nathan, 1985). Not surprisingly, a few studies suggest that military children of veteran parents
with war-related trauma develop mental health and secondary trauma issues. Moreover, some
families dealing with veteran war trauma have difficulty seeking help due to lack of access
and/or cultural stigma (Burrell et al., 2006; Caselli & Motta, 1995; Solomon, 1990).
Victimization of Military Children in Schools
Research has indicated that minority children (i.e. race and ethnicity, gender, cultural
group) and psychologically strained student populations often feel alienated from peers and
school adults and experience challenges with social and academic functioning; consequently,
they often become targets of physical and non-physical victimization (Eisenberg, Neumark-
Sztainer, & Perry, 2003; Lynch & Cicchetti, 1997; Townsend, 2000). In the case of military
children, recent qualitative studies have shown that they struggle more with social and academic
functioning within the school and classroom contexts and may feel a greater sense of
disengagement and alienation when compared to their civilian peers. These findings suggest that
military children are more vulnerable to school violence and victimization than civilian peers
(Chandra, Martin, Hawkins, & Richardson, 2010; Mmari et al., 2009). In a study of children of
deployed soldiers, school staff and parents reported that some military students struggled with
maintaining focus in the classroom as well as emotional regulation with teachers and peers
(Chandra et al, 2010). Participants in the study attributed academic and social functioning
problems of military students to deployment-related parental absence, increased responsibilities
at home, poor mental health of some non-deployed parents, and difficulty accessing mental
health services. In addition, Bradshaw and colleagues (2010) found that due to multiple
transitions, military students had difficulty with making friends, adjusting to new physical school

103
environments, and proactively seeking academic and emotional support from school staff and
peers.
Not surprisingly, research shows early indicators that military children are vulnerable to
physical and non-physical victimization in schools (Chandra et al., 2010; Mmari et al., 2009). In
a study of military adolescents with deployed parents, school personnel and home caregivers
reported that military adolescents exhibited increased behavior problems, including fights with
other students at school (Mmari et al., 2009). Moreover, Chandra and colleagues (2010) found
that military children were bullying victims of anti-war student perpetrators. While the preceding
studies have uncovered concerning findings about military children and their risk of
victimization, these studies had exploratory objectives and small purposive samples. Hence, it is
difficult to make generalizations about the victimization rates of military children and risk and
protective factors within the school context. Future studies concerning the victimization rates
among military children in public schools and potential protective factors in the school context
must utilize large population samples of students in military-connected schools with reliable and
detailed measures of victimization.
School Climate and Victimization
A large body of research has indicated that a school’s social and emotional climate
influence rates of student victimization and violence (Eisenberg et al., 2003; Elliot, Gregory, &
Fan, 2010; Hoy, Smith, & Sweetland, 2002; Hoy, Tarter, & Bliss, 1990). A school’s climate has
multiple components, including supportive teacher-student relationships, high levels of
meaningful participation, a safe school environment, a sense of belonging, and a school’s respect
for a student’s family and cultural background (Cohen, McCabe, Michelli, & Pickerall; 2009;
Zullig, Koopman, Patton, & Ubbes, 2010).

104
Relationships and Risky Behavior Disapproval. In studies conducted in diverse
locales, supportive teacher and peer relationships have been found to create safer school
environments and a school-wide culture that discourages various risky behaviors on school
grounds (i.e. violence and drugs). In a study of Israeli high school students, Marachi, Astor, &
Benbenishty (2006) found that higher levels of teacher support were associated with lower rates
of victimization. Similarly, Wilson (2004) found that caring student-teacher relationships,
student-peer relationships, and respect for authority are related to decreased physical and
relational aggression (e.g. verbal harassment). In addition, research has suggested that in school
environments comprised of trusting and nurturing teacher-student relationships, students adopt
attitudes and behaviors that prevent school violence. In a study of high schools in Virginia, Eliot,
Cornell, Gregory, & Fan (2010) found that in schools where students perceived more peer and
teacher social and emotional support, students tended to endorse positive attitudes toward
seeking help from school staff for bullying and threats of violence among peers.
Research has also suggested that when supportive teacher-student relationships are
present, teachers and students are able to establish a collective understanding of school-level
procedures that address school violence and other risky behaviors (Brand, Felner, Shim,
Seitsinger, & Dumas, 2003). Teachers are also able to facilitate active student participation in
monitoring and reporting the violent behaviors of their peers, and pro-social relationships among
students (Brand, Felner, Shim, Seitsinger, & Dumas, 2003). In a national study of Israeli
secondary students, Khoury-Kassabri, Benbenishty, & Astor (2005) found that in both Jewish
and Arab schools, a student’s understanding of school violence policies and procedures and
positive relationships with teachers was negatively associated with several victimization types,
including serious physical victimization. In addition, when students are involved in the decision-

105
making process for responding to school violence, Khoury-Kassabri and colleagues (2005) found
significantly lower rates of all victimization types. In a study of American secondary students in
urban schools, Brand and colleagues (2003) found that consistently implemented school
discipline policies and procedures and students’ perceptions of a clarity of rules and expectations
facilitated students' behavioral adjustment. In contrast, disciplinary harshness was negatively
associated with students’ behavioral adjustment.
Respect for Cultural and Family Background. A large body of literature on culturally
responsive pedagogy and classroom management has theorized that teachers and other school
staff need to be more sensitive and conscious of cultural expressions and historical experiences
of racial discrimination before judging the school behaviors of racial and ethnic minority youth.
Similarly, a recent review of literature on the children of military personnel suggests that school
staff need training on military families and culture to be more responsive to the academic and
emotional needs of military children (De Pedro et al., 2011). Empirically, culturally relevant
responses have been recently examined among youth who self-identify as transgendered. In a
mixed-methods study of transgendered youth and school climate, McGuire, Toomey, & Russell
(2010) found that specific school violence prevention strategies designed for transgendered
youth such as teacher training on transgendered identity were associated with feelings of safety,
connectedness to school staff, and less harassment among transgendered students.
Belonging. Belonging has also been shown to be associated with reduced rates of student
victimization. Studies have indicated that schools with high connectedness have lower
victimization rates, including physical aggression, than schools with low connectedness (Wilson,
2004). The prevalence of destructive social behaviors is also affected by connectedness. For
example, highly connected middle school students were less likely than less connected students

106
to report regular smoking, drinking, and marijuana use outside of school (Bond et al., 2004). In
addition, Bonny and colleagues (2000) found that highly connected middle school students in
urban districts reported fewer psychosomatic health symptoms, fewer nurse visits, and less
regular use of cigarettes, and alcohol.
Meaningful Participation and Safety. Other components of school climate such as
meaningful participation and safety can potentially exert a positive influence on creating positive
school climates and academic outcomes. This is especially evident when teachers, students, and
other members of a school community have active, participatory roles in organizing how spaces
in schools are utilized for instruction. In a qualitative study of a charter school specializing in
environmental studies and cooperative learning, Gislason (2009) found that the school’s open
house space and extended instructional block scheduling demanded collaboration among the
teachers and students in planning and instruction. However, Gislason (2009) found that open
spaces created barriers to the efficient use of classroom time and space (i.e. more noise, traffic,
students having difficulty hearing at the periphery of spaces, and extra effort to control
acoustics). In response to these barriers, Gislason found that students exerted a high degree of
ownership over the house setting and high level of self-motivation as a result of a cooperative
model and the need to create an orderly classroom space. In contrast, research has also found that
spaces that are unmonitored or “unowned” by teachers, students, and other school staff are
spaces prone to school violence. In a qualitative study of five high schools, Astor and colleagues
(1999) found that violent spaces occurred primarily in spaces such as hallways, dining areas, and
parking lots when school staff members were absent.
The preceding studies indicate that multiple components of school climate promote lower
rates of victimization. Conducted in different geographic contexts, the reviewed studies also

107
suggest that school climate’s effect on victimization can be generalized in almost any school
context. To date, studies have not yet examined the social and emotional qualities of military-
connected schools and associations between school climate and victimization among children
enrolled in these schools. Also, as previously mentioned, studies have begun to suggest that
military students may be engaged in school violence at higher rates than their civilian peers,
however, to date, only one study has examined these rates using a normative sample (Reed et al.,
2011). Further, no studies have yet examined associations between school climate and rates of
student victimization. Also, it is not clear if associations between school climate and
victimization are different for military children, when compared to non-military children or if
school climate can still influence victimization rates even when students are experiencing the
emotional toll of the deployment of a family member.
This study serves to fill the current gaps in studies examining the victimization of
military children and the role of school climate. I evaluate and compare the victimization rates of
military children (with a parent or sibling in the military) and non-military children as well as
examine relationships between school climate and mental health. Drawing from research on the
victimization of minority children, it is expected that military children have higher rates of
victimization than non-military children. There has also been a wide breadth of research on
school climate and victimization, and it is expected that school climate is associated with
reduced victimization rates. Moreover, considering research on the protective effects of school
climate on the victimization of psychologically strained student populations, it is also expected
that a supportive school climate would promote positive mental health outcomes even if students
have experienced military stressors such as a deployment.

108
Multiple objectives guide this study. First, this study compares the victimization rates of
military and non-military students (i.e. having a parent in the military or having a sibling in the
military). Second, this study assesses associations between multiple components of school
climate and victimization. Third, this study evaluates the effects of school climate on
victimization among military and non-military students. Last, this study seeks to examine
associations between deployment and victimization. The data from this dissertation is drawn
from two sources, the California Healthy Kids Survey and Dataquest, school-level data and
statistics on academic achievement and school demographics, from the California Department of
Education.
Methods
California Healthy Kids Survey (CHKS)
Core and military-connected modules. One of the data sources utilized in this study is
the 2011 California Healthy Kids survey (CHKS). The data from the California Healthy Kids
Survey (CHKS) is comprised of individual level data where individuals are also identified by
school, county, and local educational agency (i.e. public school district). The CHKS is the largest
statewide survey of resiliency, protective factors, and risk behaviors in the United States and is
collected from students at the elementary, middle, and high school levels (Austin, Bates & Duerr,
2011). The CHKS is comprised of several modules, including the core module. The core module
includes data on demographics (e.g. age, grade, gender, race and ethnicity), health-related
behaviors, tobacco use, drug use, violence behaviors, bullying, victimization, and school climate
(Austin et al., 2011).
The CHKS also includes the military-connected module. In 2010, a research team at the
University of Southern California and West Ed collaborated to create the CHKS military-

109
connected module in response to a growing awareness among researchers and educators on the
social, emotional, psychological and academic needs of military children. Congress directed the
U.S. Department of Education (DoE) and U.S. Department of Defense (DoD) to address these
gaps (U.S. Department of Defense, 2008). The military-connected module is administered to
parents, school staff, and students to gather their views of their schools and the educational,
developmental, and health-related needs of their students (Austin, Bates & Duerr, 2011). The
military connected module includes military demographic item as well as items on exposure to
deployment, school transitions, family dynamics, parent-school relationships, mental health,
positive well-being, is comprised of demographic items that identify military students and items
that measure students’ perceptions of school supports for military children. The student military
connected module is administered to both military and non-military students in military-
connected schools. The present study utilizes data from both the core module and military-
connected modules collected in Spring 2011. The CHKS core module included 115 items in the
middle school version and 126 items in the high school version.
Data collection procedure. The CHKS core module and the military-connected module
were administered in the Spring of 2011 in eight military-connected school districts serving
elementary, middle, and high school students, surrounding military bases in the San Diego
metropolitan area. The participating schools were required to survey all students in grades 5,7, 9,
and 11. Since the focus of this dissertation is on secondary students, data collected from 7
th
, 9
th
,
and 11
th
graders will be analyzed. Prior to the survey administered at a secondary school site,
parental consent was gathered for each participating student. Surveys were also sealed in
envelopes for classrooms in schools. The envelopes reported the number of students absent in
these classes at the day of the survey. There were 599 envelopes. The surveys were administered

110
by school staff members or by USC team members. Proctoring instructions and trainings were
given to all survey administrators and an introductory script was read to the participating
students. The USC research team provided survey administrators with gift cards as incentives, in
order to facilitate high response rates. Participants were encouraged to answer questions honestly
and assured their responses would remain anonymous. Participants were allowed to withdraw
from the survey at any time. The core and military-connected modules took approximately 2
hours to complete.
The data used in this study is a subsample of 14,943 7
th
, 9
th
and 11
th
graders, The overall
number of enrolled students eligible to participate in the survey was 18,701. Absent students and
students whose parents refused permission to participate in the study were excluded from the
study sample. The final response rate was 86.73%.
California Department of Education School Level Data
The CHKS dataset used for this dissertation was merged 2010 Accountability Progress
data, which is made publically available by the California Department of Education. California’s
comprehensive accountability system has monitored the academic achievement of every public
school in California. The accountability system is based on state requirements established by the
No Child Left Behind Act of 2001 (California Department of Education, 2011). The data
includes information pertaining to state and federal accountability requirements, including on
academic performance index (API), adequate yearly progress (AYP), and program improvement
(PI). In addition, API, AYP, and PI school-level data by subgroup (e.g. racial or ethnic group,
special education and English Learner status) is disaggregated.
API can be used as a measure of a school’s academic achievement level. A base API
score is calculated from the following indicators: Standardized Testing and Reporting (STAR)

111
test indicators in core academic subjects (English language arts, math, science, and history-social
science), California Modified Assessment (CMA) test results, California Alternate Performance
Assessment (CAPA) results, and California High School Exit Examination (CAHSEE) results
(California Department of Education, 2011). The API scores fall into a numeric index ranging
from 200 to 1000. After calculation of API, schools receive accountability information in API
reports. These reports include base API and a growth API. The Base API starts the reporting
cycle and is released approximately a year after testing. For example, the 2010 Base is calculated
from results of statewide testing in spring 2010 but is released the following school year. The
Growth API, released after Base API, is calculated with the same indicators as the prior year
Base API but from test results of the following year. For example, the 2011 Growth is calculated
from results of statewide testing in spring 2011 and is released the following school year.
Data related to socio-economic status (SES) of school is provided. This includes percent
of students qualifying for free and reduced priced meals. The USDA determines federal
eligibility requirements for students to receive free and reduced price lunch (California
Department of Education, 2011). The USDA Department’s guidelines for free and reduced price
meals were obtained by multiplying the 2011 Federal Income poverty guidelines by 1.30 and
1.85 respectively, and by rounding the result upward to the next whole dollar. The data in this
dissertation includes percent of students who qualify for free and reduced priced meals by
school. This variable is also commonly used by educational researchers as an indicator of SES
and poverty concentration in a school.

112
Measures
Dependent Variables
Victimization. As seen in Table 25, participants were asked about their victimization on
school property in the past year (α = .792). Eight items were used for this construct: During the
past 12 months, how many times on school property have you…been pushed, shoved, slapped,
hit, or kicked by someone who wasn’t just kidding around; been afraid of being beaten up; been
in a physical fight; had mean rumors or lies spread about you; had sexual jokes, comments
gestures made to you; been made fun of because of your looks or the way you talk; had your
property stolen or deliberately damaged, such as your car, clothing, or books; been threatened or
injured with a weapon (gun, knife, club, etc). The responses to these items were on a four-point
Likert scale and ranged from 1 = 0 times, 2 = 1 time, 3 = 2-3 times. A composite victimization
scale was created by summing the responses, and then, the scale was dichotomized—0 no
victimization, 1 victimization.
Table 25. Domains, Scales and Items for the Victimization
Variable Items
Victimization “During the past 12 months, how many times on school property
have you…
…been pushed, shoved slapped, hit or kicked by someone who
wasn’t just kidding around?”
…been in a physical fight?”
…had mean rumors or lies spread about you?”
…had sexual jokes, comments, gestures made to you?”
…been made fun of because of your looks or the way you talk?”
…had your property stolen or deliberately damaged, such as your
car, clothing, or books?”
…been threatened or injured with a weapon (gun, knife, club, etc)?”
The possible responses to these items were A) 0 times; B) 1 time; C)
2-3 times.

113
Independent Variables
School Climate Perceptions. According to the school climate literature, there are
multiple dimensions of school climate in the school context. The CHKS core module provides
items covering multiple dimensions of school climate. These include sense of safety, relationship
with adults, belonging, and peer risky climate, meaningful participation, and school’s respect for
a student’s family. Table 26 shows the items that comprise the composite scales that represent
school climate.
Safety. Two items asked respondents about their perceptions of safety in their schools (α
= .812). One item is “I feel safe in my school.” The responses to this item were on a five point
Likert scale and ranged from 1 = Strongly disagree, 2 = Disagree, 3 = Neither disagree or agree,
4 = Agree, 5 = Strongly agree. The second item is “How safe do you feel when you are at
school?” The responses to this item were 1 = Very safe, 2 = Safe, 3 = Neither safe nor unsafe, 4
= Unsafe, 5 = Very unsafe. The item responses were not congruent. The second item was
recoded as 1 = Very Unsafe, 2 = Unsafe, 3 = Neither safe nor unsafe, 4 = Safe, 5= Very Safe.
The composite score of each participant was derived by summing up the two items and dividing
by two. Therefore, the safety scale scores range from 1-5.
Relationship with Adults. In this scale, participants were asked about their relationships
with adults in the school (α = .880). Six items were used for this construct: At my school, there is
a teacher or some other adult who tells me I do a good job; who tells me when I do good; who
notices when I’m not there; who always wants me to do my best; who listens to me when I have
something to say; who believes I will be a success. The responses to these items were on a four-
point Likert scale and ranged from 1 = Not at all true, 2= A little true, 3 = Pretty much true, 4 =

114
Very much true. The composite score of each participant was derived by summing up the six
items and dividing by six. Therefore, the relationships scale scores range from 1-4.
Belonging. In this scale, participants were asked about their sense of belonging (α =
.797). Three items were used for this construct: Strongly agree or disagree with, I feel close to
people at this school; I am happy to be at this school; I feel like I am part of this school The
possible responses to these items were on a five-point Likert scale and ranged from 1 = Strongly
disagree, 2= Disagree, 3 = Neither disagree or agree, 4 = Agree, 5 = Strongly agree. The
composite score of each participant was derived by summing up the three items and dividing by
three. Therefore, the safety scale scores range from 1-5.
Risky Behavior Approval. In this scale, participants were asked about their perceptions of
risky behaviors (α = .900). Six items were used for this construct: How do you feel about
someone your own age smoking one or more packs of cigarettes a day; having one or more
drinks of alcohol every day; trying marijuana or hashish; using marijuana once a month or more,
carrying a weapon to school; how would your friends feel about you smoking one or more packs
of cigarettes a day? The possible responses to these items were on a three-point Likert scale and
ranged from 1 = Neither approve or disapprove, 2= Somewhat disapprove, 3 = Strongly
disapprove. The composite score of each participant was derived by summing up the six items
and dividing by six. Therefore, the safety scale scores range from 1-3.
Meaningful Participation. In this scale, participants were asked about their meaningful
participation at school (α = .742). Three items were used for this construct: I do interesting
activities at school; At school, I help decide things like class activities; I do things that make a
difference. The possible responses to these items were on a four-point Likert scale and ranged 1
= Not at all true, 2= A little true, 3 = Pretty much true, 4 = Very much true. The composite score

115
of each participant was derived by summing up the three items and dividing by 3. Therefore, the
safety scale scores range from 1-4.
Respect for Family. Two items asked participants to report their school’s for their family
life. The first item was “Adults in this school respect my family.” The second item asked
participants to report school adult’s understanding for their family life. The item was “Other
students in school do not really understand my family life.” The possible responses to this item
were 1 = Not at all true, 2= A little true, 3 = Pretty much true, 4 = Very much true. The items are
not congruent. The second item was reverse coded. The composite score of each participant was
derived by summing up the two items. Therefore, the safety scale scores range from 1-4.
Table 26. Domains, Scales and Items for the School Climate Variables
Variable Items
Belonging “Strongly agree or disagree with…
…I feel close to people at this school.”
…I am happy to be at this school.”
…I feel like I am part of this school.”
The possible responses to these items were A) Strongly disagree; 2)
Disagree; 3) Neither disagree or agree; 4) Agree; 5) Strongly agree.
Relationships
with Adults
“At my school there is a teacher or some other adult who…
…tells me I do a good job.”
…tells me when I do good.”
…who notices me when I’m not there.”
…who always wants me to do my best.”
…who notices when I’m not there.”
…who always wants me to do my best.”
…who listens to me when I have something to say.”
…who believes I will be a success.”
The possible responses to these items were A) Not at all true; 2) A little
true; 3) Pretty much true; 4) Very much true;
Safety “I feel safe in my school.”
A) Strongly Disagree; 2) Disagree; 3) Neither disagree or agree; 4)
Agree; 5) Strongly agree.
“How safe do you feel when you are at school?”
A) Very safe; B) Safe; C) Neither safe nor unsafe; D) Unsafe; E) Very
unsafe.
(Table 26 continued)

116
(Table 26 continued)
Meaningful
Participation
“At school…
…I do interesting things at school.”
…I help decide things like class activities.”
…I do things that make a difference.”
The possible responses to this item were A) Not at all true; B) A little
true; C) Pretty much true; D) Very much true; E) Don’t know.
Approval of
Risky
Behaviors
“How do you feel about…
…someone your own age smoking one or more packs of cigarettes a day.”
…having one or more drinks of alcohol every day.”
…trying marijuana or hashish.”
…using marijuana once a month or more.”
…carrying a weapon to school.”
The possible responses to this item were A) Neither approve nor
disapprove; B) Somewhat disapprove; C) Strongly disapprove
Respect for
Family
“Adults in this school respect my family.”
“Other students in school do not really understand my family.”
The possible responses to this item were A) Not at all true; B) A little true;
C) Pretty much true; D) Very much true; E) Don’t know.

Deployment (Past Ten Years). As seen in Table 27, one item asked participants to report
their experiences with the deployment of a family member. The item was “About your family
and the military, as far as you can remember, how many times in the last 10 years did any
member of your family leave home and serve (deploy) outside the USA?” The possible
responses to this item were 1 = Never, 2 = Once, and 3 = Twice or more. Two dummy variables
were created: Deployment (1 time) and deployment (2 or more times).
Table 27. Deployment Item
Variable Item
Deployment “As far as you can remember, how many times in the past 10 years
did any member of your family leave home and serve (deploy) outside
the USA?”

The possible responses to this item were A) Never; B) Once; C) Twice
or more; D) Don’t know.

117
Demographic Variables. As seen in Table 28, four main demographic characteristics
were represented in this study—race and ethnicity, gender, grade level, and military connection.
Race and Ethnicity. Two items were used to determine the racial and ethnic backgrounds
of the participants. Participants were asked, “What is your race?” the possible responses were 1 =
American Indian or Alaska Native, 2 = Asian, 3= Black or African-American, 5 = Native
Hawaiian or Pacific Islander, and 5 = White. Another item asked, “Are you of Hispanic or Latino
origin?” The possible responses were 1) Yes and 2) No. A variable was computed for race and
ethnicity and the categories were 1) Asian/AI/NH/PI 2) Black 3) White 4) Mixed 5)
Hispanic/Latino. Then, race and ethnicity was represented by five dummy variables—
Asian/AI/NH/PI, Black, White, Mixed, and Hispanic/Latino.
Grade level. One item was used to determine the grade of the participants. Participants
were asked, “What grade are you in? The possible responses were 1) 6
th
grade, 2) 7
th
grade, 3) 8
th

grade, 4) 9
th
grade, 5) 10
th
grade, 6) 11
th
grade, 7) 12
th
grade, 8) Other grade. Grade level was
represented by a series of dummy variables: 7
th
grade, 9
th
grade, and 11
th
grade.
Gender. Participants were asked to report their gender. They were asked, “What is your
sex?” and the possible responses were 1 = male and 2 = female. The items were recoded as 0 =
female and 1 = male. Gender was represented by a dummy variable coded as 1= female and 0 =
male.
Military-Connection. One item was used to determine a respondent’s military connection.
Participants were asked to report their connection to the military in one item. The item asks
participants, “Who in your family is in the military (Army, Navy, Marines, Air Force, national
Guard or Reserves)?” the possible responses were 1 = No one in my family is in the military, 2 =
Father, 3 = Mother, 4 = Brother or sister, and 5 = Grandparent or other relative. Based on the

118
literature, military connection is defined as having a parent or sibling serving in the military.
Thus, the variable was recoded with the following categories: 0) Not in the military 2) Having a
military parent and 3) Having a military sibling. Military connection was represented by two
dummy variables: Having a military parent, having a military sibling.
Table 28. Domains, Scales and Items for the Demographic Variables
Variable Items
Gender “What is your sex?”
Respondents chose one of the following: A) Male and B) Female.
Grade “What grade are you in?”
Respondents chose one of the following: A) 6
th
grade; B) 7
th
grade; C)
8
th
grade; D) 9
th
grade; E) 10
th
grade; F) 11
th
grade; G) 12
th
grade; H)
Other grade; I) Ungraded
Race and
Ethnicity
“What is your race?”
Possible responses included A) American Indian or Alaska Native; B)
Asian; C) Black or African American; D) Native Hawaiian or Pacific
Islander; E) White; F) Mixed (two or more) races.
“Are you of Hispanic or Latino origin?”
Respondents chose one of the following: A) No; B) Yes
Military
Connection
“Who in your family is in the military (Army, Navy, Marines, Air Force,
National Guard, or Reserves)?
Respondents could mark more than one of the following answers: A) No
one in my family is in the military; B) Father; C) Mother; D) Brother or
sister; E) Grandparent or other relative; F) Don’t know

School Level Variables
Taken from publically available CDE data, two school-level variables were also used to
represent two school contextual factors surrounding school climate effects on mental health and
victimization—academic performance index (API) and free and reduced price lunch. A third
school-level variable, military concentration, was generated from student level data from the
CHKS military-connected module.
Academic Performance Index (API). To obtain a school’s base API score, the
following indicators are calculated: Standardized Testing and Reporting (STAR) test indicators
in core academic subjects (English language arts, math, science, and history-social science),

119
California Modified Assessment (CMA) test results, California Alternate Performance
Assessment (CAPA) results, and California High School Exit Examination (CAHSEE) results.
The API scores fall into a numeric index ranging from 200 to 1000.
Free and Reduced Price Meals. The California Department of Education also publishes
the percentage of students who qualify for free and reduced price meals in each school. The
USDA determines federal eligibility requirements for students to receive free and reduced price
lunch. The USDA Department’s guidelines for free and reduced price meals were obtained by
multiplying the 2011 Federal Income poverty guidelines by 1.30 and 1.85 respectively, and by
rounding the result upward to the next whole dollar.
1. Military Concentration. The military-connected module in the CHKS includes an
item that asks participants the following: “Who in your family is in the military
(Army, Navy, Marines, Air Force, National Guard, or Reserves?” The possible
responses are A) No one in my family is in the military, B) Father, C) Mother, D)
Brother or sister; E) Grandparent or other relative; F) Don’t know. The item was
recoded as 1 - Having a parent in the military and 0 - No parent in the military. A
variable aggregated at the school level was created. This aggregate variable represents
the proportion of students with a military parent in a school.
Analytical Plan
The goal of this study is to examine the relationship between school climate and mental
health—well-being, depression, and suicidal ideation—among students in military-connected
schools. Associations between school climate and mental health among all students and within
military and non-military student populations were examined. In addition, the relationship
between school climate and mental health, controlling for deployment was assessed.

120
The data was analyzed using bivariate, and multivariate analyses. First, cross-tabulations
were generated to compare the victimization rates among non-military students, students with a
military parent and students with a military sibling. Chi-square analyses were also conducted to
determine significant associations between military connection and victimization. Correlational
analyses were then conducted to assess associations between school climate and victimization.
Next, multivariate analyses using logistic regression were then conducted. The logistic
regression technique is used when the dependent variable is limited or constrained to a
dichotomous outcome. The dependent variable in this study was nominal and dichotomous and
coded as 0 or 1. A direct method procedure was used to assess the predictive value of
independent variables of interest, while controlling for other independent variables. Separate
logistic regression analyses were also conducted to assess relationships between school climate
and victimization within each group of students by military-connection—non-military students,
students with a military parent, and students with a military sibling. Logistic regression analyses
were also conducted to assess relationships between deployment and victimization, controlling
for demographics and school climate.
Prior to conducting the logistic regression analyses, several decisions were made to
evaluate the logistic regression models. First, statistical significance tests were set at p < .05.
Second, it was determined that the strength of each independent variable would be measured by
the odds ratios and confidence intervals of each predictor. Odds ratios and confidence intervals,
instead of individual regression coefficients, provide the researcher easier interpretation of
results and are generally used in studies with large samples. Third, a series of indicators, the
likelihood ratio test, Cox & Snell R
2
, Nagelkerke R
2
,

were used to determine if each logistic
model was a model fit. A logistic model is said to provide a better fit to the data if it

121
demonstrates an improvement over the intercept model, which contains no predictors. Fourth,
goodness-of-fit statistics were used to assess the fit of a logistic model against observed
outcomes. The inferential goodness-of-fit statistic used in this study is the Hosmer-Lemeshow
(H-L) test. The H-L statistic is a Pearson chi-square statistic calculated from a 2 x g table of
observed and estimated expected frequencies, where g is the number of groups formed from the
estimated probabilities. A non-significant H-L statistic indicates that there is no significant
difference between the observed frequencies and the expected frequencies derived from the
model, and hence, the model fits the data well. Fifth, the percent correct prediction statistic was
evaluated to determine the percent of all cases correctly predicted in each logistic regression
model. This statistic is an indicator of predictive power of each logistic regression model. . In
this study, overall percent correct prediction statistics for all logistic regression models ranged
from 71.3% to 77.8%. Also, the number of cases observed to be 0 that were correctly predicted
to be 0 ranged from 76 to 113. The number of cases observed to be 1 that were correctly
predicted to be 1 ranged from 292 to 7579.
Sixth, listwise case deletion was used for the logistic regression models. Last, additional
analyses were also conducted to account for school-level factors, and statistical interactions
incorporating school-level factors and military connection were also conducted; however, no
significant statistical interactions. SPSS 19 was utilized in all univariate, bivariate, and
regression analyses.
Other advanced multivariate statistical techniques was considered for this study. First, a
multivariate linear regression (MLR) approach was considered. MLR is utilized for continuous
outcome variables. It was determined that a multivariate model with a dichotomous variable
representing victimization (yes/no) would have greater interpretability than a model utilizing a

122
continuous victimization variable. Second, a two-level hierarchical linear model (HLM) was
prepared to assess the effects of school-level and individual-level variables on victimization;
however, preliminary analyses indicated an intra-class correlation coefficient of .035. Due to the
lack of practical significance for explaining 3.5% of between school variation, HLM was deemed
inappropriate for this study.
Results
Descriptive Statistics of the Sample
In the first step of analysis, descriptive statistics were conducted to determine the
demographic characteristics of the overall sample and demographic differences between military
and non-military students. Table 29 shows that the sample is almost evenly split by male and
female and by grade level (7
th
, 9
th
, and 11
th
grades). The results indicate a diverse sample of
students by race and ethnicity. Almost three-fourths of the sample (72.7%) is non-white.
Hispanic students comprise largest racial/ethnic group in the sample (50.3%), while Black
students comprise the smallest proportion of students (3.0%). Table 29 also shows that military-
connected students comprised almost 15% of the sample. About 4.2% of the sample is comprised
of students with a military sibling, while 8.8% of the students have a parent in the military.
86.9% of the sample are classified as non-military.
Bivariate analyses were also conducted within three subsamples—non-military students,
students with a parent serving in the military and students with siblings serving in the military.
When compared to the proportions of students by race and ethnicity for the non-military sample,
Table 29 indicates that among students with a military parent, Asian, Black, and Mixed students
represent a larger proportion of total students (12.1%, 7.2%, and 18.1%, respectively), while
Hispanic students represent a smaller proportion (36.9%). Among students with a sibling, white

123
students are slightly overrepresented (31.7%). When compared to the proportions of students by
grade in the non-military group, Table 29 indicates that among students with a military parent,
students are slightly younger overall. 40.6% of students with a military parent are 7
th
graders,
while 25.8% are 11
th
graders. In contrast, the results suggest that students with siblings serving in
the military are a slightly younger group when compared to non-military students and students
with military parents. Table 29 shows that 25.6% of students with military siblings are 7
th

graders, while 37.5% are in the 11
th
grade.
Table 29. Sample Characteristics for Study #3 (n= 14,943)
Total
N(%)
No one in the
military
(N=12,990)
%
Parent
(N=1396)
%
Sibling
(N=649)
%
Gender
Female
Male

7606 (51.7%)
7181 (48.6%)

51.5
48.5

51.1
48.9

50.9
49.1
Grade
7
th

9
th

11
th

4588 (32.9%)
4908 (35.2%)
4446 (31.9%)

32.3
35.4
32.3

40.6
33.6
25.8

25.6
33.9
37.5
Race/Ethnicity
Asian/AI/HI/AN
Black
White
Mixed Race
Hispanic

1189 (8.2%)
432 (3.0%)
3948 (27.3%)
1606 (11.1%)
7261 (50.3%)

7.9
2.5
27.3
10.3
52.0

12.1
7.2
25.6
18.1
36.9

7.1
4.0
31.7
12.3
44.8

Military Connection and Victimization
Since the research literature suggests that military and non-military students have
different social and emotional experiences in school, analyses were required to compare their
victimization rates. Detailed bivariate statistical analyses and in later sections, multivariate
analyses, were conducted to determine if there are different victimization rates among non-

124
military students, students with a military parent, and students with a military sibling. Table 30
shows that there is a significant association between overall victimization and military
connection (show chi-square). The highest rates of overall victimization occurred among
students with a military parent (77.1%), slightly higher than students with a military sibling
(75%) and non-military students (70.3%).
Analyses were then conducted to compare the rates of each form of victimization among
the three groups of students. As seen in Table 30, on all seven indicators of victimization,
students with military parents had the highest rates. Significant chi-square tests of association
also were found between military connection and all forms of victimization. About 39.2% of
students with a military parent and reported being shoved, pushed, slapped, hit or kicked, slightly
higher than students with a military sibling (32.3%) and non-military students (39.2%). Also,
students with military parents (14.4%) and reported higher rates of being threatened or injured
with a weapon (e.g. gun, knife, or club) than students with a military sibling (11.8%) and non-
military students (7.7%). Students with a military parent (28.2%) also reported higher rates of
being afraid of being beaten up than students with a military sibling (20.7%) and non-military
students (28.9%). In addition, about 33.7% of students with a military parent reported higher
rates of their property stolen or damaged, slightly higher than students with a military sibling
(29.8%) and non-military students (25.5%).
Table 30 also shows that students with a military parent experience higher rates of non-
physical victimization than students with a military sibling and non-military students. When
compared to non-military students (40.5%) and students with military sibling (43.4%), students
with military parents (46.3%) reported higher rates of having mean rumors or lies spread about
them. Similarly, students with a military parent (51.7%) also reported higher rates of being

125
victimized by sexual jokes, comments, or gestures than students with a military sibling (49.4%)
and non-military students (45.5%). About 47.1% of students with military parents reported the
highest rates of being made fun of because of your looks/way you talk, higher than students with
a military sibling (42.4%) and non-military students (37.6%).
Table 30. Victimization Rates by Military Connection Status (n=14,943)
Total
(%)
Not in the
Military
(%)
(n=12,990)
Parent
(%)
(n=1,318)
Sibling
(%)
(n=635)
In the past 12 months at school (Yes)
Shoved, pushed, slapped, hit or kicked**

30.0

28.9

39.2

32.3
Afraid of being beaten up** 19.9 18.9 28.2 20.7
Had mean rumors or lies spread about you** 41.1 40.5 46.3 43.4
Sexual jokes, comments or gestures** 45.4 45.5 51.7 49.4
Made fun of because of your looks/way you
talk**
38.6 37.6 47.1 40.2
Property stolen/deliberately damaged** 26.4 25.5 33.7 29.3
Mean rumors or lies on the Internet 21.9 21.0 28.2 26.0
Been threatened or injured with a weapon** 8.5 7.7 14.4 11.8
Victimization** 71.1 70.3 77.1 75.0
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001
School Climate and Victimization in Overall Sample
In the next steps of analyses, bivariate and multivariate statistics were conducted to
evaluate relationships between school climate and victimization among students in military-
connected schools. Correlation analyses were first conducted to assess potential bivariate
associations between school climate and the composite victimization scale. These analyses
suggest that school climate plays a role in reducing rates of victimization. As seen in Table 31,
all school climate variables were significantly associated with victimization. Moderate negative
associations were found between belonging and victimization (r=-.17) and belonging and
relationships (r=-.12). In other words, greater degrees of belonging as well as relationships are
related to lower rates of victimization. In addition, table 31 shows that significant associations,

126
though weak, were detected between other school climate variables (safety, meaningful
participation, school awareness, and risky behavior approval) and victimization. While these are
weak associations, multivariate analyses later demonstrate that these components of school
climate play a significant role in decreasing the likelihood of being victimized in school.
Table 31. Correlational Analyses of School Climate, Deployment and Victimization
1 2 3 4 5 6 7 8
1. Belonging -- .41** .24** .36** .07** .22** -.03** -.17**
2. Relationships .41** -- .08** .43** .19** .26** .02* -.12**
3. Safety .24** .08** -- .01 .03** .08** .01 -.06**
4. Meaningful
Participation
.36** .43** .01 -- .14** .19** .02 -.04**
5. School
Awareness
.07** .19** .03** .14** -- .13** .04** .07**
6. Risky
Behavior
.22** .26** .08** .19** .13** -- .01 -.05**
7. Deployment -.03** -02* -.01 .02 .04** .01 -- .14**
8. Victimization -.17** -.12** .06** .04** .07** .05** .14** --
*Significant X
2
for differences by military-connected status, p < .05
**Significant X
2
for differences by military-connected status, p <.001

In order to examine the complexity of school climate’s relationship with victimization,
multivariate analyses were conducted. Logistic regression analyses provided detailed evaluations
of school climate and victimization, since these models control for other key variables present in
military-connected schools (i.e. military connection and experiencing the deployment of a family
member. Prior to conducting the logistic regression analyses, a multiple linear regression
analysis was performed to evaluate multicollinearity among the variables in this study. The
regression coefficients were all greater than 1, signifying that multicollinearity was not violated.
Table 32 shows results for the logistic regression of victimization by school climate. The
results suggest that a supportive school environment is associated with decreased victimization

127
rates. Significant associations were found between victimization and all components of school
climate, except risky behavior approval. As expected, belonging and caring relationships are
associated with decreased odds of victimization. An increasing level of belonging is associated
with a decreasing likelihood of victimization (OR = .80; 95% CI = .76-.85). In addition, an
increasing level of caring relationships is associated with a decreased likelihood of victimization
(OR = .83; 95% CI = .77-.88). Table 32 also shows that students who attend schools where they
feel unsafe and feel that students and teachers lack respect for their family background have
slightly higher rates of victimization. Lack of safety is associated with an increased likelihood of
victimization (OR = 1.15; 95% CI = 1.09-1.20). Lack of respect for family is associated with an
increasing likelihood of victimization (OR = 1.25; 95% CI = 1.19-1.31). Unexpectedly, table 32
shows that increasing levels of meaningful participation is associated with an increasing
likelihood of victimization (OR = 1.07; 95% CI = 1.01-1.13), however, this odds ratio suggest
only a 7% increased probability of being victimized.
Even though school climate was accounted for in this logistic regression (see Table 32),
military connection still remained a significant predictor of victimization. These results suggest
that the stressors that surround a student who has a military parent or sibling lead to victimization
even when they attend schools with supportive school climates. As seen in table 32, controlling
for the effects of school climate, significant associations were found between demographics and
victimization. Table 32 indicates that having a parent in the military still plays a significant role
in being victimized. When compared to non-military students, students with a military parent are
27% more likely than non-military students to be victimized (OR=1.27; 95% CI = 1.08-1.49).
Also, while the result is not significant, having a military sibling is still associated with a 16%
increased odds of being victimized (OR=1.16; 95% CI=.93-1.44).

128
Notably, other demographic characteristics are also significant predictors of
victimization. With regards to gender, table 32 shows that male students are 35% less likely than
female students to be victimized in a school setting (OR = .65; 95% CI = .59-.70). In addition, in
reference to race/ethnicity Hispanic students are less likely to experience victimization than
white students (OR = .68; 95% CI = .61-.75). Moreover, grade level still plays a role in
predicting victimization, controlling for school climate. When compared to 7
th
graders, 9
th

graders and 11
th
graders are less likely to be victimized (OR = .70; 95% CI =.70-.88, OR = .60;
95% CI = .53-.66, respectively).
Table 32. Logistic Regressions of Victimization by Demographics and School Climate
(n=14,943)
Predictors Victimization
OR (95% CI)
Constant
Grade
7
th

9
th

11
th

5.07**

1.00
.79** (.70-.88)
.60** (.53-.66)
Gender
Female (reference)
Male

1.00
.65** (.59-.70)
Race/Ethnicity
White (reference)
Asian/AI/AN/PI/HI
Black
Mixed
Hispanic

1.00
.93 (.78-1.10)
1.01 (.77-1.38)
.99 (.84-1.16)
.68** (.61-.75)

Military-Connection
None (reference)
Parent
Sibling

1.00
1.27* (1.08-1.49)
1.16 (.93-1.44)
(Table 32 continued)

129
(Table 32 continued)
School Climate
Belonging
Relationships
Safety
Meaningful Part.
Respect for Family
Risky Behavior

.80** (.76-.85)
.83** (.77-.88)
1.15** (1.09-1.20)
1.07* (1.01-1.13)
1.25** (1.19-1.31)
.94 (.87-1.01)
Model Evaluation
% Correctly Predicted
-2 Log Likelihood
Cox & Snell R
2
Nagelkerke R
2
Block X
2
(df)
Model X
2
(df)
Goodness of fit
Homer/Lemeshow

72.2%
11990.58
.04
.06
448.39 (15)
448.39 (15)

15.14

Deployment, School Climate, and Victimization
In the next series of analyses, I accounted for another key variable unique to military-
connected schools, the experience of a family member’s deployment. A substantial literature
base identifies the deployment cycle as the most significant military stressor affecting the
schooling experiences (i.e. victimization) of students in military-connected schools. Hence, a
logistic regression was conducted to assess the effects of single and multiple deployments on
victimization, also accounting for the effects of school climate. As seen in Table 33, significant
associations between single and multiple deployments and victimization, even after controlling
for school climate, provide more evidence of the persisting effects of deployment on
victimization despite the experience of being in a supportive school. Table 33 shows that odds
ratios for multiple deployments are slightly larger than odds ratios for single deployments. A
single deployment is associated with an increasing likelihood of being victimized in the school
context (OR = 1.56; 95% CI = 1.31-1.86). Experiencing multiple deployments (2 or more) is
associated with a 60% increase in odds of being victimized (OR = 1.60; 95% CI = 1.40-1.84).

130
However, despite the experience of deployment, the results in Table 33 suggest that
school climate plays a significant role in likelihood of victimization. Table 33 indicates that all
components of school climate, except for risky behavior approval, are significantly associated
with victimization. An increasing level of belonging is associated with a decreased likelihood of
being victimized (OR = .79; 95% CI = .75-.85), while a higher degree of caring relationships is
associated with a decreased likelihood of being victimized (OR = .83; 95% CI = .77-.89). A
lower sense of safety is associated with an increased likelihood of being victimized (OR = 1.14;
95% CI = 1.09-1.20), while a lack of respect for family is associated with a 25% increase in odds
of being victimized in school (OR = 1.25; 95% CI = 1.18-1.32). Unexpectedly, an increasing
level meaningful participation is associated with increased likelihood of being victimized (OR =
1.08; 95% CI = 1.01-1.15).
Table 33. Logistic Regressions of Victimization by Demographics, Deployment, and School
Climate (n=14,943)
Predictors Victimization
OR (95% CI)
Constant
Grade
7
th

9
th

11
th

4.45**

1.00
.79** (.70-.90)
.62** (.55-.70)
Gender
Female (reference)
Male

1.00
.65**(.59-.71)
Race/Ethnicity
White (reference)
Asian/AI/AN/PI/HI
Black
Mixed
Hispanic

1.00
.88 (.72-1.06)
.97 (.71-1.32)
.98 (.82-1.16)
.68** (.61-.76)
(Table 33 continued)

131
(Table 33 continued)
Deployment
None (reference)
One
Two or more

1.00
1.56** (1.31-1.86)
1.60** (1.40-1.84)
School Climate
Belonging
Relationships
Safety
Meaningful Part.
Respect for Family
Risky Behavior

.79** (.75-.85)
.83** (.77-.89)
1.14** (1.09-1.20)
1.08* (1.01-1.15)
1.25** (1.18-1.32)
.96 (.88-1.04)
Model Evaluation
% Correctly Predicted
-2 Log Likelihood
Cox & Snell R
2
Nagelkerke R
2
Block X
2
(df)
Model X
2
(df)
Goodness of fit
Homer/Lemeshow

71.9%
10150.50
.05
.07
440.52 (15)
440.52 (15)

9.83

School Climate and Victimization among Military and Non-Military Students
As previously discussed in the section on bivariate analyses, multivariate analyses were
conducted to assess different victimization rates among non-military students, students who have
a military parent, and students who have a sibling in the military. As seen in Table 34, logistic
regression models of victimization by school climate were constructed in three subsamples.
These analyses made it possible to compare the potentially different effects of school climate on
victimization on each subsample of students. Contrary to expectations, the effects of belonging
on victimization are similar in magnitude in all three groups of students. In the case of belonging,
an increasing level of belonging is associated with a 21% decreased probability of being
victimized (OR=.79; 95% CI=.74-.85) in non-military students. Among students with a military
parent, the odds ratio for belonging is non-significant, yet similar to non-military students, an
increasing level of belonging is associated with a 13% decreased odds of being victimized

132
(OR=.87; 95% CI =.70-1.08). Among students with a military sibling, the odds ratio for
belonging is also non-significant, however, the odds ratio is similar to non-military students and
students with a military parent (OR=.86; 95% CI=.63-1.16).
Among non-military students, there is a slightly weaker effect of caring relationships on
victimization among military students (students with a military parent and students with a
military sibling) when compared to non-military students. An increasing degree of caring
relationships is associated with 27% decreased odds of victimization (OR=.73; 95% CI=.50-
1.07) among students with a military sibling and a 34% decreased odds of victimization
(OR=.66; 95% CI=.50-.86) among students with a military parent. However, among non-military
students, an increasing degree of caring relationships is associated with a 15% decreased odds of
victimization (OR=.85; 95% CI=.79-.92).
The effect of safety on victimization is similar in magnitude across the three groups of
students. Among non-military students, lack of safety is associated with a 14% increased
likelihood of victimization (OR=1.14; 95% CI = 1.0. Lack of safety is also associated with a
20% increased likelihood of victimization among students with a military parent (OR=1.20 95%
CI = 1.01-1.41). Although the odds ratio was non-significant, lack of safety is also associated
with a 18% increased likelihood of victimization among students with a military sibling
(OR=1.18; 95% CI = .91-1.53). The effect of respect for family on victimization is also similar
in magnitude across the three groups of students. As seen in Table 34, lack of respect for family
is associated with a 25% increased likelihood of victimization for non-military students
(OR=1.25; 95% CI = 1.19-1.32), a 22% increased likelihood of victimization for students with a
military parent (OR=1.22; 95% CI = 1.01-1.48), and a 30% increased likelihood of victimization
for students with a military sibling (OR=1.30; 95% CI = .99-1.70).

133
Table 34. Logistic Regressions of Victimization by School Climate (n=14,943)
Predictors Non-Military
OR (95% CI)
Military Parent
OR (95% CI)
Military Sibling
OR (95% CI)
Constant
Grade
7
th

9
th

11
th

4.53**

1.00
.81** (.71-.93)
.62** (.54-.70)
2.40

1.00
.78 (.52-1.17)
.82 (.54-1.26)
12.99**

1.00
.45* (.21-.96)
.31** (.15-.64)
Gender
Female (reference)
Male

1.00
.64** (.58-.71)

1.00
.65* (.46-.90)

1.00
.69 (.42-1.13)
Race/Ethnicity
White (reference)
Asian/AI/AN/PI/HI
Black
Mixed
Hispanic

1.00
.89 (.72-1.10)
.98 (.68-1.43)
.96 (.79-1.16)
.65** (.58-.74)

1.00
1.09 (.63-1.90)
1.18 (.61-2.29)
1.26 (.77-2.07)
1.37 (.89-2.10)

1.00
.41 (.15-1.09)
1.14 (.29-4.40)
1.00 (.45-2.20)
.49 (.28-.88)
Deployment
None
One
Two or more

1.00
1.62** (1.32-1.99)
1.59** (1.32-1.91)

1.00
1.73* (1.01-3.00)
1.92* (1.27-2.91)

1.00
1.44 (.77-2.69)
1.63 (.94-2.81)
School Climate
Belonging
Relationships
Safety
Meaningful Part.
Respect for Family
Risky Behavior

.79** (.74-.85)
.85** (.79-.92)
1.14** (1.07-1.20)
1.07 (.1.00-1.15)
1.25** (1.19-1.32)
.96 (.88-1.05)

.87 (.70-1.08)
.66* (.50-.86)
1.20* (1.01-1.41)
1.13 (.90-1.42)
1.22* (1.01-1.48)
1.02 (.76-1.35)

.86 (.63-1.16)
.73 (.50-1.07)
1.18 (.91-1.53)
1.10 (.79-1.53)
1.30 (.99-1.70)
.76 (.49-1.18)
Model Evaluation
% Correctly Predicted
-2 Log Likelihood
Cox & Snell R
2
Nagelkerke R
2
Block X
2
(df)
Model X
2
(df)
Goodness of fit
Homer/Lemeshow

71.3%
8823.12
.05
.07
368.29 (15)
368.29 (15)

9.72

77.8%
881.84
.05
.07
43.02 (15)
43.02 (15)

4.32

75.6%
417.52
.08
.12
35.11 (15)
35.11 (15)

13.33

School-Level Effects
Additional analyses were conducted to address the aim of assessing the effects of school-
level factors on victimization. Contrary to expectations, there were no significant associations
detected between military concentration and each form of victimization. In addition, analyses

134
assessing the role of school academic achievement and school poverty level found no significant
effects on victimization.
Discussion
Utilizing a population sample of secondary students in eight military connected school
districts, this study assesses victimization rates among students by military connection (non-
military students, students with a military parent, and students with a military sibling).
Associations between multiple components of school climate and victimization among military
and non-military students were also examined. In addition, relationships between deployment
and victimization are also evaluated. The results of this study indicate that multiple components
of school climate reduce the likelihood of being victimized, even when accounting for students’
experiences with the deployment of a family member. In addition, multivariate analyses indicate
that the magnitude of associations between school climate and victimization are similar among
non-military students, students with a military parent, and students with a military sibling.
Victimization Rates by Military Connection
The bivariate analyses in this study show that students with a military parent have higher
rates of victimization than students with a military sibling and non-military students. This finding
supports a recent epidemiological study that found that adolescents with a military parent have
higher rates of risky behaviors, including violent behavior, than non-military students. Reed and
colleagues (2011) found that adolescent boys with at least one parent in the military are at
elevated risk of engaging in physical fighting, carrying a weapon, and joining a gang.
These findings suggest that compared to civilian secondary students, secondary students
with a military parent are targets of non-physical and physical victimization in civilian public
schools. One explanation may be that civilian peers and school staff lack awareness and

135
sensitivity to the military-specific challenges of military children, and in turn, how these
challenges are manifested in social, emotional, and behavioral problems in school. Studies have
found that military children experience unique stressors stemming from military-specific life
events (i.e. deployment, reintegration, veterans’ war trauma) and as a result, often have
psychological challenges as well as academic and social functioning issues at school (Chandra et
al., 2010; Chartrand & Seigel, 2007; Hoshmand & Hoshmand, 2007; Huebner et al., 2009;
Jordan et al., 1992; Rosenheck & Nathan, 1986; Solomon, 1988). One significant military
stressor is multiple school transitions. Studies have found that military children cope with more
school transitions than civilian children and constantly readjust to new school and community
contexts (Bradshaw et al., 2010). Furthermore, qualitative studies have suggested that some
civilian school environments lack the knowledge and resources to facilitate healthy school
transitions and respond to parental deployment (e.g. excused absences) (Chandra et al., 2010:
Mmari et al., 2009). In some cases, a recent study found that civilian peers ridicule and
physically victimize military children (Mmari et al., 2009).
School Climate and Victimization
The findings in this study suggest that a supportive school environment can reduce the
probability of being victimized. Unlike previous studies, this study provides multiple constructs
to represent school climate—belonging, caring relationships, meaningful participation, risky
behavior disapproval, and safety. Multiple components of school climate, including belonging,
caring relationships, and safety, were significantly related to victimization.
Not surprisingly, this study indicates that caring relationships had significant associations
with victimization. This finding supports research that has linked caring-teacher relationships
with the reduced probability of physical and non-physical victimization. When students perceive

136
that teachers and other adults are providing consistent social and emotional support, students are
more likely to adopt positive attitudes about seeking help from school staff for bullying and
threats of violence among peers (Elliot et al., 2010). Caring teacher-student relationships also
help facilitate active student engagement in monitoring and reporting school violence
(Brookover, Schweitzer, Scheider, & Beady, 1978; Hoy & Hunnum, 1997). This study also
found a significant association between belonging and victimization. This finding suggests that
when students feel more connected to and/or less alienated from a school community, they are
less likely to be victimized by peers.
In addition to caring-relationships, respect for student’s family background played a
significant role in reducing the likelihood of victimization. This finding provides empirical
evidence for a large body of theoretical work on culturally responsive schooling. These theorists
posit that teachers and other school staff and peers need to be more knowledgeable of the unique
cultural and family backgrounds of minority students to ensure more positive social and
emotional outcomes in school (Hernandez-Sheets, 2003 and 2009; Ladson-Billings, 1994). In
culturally-responsive schools, peers, teachers, and other school adults are aware of the unique
family and cultural backgrounds of students, and hence can alter school and classroom
environments to address their unique needs (Ladson-Billings, 1994). In the case of military
children, this may include adding military families to curriculum (i.e. history lessons), creating
bulletin boards and other visual displays to spotlight deployed family members, assemblies, and
transition rooms (De Pedro et al., 2011).
Unexpectedly, this study found a weak association between meaningful participation and
victimization. This finding challenges theoretical work and empirical studies on school climate,
which assumes that a high level of school engagement is associated with academic success and a

137
reduced probability of victimization (Brookover et al., 1978; Hoy & Hunnum, 1997). However,
this body of literature does not account for other components of school climate, which was
accomplished in this study. This suggests that belonging, caring relationships, safety, and respect
for student’s family and cultural background may be more significant elements of school climate
with regards to victimization. In addition, this study found no association between risky behavior
disapproval and victimization. This finding challenges studies that have found lower rates of
school violence when schools utilize interventions for adopting an anti-drug and substance abuse
culture.
Deployment and Victimization
The findings indicated that single and multiple deployments were significantly associated
with victimization. As expected, the findings also indicate that the probability of being
victimized in school becomes greater when students experience multiple deployments. These
findings support recent research where the experience of a parental deployment was a significant
predictor of school violence perpetration and victimization (Reed et al., 2011). There is a clear
need for theoretical frameworks explaining the association between deployment and school
victimization. It is possible that the psychological strain of deployment may have a direct
influence on the behavior of military children and adolescents. Research has found that military
children and adolescents exhibit elevated externalizing behaviors and aggression in the
household and school contexts, hence attracting opportunities to be victimized (Flake et al.,
2009). In addition, military adolescents have had multiple school transitions through elementary
and secondary schools (Bradshaw et al., 2010). Hence, students may be more likely to victimize
military children, who are often newcomers in a school community. This finding may also
suggest that the emotional and psychological strain of prolonged and multiple deployments may

138
weaken the resiliency of children and adolescents with deployed family members, making them
vulnerable to victimization. More research is necessary to explain how and why having a
deployed parent can be related to being victimized or engaged with violent behaviors at school.
Another significant finding is that school climate still plays a significant role in reducing
victimization rates, even accounting for deployment. In fact, the odds ratios between each school
climate component and victimization in the deployment model (see Table 33) are similar to the
odds ratios in the non-deployment model (Table 32). This finding supports theoretical work and
empirical studies that have found a supportive school environment significantly reduces the
likelihood of victimization, even among at-risk and high-stress student populations (Astor,
Meyer, & Behre, 1999).
School Climate and Victimization among Military and Non-Military Students
Detailed analyses of the relationships between multiple components of school climate
and victimization within subsamples of non-military students, students with a military parent and
students with a military sibling were conducted in this study. A recent study provided normative
rates comparing the school violence and victimization rates of military-connected and civilian
youth (Reed et al., 2011). To date, there have been no studies that have compared associations
between school climate and victimization between different demographic groups. Given the
distinct stressors that military adolescents experience, it is notable that a supportive school
environment (belonging, caring relationships, safety, meaningful participation, and respect for
family and cultural background) can influence victimization of military-connected adolescents
similarly to non-military students. It is possible that school staff and students in military-
connected schools are attuned to the challenges of military life due to their proximity to military
installations and exposure to military students and their culture. Daily experiences with military

139
culture may have enhanced the awareness of school staff and peers to respond to the needs of
military children in culturally sensitive ways. Future empirical investigations can focus on a
school’s awareness of military culture and its impact on the victimization and violence rates of
military children.
Unique to this study are comparative multivariate analyses of school climate and
victimization among three groups of students, non-military students, students with a military
parent and students with a military sibling. There is a growing body of research on the emotional
and psychological outcomes of military children and adolescents; only one study has directly
assessed the outcomes of students with a military sibling (Rodriguez & Margolin, 2011).
Rodriguez and Margolin (2011) found that adolescents have unique social and emotional
challenges when a sibling is deployed, stemming from parental stress and social alienation and
the disruption of a sibling bond. More research is needed to further examine how other school
environmental factors (in addition to school climate) can help promote positive social and
emotional outcomes among students with siblings serving in the military.
Summary
Overall, the results of this study show that there are significant relationships between
school climate and victimization. These relationships are still significant, even accounting for
single and multiple deployments. This study also indicates that students with a military parent
had higher rates of physical and non-physical victimization. Detailed multivariate analyses show
that associations between school climate and victimization are similar in magnitude among
military-connected students (students with a military parent, students with a military sibling) and
non-military students. In addition, this study presents the first quantitative study, to date,
examining relationships between school climate and victimization among students with a

140
military sibling. The overall results provide further evidence that future theories, educational
policies, and school-based interventions be focused on enhancing the social and emotional
climate of military-connected school environments.
Limitations
There are a few limitations in this study that are worthy noting before generalizing these
findings to school climate interventions in military-connected schools. First, due to the cross-
sectional nature of this study, cause and effect relationships could not be assessed. Future studies
utilizing longitudinal data can be conducted to determine if school climate has a causal
relationship with victimization among military and non-military students. Second, this study
examined associations between school-level factors and victimization, and no significant
relationships were found. Academic achievement and poverty indicators were school-level
variables, assigned to all sampled students within a school. Future empirical work would need to
include student-level academic achievement and poverty data, to ensure that variation among
students within a school can be more accurately represented.
Implications
The findings from this study provide a detailed understanding of school climate and
victimization in military-connected schools. A large number of studies have found school
climate can reduce rates of school violence and student victimization in different contexts. This
is the first empirical investigation of school climate and victimization in military-connected
schools. Future studies are needed to understand this relationship among military and non-
military student populations in other military-connected school and district contexts. These
studies would require the use of a theoretically-driven model of school climate to determine
detailed analyses concerning caring relationships, belonging, family respect, safety, and risky

141
behavior disapproval. More importantly, these multivariate analyses would need to account for
military-specific stressors such as deployment and school transitions.
The findings in this study provide further support for including school climate in current
state accountability systems. Schools are embedded in a high stakes accountability context (e.g.
No Child Left Behind) intended to leverage the academic achievement of traditionally
underserved and historically low achieving minority groups (Cohen et al., 2010; Cochran-Smith,
2005). In a survey of state school climate policies, Cohen and colleagues (2010) found that
school climate improvement and social and emotional skill building are excluded from general
state accountability systems in most U.S. states, leaving it up to schools to voluntarily pursue
climate issues. This policy context stands in conflict with a large body of educational research
linking supportive school environments with reduced rates of bullying and victimization and
links between positive social and emotional outcomes with academic achievement (Cohen et al.,
2011). The results of this study provide more support for a growing research agenda outlining the
relevance of social and emotional outcomes (i.e. victimization and violence) to the current
educational policy trend of improving academic outcomes among minority and at-risk student
populations.
In summary, this study fills gaps in the military-connected school and school climate
literatures and has implications for future research, theory, and educational practice. Future
studies are needed to investigate how multiple components of school climate reduce rates of
victimization among military and non-military students, in civilian public schools. Studies with
longitudinal research designs are needed to understand the cause and effect relationships between
multiple components of school climate and victimization. Further, qualitative studies are needed
to understand how teachers and other school staff in military-connected schools create

142
supportive, military culture-friendly environments. These studies can help generate effective
local school climate practices and strategies specific to military children.

143

Chapter Five:
Integration and Implementation of Findings from Three Studies
Purpose of the Studies
Research and theoretical work conducted across academic disciplines has found that a
supportive and caring school climate can promote positive academic, social and emotional, and
psychological outcomes among students (Brand, Feldner, Shim, Seitsinger, & Dumas, 2003;
Cohen, McCabe, Michelli, & Pickeral, 2009; Zullig, Koopman, Patton, & Ubbes, 2010). In
particular, these studies have found that a supportive school climate can promote well-being and
curb victimization and negative mental health outcomes (i.e. depression, suicidal ideation)
among students from different geographic and cultural contexts, even among students
experiencing psychological strain in their households and neighborhoods (Hoy & Hunnum,
1997; Jia, Way, Ling, Yoshikawa, Chen, Hughes, Ke & Lu, 2009; Modin & Ostberg, 2009).
Recent reviews in educational research have generated theoretically driven models that
incorporate multiple components of school climate (Cohen et al., 2009; Zullig et al., 2010).
These components include belonging, caring relationships, safety, and respect for a student’s
family background, meaningful participation, and risky behavior disapproval. To date, empirical
studies have not yet utilized conceptual models and measures incorporating these components of
school climate.
Researchers have not yet examined the role of a supportive and caring school climate on
the social and emotional outcomes of military students. This is surprising given the long wars in
Iraq and Afghanistan and the significant presence of military students in civilian public school
districts throughout the United States. According to the Department of Defense Educational
Activity (DoDEA), since 2002, over 2 million children have experienced a parental deployment.

144
Presently, there are an estimated 1.2 million school-aged military children, while another
625,000 of National Guard, and 705,000 children of Reserve members attend civilian public
schools. In addition, 214 civilian public school districts are comprised of a significant minority
of military students (about 400 or more than 4% of total student enrolment) (Kitmitto et al.,
2011).
A large number of studies have found that military children experience military life
stressors that could potentially impact their social and academic functioning in schools
(Chartrand and Seigel, 2007; Hoshmand & Hoshmand, 2007; Huebner, Mancini, Bowen, &
Orthner, 2009). During periods of deployment, studies have found adolescents adopt more
household roles and responsibilities, cope with the fear of a family member’s death or injury, and
are exposed to the stress and anxiety of a left-behind parent (Faber, Willerton, Clymer,
MacDermid, & Weiss, 2008; Huebner et al., 2007). In the reintegration periods, military
adolescents cope with re-establishing relationships with a deployed parent and in some cases, are
exposed to the physical and psychological war trauma of a returning veteran (Faber, Willerton,
Clymer, MacDermid, & Weiss, 2008; Huebner et al., 2007). In addition, DoDEA reports that
military students encounter multiple residential relocations and school transitions. Typically,
military students have on average nine school transitions throughout the course of their
elementary and secondary schooling. In addition, military students are three times as likely as
civilian students to experience a school transition. Despite these stressful experiences and
concerning outcomes, recent studies have found that supportive communities (i.e. military bases,
military-impacted neighborhoods) with easy access to military-specific health care and other
social services can help promote strategies among military children and adolescents to cope with

145
the stressors of deployment (Flake, Davis, Johnson, & Middleton, 2009; Hoshmand &
Hoshmand, 2007).
Recent studies have examined the social and emotional experiences of military students
in civilian operated schools. On one hand, these studies have uncovered localized attempts by
school staff to make school environments more welcoming and caring of military students. Some
school staff at military-connected schools have developed homegrown practices (i.e. clubs
geared to military students, displays and bulletin boards highlighting deployed parents) to
facilitate the belongingness of military students. However, on the other hand, these same studies
have highlighted how some school staff have struggled with responding to the social and
emotional challenges of military students (Bradshaw et al., 2010; Chandra et al., 2010; Mmari et
al., 2009). School staff often lack awareness surrounding military life events, evidence-based
strategies to address temporary separation from a deployed parent or sibling, and school-wide
procedures for facilitating easy school transitions. The preceding studies have begun to fill in the
current gap in our knowledge on the role of a supportive school climate in promoting positive
social and emotional outcomes among military children. However, these studies utilized small
samples and did not incorporate theoretically-driven conceptual models of school climate.
Hence, findings from these studies do not provide results concerning the role of school climate in
promoting positive social and emotional outcomes among military students.
So far, large gaps in our current knowledge of school climate and military-connected
schools still exist. To date, no studies have examined how school climate could promote well-
being and curb victimization and negative mental health outcomes among military adolescents. It
is also not yet known if military students have, in fact, more negative school climate perceptions
than non-military students. In addition, it is not known if school climate plays a role in

146
promoting well-being, curbing rates of negative mental health outcomes and victimization
among military and non-military students in military-connected schools. Knowledge from these
studies would expand research on school climate and military-connected schools.
This multiple manuscript dissertation addresses the current gaps in research. This
dissertation first compares the school climate perceptions of military and non-military students
enrolled in the same schools. Components of school climate, which include belonging, caring
relationships, meaningful participation, safety, risky behavior disapproval, and respect for family
and cultural background are assessed among non-military students, students with a military
parent, and students with a military sibling. Next, this dissertation investigates associations
between school climate and mental health (i.e. well-being, depression, and suicidal ideation),
accounting for the experience of deployment. More detailed multivariate analyses are conducted
to determine if relationships between school climate and mental health are different among
military and non-military students. Last, this dissertation examines relationships between school
climate and victimization, accounting for the experience of deployment. Further multivariate
analyses are performed to assess whether the relationship between school climate and
victimization is different for military and non-military students. This study utilizes a population
sample of 7
th
, 9
th
, and 11
th
graders in eight military-connected school district and unlike previous
studies, utilizes a theoretically-driven model of school climate. The major findings of this study
and implications for school reform, policy, and recommendations for future research are
discussed in this chapter.

147
Overall Findings and Links to Current Research on School Climate and Military-
Connected Schools
School Climate Perceptions in Military-Connected Schools
Study 1 of this dissertation provided detailed analyses of school climate perceptions
among non-military students, students with a military parent, and students with a military sibling.
This study examined students’ perceptions of multiple components of school climate—
belonging, caring relationships, safety, meaningful participation, respect for a student’s family,
and risky behavior disapproval. One important finding is that within the same schools military-
connected students had more negative perceptions of each component of school climate than
non-military students. This finding supports existing research on military-connected schools.
These studies have found that military-connected students experience social and emotional
challenges in their school environments, stemming from military life events (i.e. deployment)
(Bradshaw et al., 2010; Chandra et al., 2010; Mmari et al., 2009). This finding also suggests that
school staff in civilian public schools may lack awareness and strategies for responding to the
social and emotional challenges of military children.
While military students consistently reported more negative school climate perceptions,
the differences between military and non-military were small (less than 5% percent difference).
This finding suggests that while some military students may have negative social and emotional
experiences of their school environments, there may also be a sizeable proportion of resilient
military students who have positive experiences of their school environments. In addition, there
may also be a significant proportion of non-military students in the same schools who have
negative experiences of their school environments. This finding also uncovers a limitation in
recent studies on military-connected schools and military students. These studies utilized

148
purposive samples of military students and examined their social and emotional challenges of
military students, while ignoring the potential challenges of non-military students in the same
schools (Bradshaw et al., 2010; Chandra et al., 2010; Mmari et al., 2009. The authors’ exclusive
investigation of the social and emotional challenges of military-connected students, but not their
social and emotional strengths, supports a current assumption by researchers and practitioners
that military children have negative outcomes.
School Climate and Mental Health among Military and Non-Military Students
Study 2 of this dissertation provided multivariate analyses examining the role of school
climate in the well-being, depression and suicidal ideation rates of military and non-military
students in the same schools. One important finding in this study is that there are differences in
the mental health outcomes of students with military parent and students with a military sibling.
Most previous empirical studies have focused on the psychological outcomes and experiences of
children with a military parent. To date, only one published study has focused on the experiences
of adolescents or children with a military sibling (Rodriguez & Margolin, 2011). Rodriguez &
Margolin (2011) found that adolescents with a military sibling have unique military-related
stressors, such as coping with the repeated and prolonged separation from a sibling, the
emotional distress of left-behind parents, and alienation from social supports. Future studies need
to analyze the mental health outcomes of adolescents with a military sibling and adolescents with
a military parent and assess potential differences.
Results from this study also indicated that a supportive and nurturing school climate may
benefit the mental health outcomes of all students. This study is the only known empirical
investigation to show that school climate has an impact on the mental health outcomes of
students in military-connected schools. For all students in this study, caring relationships and

149
belonging played a major role in all three indicators of mental health (well-being, depression,
and suicidal ideation) among students in military-connected schools. This finding supports
research and theoretical work on the protective roles that support from teachers and other school
adults and belonging play in the mental health outcomes of students in different school contexts
(Cohen et al., 2009; Zullig et al., 2010).
Unexpectedly, school safety played a small role in promoting positive mental health
outcomes and reducing negative mental health outcomes among students in this study. This
finding challenges numerous studies that have found that a perception of school order can serve
as significant protective factors for internalizing and externalizing behavioral and mental health
problems among middle and high school students (Kuperminc, Leadbeater, & Blatt, 2001). This
finding also challenges the current emphasis among schools that focus exclusively on school
safety interventions as a means to improve the overall school’s climate. Future studies need to
continue evaluating the role that school safety play in mental health and risky behaviors, while
also accounting for other components of school climate.
Results in this study further indicated that school climate had a significant impact on the
mental health outcomes of military students to the same extent as non-military students. For both
military and non-military students, the most significant components of school climate were
caring relationships and belonging. Both components promoted well-being and reduced rates of
depression and suicidal ideation. These findings suggest that school climate interventions
designed to address the mental health outcomes of non-military students may also be beneficial
to the mental health of military students. In the same vein, school climate interventions tailored
to the unique challenges of military life (i.e. transition rooms and centers) may also promote
well-being among non-military students. These results challenges research on military students

150
and military-connected schools (Bradshaw et al., 2010; Chandra et al., 2010; Mmari et al., 2009).
These studies explored the social and emotional supports of civilian public school environments,
yet concluded that they did not significantly affect the mental health issues of students from
military families.
The findings of this study also indicated that school climate even plays an important role
in the mental health outcomes of students experiencing the stress of a family member’s
deployment. These results support recent empirical and theoretical work showing that supportive
and responsive schools have a protective effect on the mental health outcomes of students,
experiencing psychological strain (Astor et al., 2011). Astor and colleauges (2011) found that
teachers, principals, school psychologists, counselors, and other school staff utilized a data-
driven decision making procedure to identify students with post-trauma symptoms in the
aftermath war. This process enabled teachers and other school staff to develop more caring
relationships between teachers and students. In addition, this process helped prevent the
academic failure and disengagement of students diagnosed with post-trauma symptoms by
avoiding mislabeling and placement in special education and other inappropriate academic and
behavioral intervention programs.
Victimization in Military-Connected Schools
Study 2 of this dissertation examined the role of school climate in the victimization of
students in military-connected schools. One important finding is that military-connected
students, especially students with a military parent, had consistently higher rates of victimization,
when compared to non-military students. These findings support a recent epidemiological that
found that students who have a parent serving in the military engage in violent behaviors in

151
schools and are physically and non-physically victimized to a greater extent than their civilian
peers (Reed, Bell, & Edwards, 2011).
Findings from recent qualitative studies may provide some explanation for the higher
rates of victimization and violence of military students. These studies suggested that lack of
awareness and support by schools staff and peers contribute to military students being more
vulnerable to becoming victims of non-physical and physical acts of violence. Mmari and
colleagues (2010) found that anti-war sentiments provided impetus for civilian students to
commit acts of violence against military adolescents. Also, military students had difficulty
making friends and developing caring and nurturing relationships at school due to teachers’ and
peers’ lack of awareness of military life and culture. In addition, military students cope with
multiple school transitions and must constantly adjust to new school environments, including
school and classroom rules and procedures (Bradshaw et al., 2010). Often military-connected
school environments lack appropriate transition supports and procedures, and hence, lead to
academic issues and social challenges with peers.
Similar to Study 2’s results on mental health outcomes, Study 3 also indicated that there
are differences between students with a military parent and students with a military sibling with
regard to victimization rates. Consistently, on almost every item, students with a military parent
had slightly higher rates of being victimized at school. This finding provides researchers with
more evidence that having a parent in the military and having a sibling in the military are
qualitatively different experiences that lead to different outcomes. Students with a military parent
may be dealing with multiple and prolonged parental separation, adopting household roles and
responsibilities typically reserved for adults, and in some cases, a veteran parents physical or
psychological trauma. While some military children and adolescents develop resilience from the

152
stressors of having a military parent, other military students may develop externalizing
behaviors. Future empirical studies need to examine the risky behavior outcomes of students
with parents serving in the military and students with siblings serving in the military as separate
groups. Moreover, future empirical studies need to develop explanations for why outcomes may
be different for students who have a military parent and students who have a military sibling.
In addition, this is one of the first empirical studies to assess the role of school climate in
the victimization of students attending military-connected schools. Results from this study
indicate that secondary students in military-connected schools are victimized less frequently
when they are in a supportive and caring school environment. Several findings add to current
research linking school climate to violence and victimization. Similar to Study 2’s findings, it is
the dimensions of belonging and caring relationships played the most important role in reducing
victimization among all students in this study. This finding supports research on the role of
teachers’ social and emotional support and belonging in promoting pro-social behaviors and
conflict resolution skills (Bond, Butler, Thomas, Carlin, Glover, Bowes, & Patton, 2007; Wilson,
2004; Bonny, Britto, Klostermann, Hornung, & Slap, 2000).
Also, as expected, a lack of safety increased victimization. This supports a substantial
literature on school safety and violence, which has found that unsafe schools where school
violence rules and procedures are not commonly understood by school staff and students are
characterized by high rates of both physical and non-physical victimization (Marachi, Astor, &
Benbenishty, 2006; Wang, 2009). In addition, unique to this study, lack of respect for a student’s
family background played a substantial role in reducing victimization in military-connected
schools. This finding provides empirical support for theoretical work on culturally responsive
schooling (Hernandez-Sheets, 2003 and 2009; Ladson-Billings, 1994). This area of research has

153
found that schools prevent disengagement of minority students when teachers and peers are
aware of their family dynamics and cultural issues.
This is also one of the first empirical studies that compares the degree to which school
climate impacts victimization among non-military students, students with a military parent, and
students with a military sibling. For all three groups of students, belonging, safety, respect for
student’s family background played major roles in reducing victimization. Also, while caring
relationships also played a significant role in the victimization outcomes of all three groups of
students, the impact is slightly larger among students with a military parent. These findings
support past research, which indicates that a supportive and responsive school climate has a
significant impact on different groups of students by race, ethnicity, gender, and other cultural
identities (i.e. military connection) (Cohen et al., 2009; Zullig et al., 2009). In addition, this study
also shows that school climate plays a significant role in reducing victimization even when
students are experiencing the stress of a family member’s deployment. Findings from Study 3
indicate that multiple components of school climate, especially caring relationships, belonging,
and respect for a student’s family background, may lead to less violent schools even when
students are psychologically impacted by deployment.
Implications for Future Educational Reform and Practice
Research conducted in this dissertation is positioned within the context of a research
agenda that seeks to examine the role that supportive schools play in the social, emotional and
academic outcomes of military-connected students. As seen in Figure 1, De Pedro and colleagues
(2011) found that the bulk of research studies on military children and adolescents have focused
on risk issues related to military life events (i.e. deployment), psychological and social supports
in the family and community, and their social, emotional and psychological development. As

154
seen in the right hand of Figure 1, future research on military children needs to focus on how
factors within the school environment such as principal leadership, peer and teacher awareness
and support, and a supportive school climate can influence the social, emotional and academic
outcomes of military children. The studies in this dissertation specifically examined the role that
a supportive school climate plays in key social and emotional outcomes known to impact
academic functioning—mental health and victimization. In order to more accurately assess how
school climate influences the mental health and victimization outcomes of military-connected
students, this dissertation utilizes a conceptualization of school climate, driven by decades of
theoretical and empirical work. This model is comprised of multiple components, including
belonging, caring relationships, meaningful participation, safety, respect for a student’s family,
and risky behavioral disapproval.
Findings from this dissertation begin to fill in the knowledge gap on the role of
supportive school environments in the social and emotional outcomes of military-connected
students. These studies suggest that a supportive school climate plays a significant role in the
social and emotional outcomes of military-connected students, specifically promoting well-being
and reducing rates of depression, suicidal ideation, and victimization. As seen in the right hand
side of Figure 1, future studies could examine how principal leadership and peer and teacher
awareness and support can enhance a supportive school climate. Moreover, future studies and
practice could focus on factors outside the school context. Such research could focus on how
school reform initiatives and support from universities, community organizations, and military
bases enhance the capacity of military-connected schools to develop supportive school climates.

155
A Whole School Approach. Findings in these studies also have relevance for a large
body of research that examines the role of supportive school climates in the social and emotional
outcomes of all students. The results of this dissertation showed that school climate still
promoted positive mental health outcomes and curbed negative mental health outcomes and
victimization rates among all students (military and non-military, with a deployed family
member and without a deployed family member). Being in a supportive and caring school
environment may be beneficial to all students, not only students experiencing the most severe
psychological stress, but also students throughout the mental health continuum. Hence, to benefit
the mental health outcomes of all students in a school, a whole school, social and emotional
climate intervention students could lead to large-scale reductions in depression, suicidal ideation,
and victimization rates.
However, schools typically apply a narrow intervention oriented approach to alleviating
mental health issues and victimization. Predominant in the clinical psychological, psychiatry, and
social work practice fields, an intervention approach only targets a small number of students with
the most severe levels of negative mental health and risky behaviors, while ignoring the social
and emotional needs of other students (Espelage & Swearer, 2009). Narrow school-based
interventions include school counseling, referrals to district mental health clinics, special
education services for severe behavioral and social issues, and isolated evidence-based practices
in classroom and counseling sessions.
A whole-school approach stems from primary prevention approaches already
implemented in public health practice and research. In the context of public health, a primary
prevention approach occurs when measures are taken to prevent diseases or injuries and promote
general well-being instead of exclusively curing or treating the symptoms (Espelage & Swearer,

156
2009). Schools that adopt a primary prevention approach seek reductions in mental health issues
and violence among all students. School wide approaches can be used to transform a school’s
social and emotional climate for all students. In the context of military-connected schools, efforts
to enhance caring relationships and belonging among students can promote well-being and curb
victimization among military and non-military students.
In the context of educational research, studies have a comprehensive school reform
approach to change student outcomes on a large scale (Borman, Hewes, Overman, & Brown,
2003; Datnow, 2005; Datnow, Borman, Stringfield, Overman, & Castellano, 2003). A
comprehensive school reform (CSR) approach involves various components. First, it is a
systematic approach to planning, implementing, and evaluating school-wide efforts that
incorporate all aspects of a school, including instruction, management, and parent involvement
(CASEL, 2003). Second, CSR models suggest that the successful implementation of a school
climate improvement effort requires the active engagement of the school, community, youth,
teachers and parents (CASEL, 2003). Research has found that authentic community approaches
to youth risk outcomes may have more sustainability than top-down approaches to school
improvement (Thornton, Craft, Dahlberg, Lynch, & Baer, 2008). The concept of CSR is
reflected in current research-based and data-driven models of school climate improvement,
including models developed by the National School Climate Center (NSCC) and the Center for
Social and Emotional Learning (CASEL).
Integrating School Climate into Current Educational Reform. These studies’ overall
findings indicate that school environments provide not only academic support, but also promote
key social and emotional outcomes for students. Hence, it is necessary that current American
educational reform initiatives also focus on creating supportive school environments and

157
providing students with opportunities to build social and emotional skills. In addition, these
findings suggest that schools can broaden their societal purpose by moving beyond an academic
focus and including a clear purpose in promoting healthy social and emotional development of
students.
Historically, urban schools since the early 19
th
century have enacted visions of
developing the social and emotional development of students. In the late 19
th
century, Addams
(1902) theorized that a student’s moral and civic development depended on the fulfillment of
students’ personal needs. In practice, this approach would require public schools to partner with
community organizations to provide social services for children and their families. In 1889,
Addams established the Hull House in Chicago, while other settlement houses were established
in impoverished urban neighborhoods. The Hull House provided recently arriving immigrant
children and their families with social services, extra-curricular and after-school programs, job
training, and financial assistance (Elshtain, 2001).
More recently in the 1970’s, Comer developed and implemented the School Development
Program in the New Haven Public Schools District. Comer (1984) posited that the “tasks of the
home, social network, and school is to prepare children to function optimally as adults--to obtain
the skills necessary to earn a living for themselves and their families--to become responsible and
competent individuals and/or heads of households, including good child-rearers, and to become
responsible citizens and to find satisfaction in life” (p. 324). Comer (1984) established a school
model that responded to the psychological and financial stressors of New Haven public school
students and their families. Similarly, Dryfoos (1984) conceptualized the full service school. The
term full-service school encompassed the following: a) school-based primary health clinics,
youth service programs, community schools, and other innovative efforts to improve access to

158
health facilities for delivering services through partnerships with community agencies and b) a
shared vision of youth development and financial support from sources outside school systems,
particularly states and foundations.
While schools could potentially address social and emotional development in addition to
academic skills, the improvement of academic outcomes, especially among historically
oppressed minority student populations, has been the central aim of the American K-12
educational policy. In the No Child Left Behind (NCLB) policy context, state accountability
systems have driven the academic focus of public schools; public schools are accountable for
monitoring and improving the academic achievement of all students through extensive state
accountability mandates. In contrast, state accountability systems leave it up to schools to
voluntarily purse school climate improvement and the development of social and emotional skills
(Cohen et al., 2009).
This policy context contradicts school climate and social and emotional learning research
A report from by the Collaborative for Academic, Social and Emotional Learning (CASEL) has
theorized that a nurturing, safe, and supportive school climate facilitates social and emotional
learning (CASEL, 2003). Social and emotional skills include being effective problem solvers,
recognizing the consequences of one’s actions, taking and responsibility for one’s personal
health and well-being. In addition, students with social and emotional skills know how to
develop effective social relationships such as learning how to work in a group and how to
understand and relate to others from different cultures and backgrounds. They are also caring
individuals with concern and respect for others, develop good character, make sound moral
decisions, and behave in an ethical and responsible manner. When students adopt these skills,

159
they not only become functioning adults with technical skills necessary for employment, but they
also become engaged, responsible citizens in a democracy.
Figure 1. Dissertation Studies within the Context of a Research Agenda

Military-
specific
External
risk issues
Family and
community
supports
Social,
emotional,
and
psychological
development
of military
children
Principal
Leadership
Teacher
Awareness
and
Support
Peer
Awareness
support
Social
Outcomes
Academic
Outcomes
Emotional
Outcomes
Supportive
School
Climate
School Reform for Supportive School Climates and Military Cultural Group
Support from Universities, Community Organizations, and Military Bases
Societal Contextual Factors (e.g., popular attitudes
toward military)
Military Contextual Factors (e.g. war, military
branch)

160
Bibliography
Addams, J. (1902). Democracy and Social Ethics. New York: Macmillan.
Amen, D., Jellen, L., Merves, E., & Lee, R. (1988). Minimizing the impact of deployment
separation on military children: Stages, current preventative efforts and system
recommendations. Military Medicine, 153(9), 441-446.
Anyon, J. (1997). Ghetto Schooling: A Political Economy of Urban Educational Reform. New
York: Teachers College Press.
Astor, R. A., Meyer, H. A, & Behre, W. J. (1999). Unowned places and times: Maps and
interviews about violence in high schools. American Educational Research Journal, 36,
3-42.
Astor, R.A., Meyer, H., Benbenishty, R., Pat-Horencyk, R., Brom, D., Baum, N., Schiff, M., De
Pedro, K. (2011). The influence of the Second Lebanese War on Israeli students in urban
school settings: Findings of the Nahariya district-wide screening prototype. In Gallagher,
K.S., Brewer, D., Goodyear, R., & Bensimon, E. (eds.), International Handbook of
Research in Urban Education. New York: Taylor & Francis.
Austin, G., Bates, S., & Duerr, M. (2011). Guidebook for the California healthy kids survey part
I: Administration. San Francisco, CA: WestEd. Retrieved from
http://chks.wested.org/resources/chks_guidebook_1_admin.pdf on September 10, 2011.
Bond, L., Butler, H., Thomas, L., Carlin, J., Glover, S., Bowes, G., & Patton, G. (2007). Social
and school connectedness in early secondary school as predictors of late teenage
substance use, mental health, and academic outcomes. Journal of Adolescent Health, 40,
9-18.

161
Bonny, A.E., Britto, M.D., Klostermann, B.K., Hornung, R.W., & Slap, G.B. (2000). School
disconnectedness: Identifying adolescents at risk. Pediatrics, 106(5), 1017-1021.
Borman, G., Hewes, G., Overman, L., & Brown, S. (2003) Comprehensive school reform and
achievement: A meta-analysis. Review of Educational Research. 73(2), 125-230.
Bowen, G., Mancini, J., Martin, J., Ware, W., & Nelson, J. (2003). Promoting the adaptation of
military families: An empirical test of a community practice model. Family Relations,
52(1), 33-44.
Bradshaw, C.P., Sudhinaraset, M., Mmari, K., & Blum, R. (2010). School transitions among
military adolescents: A qualitative study of stress and coping. School Psychology Review,
39(1), 84-105.
Brand, S., Felner, R., Shim, M., Seitsinger, A., & Dumas, T. (2003). Middle school improvement
and reform: Development and validation of a school-level assessment of climate, cultural
pluralism, and school safety. Journal of Educational Psychology, 95(3), 570-588.
Brookover, W., Schweitzer, J., Schneider, J., Beady, C., Flood, P., & Wisenbaker, J. (1978).
Elementary school social climate and school achievement. American Educational
Research Journal, 15(2), 301-318.
Burrell, L.M., Adams, G.A., Durand, D.B., & Castro, C.A. (2006). The impact of military
lifestyle demands on well-being, army, and family outcomes. Armed Forces & Society,
33(1), 43-58.
Camp, W. (1990). Participation in student activities and achievement: A covariance structural
analysis. Journal of Educational Research. 83(5), 272-278.

162
Chandra, A., Martin, L.T., Hawkins, S.A., & Richardson, A. (2010). The impact of parental
deployment on child social and emotional functioning: Perspectives of school staff.
Journal of Adolescent Health, 46(3), 218-223.
Chartrand, M.M. & Seigel, B. (2007). At war in Iraq and Afghanistan: Children in US military
families. Ambulatory Pediatrics, 7(1),1-2.
Cochran, Smith, M. (2005). The new teacher education: For better or for worse? Educational
Researcher. 34(7), pp. 3-17.
Cohen, J. (2006). Social, emotional, ethical, and academic education: Creating a climate for
learning, participation, democracy, and well-being. Harvard Educational Review. 76(2),
201-237.
Cohen, J., McCabe, L., Michelli, N., & Pickeral, T. (2009). School climate: Research, policy,
practice, and teacher education. Teachers College Record, 111(1), 180-213.
Collaborative for Academic, Social, and Emotional Learning. (2003). Safe and Sound: An
Educational Leader's guide to Evidence-Based Social and Emotional Learning (SEL)
Programs. Chicago, IL.
Comer, J.P. (1984). Home-school relationships as they affect the academic success of children.
Education and Urban Society. 16, 322-337.
Cozza, S.J., Chun, R.S., & Polo, J.A. (2005). Military families and children during Operation
Iraqi Freedom. Psychiatric Quarterly. 76(4), 371-378.
Datnow, A. (2005). Sustainability of comprehensive school reform models in changing district
and state contexts. The Journal of Leadership for Effective and Equitable Organizations.
41(1), 121-153.

163
Datnow, A., Borman, G., Stringfield, S., Overman, L., & Castellano, M. (2003). Comprehensive
school reform in culturally and linguistically diverse contexts: Implementation and
outcomes from a four-year study. Educational Evaluation and Policy Analysis. 25(2),
143-170.
De Pedro, K., Astor, R.A., Benbenishty, R, Estrada, J.N., Smith, G.R., & Esqueda, M.C. (2011).
The children of military service members: Challenges, resources, and future educational
research. Review of Educational Research, 81(4), 566-618.
Dryfoos, J.G. (1995). Full service schools: Revolution or fad? Journal of Research on
Adolescence, 5(2), 147-172.
Eisenberg, M., Neumark-Sztainer, D., & Perry, C. (2003). Peer harassment, school
connectedness, and academic achievement. Journal of School Health, 73(8), 311-316.
Elliot, M., Cornell, D., Gregory, A., & Fan, X. (2010). Supportive school climate and student
willingness to seek help for bullying and threats of violence. Journal of School
Psychology, 48, 533-553.
Elshtain, J.B. (2001). The Jane Addams Reader. New York: Basic Books.
Espelage, D. & Swearer, S. (2007). A social-ecological model for bullying prevention and
internvetion. In Jimerson, S. (ed.), International Handbook of Bullying in Schools. New
York: Taylor & Francis.
Fenning, P. & Rose, J. (2007) Overrepresentation of African-American students in exclusionary
discipline: The role of school policy. Urban Education. 42(6), 536-559.
Finn, J., & Rock, D. (1997). Academic success among students at risk for school failure. Journal
of Applied Psychology, 82(2), 221-234.

164
Flake, E.M., Davis, B.E., Johnson, P.L., & Middleton, L.S. (2009). The psychosocial effects of
deployment on military children. Journal of Developmental & Behavioral Pediatrics,
30(4), 271-278.
Flanagan, C., & Stout, M. (2003). Developmental patterns of social trust between early and late
adolescence: Age and school climate effects. Journal of Research on Adolescence, 20 (3),
748-773.
Galovski, T., & Lyons, J.A. (2004). Psychological sequelae of combat violence: A review of the
impact of PTSD on the veteran's family and possible interventions. Aggression and
Violent Behavior, 9(5), 477-501.
Gilreath, T.D., Astor, R.A., Cederbaum, J.A., Atuel, H., & Benbenishty, R. (2012). School
violence and victimization among military connected youth. Manuscript submitted for
publication.
Gorman, G., Eide, M., & Hisle-Gorman, E. (2010) Wartime military deployment and increased
pediatric mental and behavioral complaints. Pediatrics. 126, 1058-1066.
Gottfredson, G., Gottfredson, D., Payne, A. P., & Gottfredson, N. (2005). School climate
predictors of school disorder: Results from a national study of delinquency prevention in
schools. Journal of Research in Crime and Delinquency, 42(4), 412-444.
Greenberg, M.T., Weissberg, R.P., O’Brien, M.U., & Zins, J.E., Fredericks, L., Resnik, H., &
Elias, M.J. (2003). Enhancing school-based prevention and youth development through
coordinated social, emotional and academic learning. American Psychologist. 58(7), 466-
474.

165
Gutman, L., Sameroff, A., & Eccles, J. (2002). The academic achievement of African-American
students during early adolescence: An examination of multiple risk, promotive, and
protective factors. American Journal of Community Psychology, 30(3), 367-399.
Hanish, L. D., & Guerra, N. G. (2000). The roles of ethnicity and school context in predicting
children’s victimization by peers. American Journal of Community Psychology, 28, 201–
223.
Hernandez-Sheets, R. (2003). Competency vs. good intentions: Diversity ideologies and teacher
potential. Qualitative Studies in Education. 16(1), pp. 111-120.
Hernandez-Sheets, R. (2009). What is diversity pedagogy theory? Multicultural Education.
16(3), pp. 11-17.
Hoshmand, L.T., & Hoshmand, A.L. (2007). Support for military families and communities.
Journal of Community Psychology, 35(2), 171-180.
Hoy, W., & Hannum, J. (1997). Middle school climate: An empirical assessment of
organizational health and student achievement. Educational Administration Quarterly,
33(3), 290-311.
Hoy, W., Smith, P., & Sweetland, S. (2002). The development of the organizational climate
index for high schools: Its measure and relationship to faculty trust. The High School
Journal, 86(2), 38-49.
Hoy, W., Tarter, J., & Bliss, J. (1990). Organizational climate, school health, and effectiveness:
A comparative analysis. Educational Administration Quarterly, 26(3), 260-279.
Huebner, A.J., Mancini, J.A., Bowen, G.L., & Orthner, D.K. (2009). Shadowed by war: Building
community capacity to support military families. Family Relations, 58(2), 216-228.

166
Jia, Y., Way, N., Ling, G., Hirokazu, Y., Chen, X., Hughes, D., . . . Lu, Z. (2009). The influence
of student perceptions of school climate on socioemotional and academic adjustment: A
comparison of Chinese and American adolescents. Child Development, 80(5), 1514-1530.
Jordan, B.K., Marmar, C.R., Fairbank, J.A., Schlenger, W.E., Kulka, R.A., Hough, R.L., &
Weiss, D.S. (1992). Problems in families of male Vietnam veterans with posttraumatic
stress disorder. Journal of Counsulting and Clinical Psychology, 60(6), 916-926.
Kagan, D. (1990). How schools alienate students at risk: A model for examining proximal
classroom variables. Educational Psychologist, 25(2), 105-125.
Khoury-Kassabri, M., Benbenishty, R., Astor, R. A., & Zeira, A. (2004). The contributions of
community, family, and school variables to student victimization. American Journal of
Community Psychology, 34(3), 187-204.
Kitmitto, S., Huberman, M., Blankenship, C., Hannan, S., Norris, D., Christenson, B. (2011).
Educational Options and Performance of Military Connected School Districts Research
Study –Final Report. San Mateo, CA: American Institutes for Research.
Kuperminc, G., Leadbeater, B., & Blatt, S. (2001). School social climate and individual
differences in vulnerability to psychopathology among middle school students. Journal of
School Psychology, 39(2), 141-159.
Ladson-Billings, G. (1994). The Dreamkeepers: Successful Teachers of African-American
Children. New York: Columbia University Press.
Loukas, A., Suzuki, R., & Horton, K. (2006). Examining school connectedness as a mediator of
school climate effects. Journal of Research on Adolescence, 16(3), 491-502.

167
Lynch, M. & Cicchetti, D. (1997). Children’s relationships with adults and peers: An
examination of elementary and junior high school students. Journal of School
Psychology. 35(1), pp. 81-99.
MacDermid, S.M., Samper, R., Schwarz, R., Nishida, J., & Nyaronga, D. (2008). Understanding
and promoting resilience in military families. West Lafayette, IN: Purdue University.
Marachi, R., Astor, R. A., & Benbenishty, R. (2006). Effects of student participation and teacher
support on victimization in Israeli schools: An examination of gender, culture, and school
type. Journal of Youth Adolescence, 36(2), 225-240.
McEvoy, A., & Welker, R. (2000). Antisocial behavior, academic failure, and school climate: A
critical review. Journal of Emotional and Behavioral Disorders, 8, 130-140.
McGuire, J., Anderson, C., Toomey, R., & Russell, S. (2010). School climate for transgender
youth: A mixed method investigation of student experiences and school responses.
Journal of Youth and Adolescence, 39, 1175-1188.
Medway, F.J., & Marchant, K.H. (1987). Adjustment and achievement associated with mobility
in military families. Psychology in the Schools, 24(3), 289-294.
Merrell, K.W., Gueldner, B.A., Ross, S.W., & Isava, D.M. (2008). How effective are school
bullying intervention programs? A meta-analysis of intervention research. School
Psychology Quarterly. 23(1), 26-42).
Mmari, K., Roche, K., Sudhinaraset, M., & Blum, R. (2009). When a parent goes off to war:
Exploring the issues faced by adolescents and their families. Youth and Society, 40(4),
455- 475.
Modin, B., & Ostberg, V. (2009). School climate and psychosomatic health: A multilevel
analysis. School effectiveness and school improvement, 20(4), 433-455.

168
Morris, A. S., & Age, T. R. (2009). Adjustment among youth in military families: The protective
roles of effortful control and maternal social support. Journal of Applied Developmental
Psychology.
Opportunities suspended: The devastating consequences of zero tolerance and school discipline.
(2000, June) Civil Rights Project. Retrieved May 14, 2009, from
www.civilrightsproject.ucla.edu/research/.../discipline_gen.php.
Pallas, A. (1988). School climate in American high schools. Teachers' College Record, 89 (4).
541-554.
Perkins, D.F. & Jones, K.R. (2003). Risk behaviors and resiliency within physically abused
adolescents. Child Abuse & Neglect, 28(5), 547-563.
Reed, S.C., Bell, J.F., & Edwards, T.C. (2011). Adolescent well-being in Washington state
military families. American Journal of Public Health, 101, 1676-1682.
Reed, S.C. Bell, J.F., & Edwards, T.C. (2011). Weapon carrying, physical fighting, and gang
membership among adolescents in Washington state military families. Paper presentation
at the American Public Health Association Annual Meeting, Washington, D.C.
Rhodes, J., Camic, P., Milburn, M., & Lowe, S. (2009). Improving middle school climate
through teacher-centered change. Journal of Community Psychology, 37(6), 711-724.
Robinson, J.P. & Espelage, D. (2011). Inequities in educational and psychological outcomes
between LGBTQ and straight students in middle school. Educational Researcher. 40(7),
315-330.
Rodriguez, A.J. & Margolin, G. (2011). Siblings of military servicemembers: A qualitative
exploration of individual and family systems reactions. Professional Psychology, 42(4),
316-323.

169
Rohall, D., Wechsler, M., & Segal, D. (1999). Examining the importance of organizational
supports on family adjustment to army life in a period of increasing separation. Journal
of Political and Military Sociology, 27(1), 49-65.
Rones, M. & Hoagwood, K. (2000). School-based mental health services: A research review.
Clinical Child and Family Psychology Review, 3(4), 223-241.
Rosenheck, R. (1986). Impact of posttraumatic stress disorder of World War II on the next
generation. Journal of Nervous and Mental Disease, 174(6), 319-327.
Solomon, R.P., Palmer, H. (2006) Black boys through the school-prison pipeline Inclusion in
Urban Educational Environments: Addressing Issues of Diversity, Equity, and Social
Justice, Information Age Publishing, 2006.
Townsend, B. (2000). The disproportionate discipline of African-American learners: Reducing
school suspensions and expulsions. Exceptional Children, 66(3), p. 381-391.
U.S. Department of Defense. (2008). Memorandum of understanding between Department of
Defense and Department of Education. Washington, DC: Department of Defense.
Wang, M.-T. (2009). School climate support for behavioral and psychological adjustment:
Testing the mediating effect of social competence. School Psychology Quarterly, 24(4),
240-251.
Wilson, D. (2004). The interface of school climate and school connectedness and relationships
with aggression and victimization. The Journal of School Health, 74(7), 293-299.

170
Yin, M., Kitmitto, S., Shkolnik, J., Hoshen, G., Hannan, S., & Arellanes, M. (2011). Military
Connection and Student Achievement: A Look at the Performance of Military-Connected
Students in Eight Public School Districts. Washington, DC: American Institutes for
Research.
Zullig, K., Koopman, T., Patton, J., & Ubbes, V. (2010). School climate: Historical review,
instrumental development, and school assessment. Journal of Psychoeducational
Assessment, 28 139-152.

Abstract (if available)

Abstract Research on school climate has found that a supportive school climate promotes positive social, emotional, psychological and academic outcomes among students, even student populations experiencing stressors in the family and community context. Studies have shed light on the stressors (i.e. deployment) and the negative mental health outcomes of children in military families. Given the significant presence of military students in over 200 civilian public school districts in the United States, it is surprising that school climate researchers have largely ignored the role that a supportive school climate plays in the social and emotional outcomes of military students. This multiple manuscript dissertation utilizes a population sample of secondary students in eight military-connected school districts to examine the role that a supportive school climate plays in two key social and emotional outcomes known to impact academic achievement—mental health and victimization. The first study provides detailed descriptive analyses of school climate perceptions (belonging, caring relationships, safety, meaningful participation, respect for student’s family, and risky behavior disapproval) among military and non-military students. The second study examines associations between multiple components of school climate and three mental health indicators—well-being, depression, and suicidal ideation—among all students as well as within military and non-military students. The third study assesses associations between school climate and victimization among all students as well as within military and non-military students. Findings from the descriptive analyses indicate that military students—those with a parent or those with a sibling in the military—consistently have more negative school climate perceptions than non-military students. In addition, findings from multivariate analyses reveal that school climate promotes well-being and curbs rates of depression, suicidal ideation and victimization among all students and within military and non-military student populations. This was true even accounting for deployment. This means that a positive school climate can have healing effects that improve mental health and reduce violence and suicide for both military and nonmilitary students. This dissertation concludes with implications for future educational reform, practice and policy.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A mixed method examination of available supports for secondary school students’ college and military aspirations

PDF

Student mental health and wellness in K-12 high-performing school districts in Northern California: best practices for educational leaders

PDF

The assessment of Latino affective and behavioral responses to racial campus climate

PDF

Student mental health and wellness in K-12 high performing school districts in Southern California: best practices for educational leaders

PDF

The effects of peer helping on participants' perceptions of school climate, school connectedness, and school violence

PDF

How successful high school students cope with bullying: a qualitative study

PDF

The effects of campus friendships and perceptions of racial climates on the sense of belonging among Arab and Muslim community college students

PDF

A case study of the academic and behavioral support provided to military-connected children in public schools

PDF

School staff perceptions of school climate: a mixed-methods multistudy examination of staff school climate at the state, regional, and school levels

PDF

Second-generation Korean-American students' mental health experiences in high school

PDF

The relationship of college climate and policy to transgender student belonging

PDF

Improving student-athlete mental health services: addressing the mental health needs of college student-athletes

PDF

Strategy and college participation: a comparative case study of two early college high schools

PDF

School social dynamics as mediators of students personal traits and family factors on the perpetration of school violence in Taiwan

PDF

The relationship between Latinx undergraduate students’ mental health and college graduation rates

PDF

Student mobility in policy and poverty context: two essays from Washington

PDF

High performing schools in high risk environments: a study on leadership, school safety, and student achievement at two urban middle schools in Los Angeles County

PDF

A mixed-methods examination of transgender and gender nonbinary students’schooling experiences: implications for developing affirmative schools

PDF

Psychological distress behavioral patterns and mental health service use among Latinos in the 2012 National Health Interview Survey: a latent class analysis

PDF

The impact of student-faculty interaction on undergraduate international students' academic outcome

Asset Metadata

Creator De Pedro, Kris Tunac (author)

Core Title The role of school climate in the mental health and victimization of students in military-connected schools

School Rossier School of Education

Degree Doctor of Philosophy

Degree Program Education

Publication Date 10/09/2012

Defense Date 07/16/2012

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

Tag bullying,educational reform,Mental Health,military students,OAI-PMH Harvest,school climate

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Astor, Ron Avi (committee chair), Coler, Darnell (committee member), Gilreath, Tamika D. (committee member)

Creator Email krisdepedro@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-103323

Unique identifier UC11289183

Identifier usctheses-c3-103323 (legacy record id)

Legacy Identifier etd-DePedroKri-1238.pdf

Dmrecord 103323

Document Type Dissertation

Rights De Pedro, Kris Tunac

Type texts

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

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA