MEDIA USE AND AFFECTIVE STATE AS MEDIATORS OF
HEALTH-PROMOTING BEHAVIORS IN SCHOOL-AGE CHILDREN
By
Panarat Chenchob

A DISSERTATION
Submitted to
Michigan State University
In partial fulfillment of the requirements
for the degree of
Nursing - Doctor of Philosophy
2013

ABSTRACT

MEDIA USE AND AFFECTIVE STATE AS MEDIATORS OF
HEALTH-PROMOTING BEHAVIORS IN SCHOOL-AGE CHILDREN
By
Panarat Chenchob
Background/Purpose: While children’s health status in the United States differs from state to
state, the health status of children is generally less than optimal. The contributions of healthpromoting behaviors (HPBs) and the factors influencing health behaviors in school-age children
have rarely been reported in literature. The objective of this study is to assess whether exposure
to television, video game, and computer (media use) mediates the relationships between peer
social support and peer hostility and bullying at school (school context), parental emotional
support, closeness and involvement (home context) and dietary intake and physical activity
(HPBs) among school-age children. Framework: The conceptual framework used for this study
is the Pender’s Health Promotion Model (HPM). Research Questions: 1) Is there a direct
relationship between events that occur in school and home contexts for school-age children and
health-promoting behaviors? 2) To what extent does children’s affective state and use of media
mediate the relationship between school and home influences on health-promoting behaviors? 3)
Is there a relationship between media use and affective state in school-age children? 4) Do
mediation effects vary as a function of sex? Methods: The study is a secondary analysis of a deidentified longitudinal data set from the “Coordinated Community Assessment (CCA).” The
sample consisted of 1,370 students. A cross-sequential design included respondents from the
initial cohort recruited from primary school students who were in fourth to sixth grades at
schools located in a Midwestern state. Instrument: The instrument used for measuring selected

2 2

variables was the Coordinated Community Student Survey (C S ) and its subscales for measuring
affective state, school context, home context, media use and health-promoting behaviors.
Procedure: Analysis will involve descriptive statistics, and structural equation modeling (SEM)
using Mplus. Result: Research Question 1: there is no direct effect of school context (β =0.000,
p=0.999) on HPBs two years after school context was assessed. Alternately, home context ((β
=0.179, p=0.000) had a significant direct effect on HPBs. Research Question 2: only children’s
affective state mediated the effects of school context on children’s HBPs (100% mediation).
Children’s media use and affective state mediated the effects of home context on children’s
HPBs (14% mediation), although home context also independently contributed to HPBs.
Research Question 3: affective state was found to be significant positively related with media
use (β <.091, p=000). Research Question 4: For both boys and girls, home context
independently contributed to HPBs while school context did not have a direct effect on HPBs.
Media use was a significant predictor of HPB’s for boys only. Affective state mediated the
relationship between home and school contexts and HPBs for boys and girls. Alternatively,
media use was a mediator in the relationship between home context and HPB for boys only.
Implication: Nurses can provide specific information to enhance support from peers and parents,
increased self-esteem for HPBs within the context of school-age children. Future research
examining health-promoting behaviors in children should identify additional external and
internal factors related to how children make decisions about dietary intake and physical activity.
Policy can be better structured to induce the provision of behavior change interventions, by
fostering linkages with supportive community based resources, and rewarding broad efforts to
improve the population health and quality of life.

Copyright by
PANARAT CHENCHOB
2013

DEDICATION
This dissertation is dedicated to the heroes in my life:
My PhD Advisors, Dr. Hiram E. Fitzgerald and Dr. Barbara A. Given, who have been involved at
various stages in my dissertation production process; my thesis advisor, Dr. Pikul Nantachaipan;
my teachers; Mrs. Prarichat Wangpan; and my family: my beloved father, Mr. Boonsong
Chenchob, my beloved mother, Mrs. Praneet Chenchob, my beloved sister, Mrs. Kanya
Somsangaun and her family and my niece Panida Somsangaun, who provide me with
unconditional love, encouragement, and support as I complete my doctoral degree.
This marathon is also dedicated to my beloved grandfathers and grandmothers, who are the role
models of strength and patience that such a journey involves.

v

ACKNOWLEDGEMENTS

My most cordial thanks to my major advisors, Dr. Barbara A. Given and Dr. Hiram E.
Fitzgerald for providing me with the incredible support in all aspects of research, funding,
guidance, and encouragement from the beginning through the end of the study. This special
gratefulness is also for my graduate committee members: Dr. Jessica V. Barnes-Najor, Dr.
Georgia B. Padonu, Dr. Debra Schutte and Dr. Linda Spence for their great guidance in terms of
professional and technical assistance. My special thanks go to Dr. KyungSook Lee for serving as
my statistical consultant. For assistant and support, I am grateful to Teresa Cherry and Joyce
Pinckney. In addition, I would like to thank the many faculty members and staff at Michigan State
University from whom I have had the opportunity to learn during my doctoral studies.
My great appreciation goes to the C.S. Mott Foundation, the Ruth Mott Foundation, U. S.
Department of Education, the Genesee Intermediate School District and the Michigan State
University Office of University Outreach and Engagement for providing me with the longitudinal
database.
My high appreciation goes to the Thai community for their kind assistance and support
throughout a year of my research processing. My effort to produce this work would not have been
successful without the continued support, kind words, and encouragement I have received from
the many Thai and American scholars. I am also grateful to the good friends I have made at MSU
for their warm support. Without their wonderful companionship, I feel I could not have made it
through the whole process. I wish I could thank all my friends, but they are too many to name
here.

vi

On this occasion, I would like to express my sincere gratitude to the Michigan State
University Graduate School and College of Nursing, which provided me with resources, support,
and funding to complete my dissertation and doctoral program. This funding included the MSU
College of Nursing Dean’s Fellowship, MSU Graduate School Summer Support Fellowship,
George and Margaret Lorimer Parsons Nursing Endowed Scholarship, International Student
Emergency Tuition Award and MSU Graduate School Dissertation Completion Fellowship. Also,
I would like to acknowledge the Royal Thai Government, Ministry of Public Health,
Praboromarajchanok Institute for Health Workforce Development and the Boromrajonani College
of Nursing, Buddhachinaraj in Thailand. They provided me with opportunities and financial
support over the past five years to help me complete my graduate studies. Special thanks to ThaiESF (Emergency Student Fund) for providing me with financial support through the educational
program.
Finally, I would like to express my love and gratitude to my parents and my relatives for
their continued love and encouragement. Their unfailing support and confidence in me provided
me with all the courage I needed to fulfill my dream of advanced education.

vii

TABLE OF CONTENTS

LIST OF TABLES

xi

LIST OF FIGURES

xii

CHAPTER 1
INTRODUCTION
Statement of the Problem
Media and health-promoting behaviors
Affective state
Purpose of the Study
Significance of the Study
Significance to Nursing Science
Significance of School Context
Significance of Home Context
Significant for Thailand
Research Questions
Summary
CHAPTER 2
CONCEPTUAL FRAMEWORK
Pender’s Health Promotion Model
Individual Characteristics and Experiences
Prior related Behavior
Personal Factors
Behavior-Specific Cognitions and Affect
Perceived Benefits of Action
Perceived Barriers to Action
Perceived Self-Efficacy
Activity-related Affect
Interpersonal Influences
Situational Influences
Commitment to a Plan of Action
Immediate Competing Demands and Preferences
Behavioral Outcome
Study Framework
Definitions of the Variables Under Study
Summary

viii

1
2
3
5
6
7
11
12
13
14
14
15

16
16
17
17
17
17
18
18
18
19
19
19
20
20
20
23
29
30

CHAPTER 3
LITERATURE REVIEW
Health-Promoting Behavior and School-Age Children
Mediating Influences on Health-Promoting Behaviors
School Context and Health-Promoting Behaviors
Home Context and Health-Promoting Behaviors
Media Use and Health-Promoting Behaviors
Affective State and Health-Promoting Behaviors
Summary
CHAPTER 4
METHODS
Design of the Original Coordinated Community
Assessment study
Procedure for Data Collection
Design of the Current Study
Sample and Sampling Procedures
Sampling Procedures
Inclusion Criteria
Exclusion Criteria
Instruments
Scale Descriptions
Data Management
Plan for Data Analysis
Protection of Human Subjects
CHAPTER 5
RESULTS

31
31
34
35
38
42
45
53

54
54
55
56
56
57
57
57
58
58
64
64
66

67
67
70
73
74
75
75
79

Descriptive Statistic
Results and Analysis
Research Question 1
Research Question 2
Research Question 3
Research Question 4
Summary

ix

CHAPTER 6
DISCUSSION AND IMPLICATIONS
Discussion of Research Question1
Home Context and HPBs
School Context and HPBs
Discussion of Research Question2
Mediational effects of School and Home
Discussion of Research Question3
Relationship of Affective State and Media Use
Discussion of Research Question4
Sex Differences
Pender’s Conceptual Model of HPM Revisited
Study Limitations
Strength of the Study
Implication for Nursing Practice
Implication for Research
Implication for Policy
Implication for Nursing in Thailand
Summary

80
81
81
82
87
87
90
90
91
89
93
94
95
95
96
97
99
102

APPENDICES
APPENDIX A
APPENDIX B

103
104
105

APPENDIX C

Covariance Metrix
Initial Model for School Context, Home Context,
Affective State, and Media Use, with Standardized
Coefficients
MSU IRB Approval Letter

REFERENCES

106
107

x

LIST OF TABLES

Table 1

The Cross-Sequential Design of the Study by Grade and Wave

56

Table 2

The Percentage af Sexual and Racial of the Sample (N = 1370)

57

Table 3

The C S Subscales Constructs, Items, Rating Scale and alpha
reliability indices

61

Table 4

The Mean, Range, Standards Deviation, Construct, Variables and Scale
Score Characteristics by Sex (N=1,370)

68

Table 5

Sex Difference Comparisons Mean Difference on Variables Used in the
Study

69

Table 6

The Correlation Matrix for All Study Variables

70

Table 7

The Comparison of the Model Fit Indices between the Initial Model and
the Alternative Model

72

Table 8

Mediation, Effect Size, Direct Effect and Indirect Effects of School
Context on HPBs

74

Table 9

Mediation, Effect Size, Direct Effect ,Direct Effect and Indirect Effects
of Home Context on HPBs

74

Table 10

The Comparison of the Model Fit Indices between the Initial Model and
the Alternative Model for Sex Differences

76

Table 11

Mediation, Effect Size, Direct Effect Direct Effect and Indirect Effects
of School Context on Health-Promoting Behaviors for Sex Differences

76

Table 12

Mediation, Effect Size, Direct Effect ,Direct Effect and Indirect Effects
of Home Context on Health-Promoting Behaviors for Sex Differences

76

Table 13

Defined School Context

83

2 2

Table 14 Covariance Matrix

104

xi

LIST OF FIGURES

Figure 1

Pender’s Health Promotion Model

21

Figure 2

Model of Health-Promoting Behaviors in School-Age Children

27

Figure 3

Path Model 1 for School Context, Home Context, Affective State, and
Media Use, with Standardized Coefficients

73

Figure 4

Path Model 2 in Boys for School Context, Home Contex, Affective State,
and Media Use, with Standardized Coefficients

77

Figure 5

Path Model 3 in Girls for School Context, Home Context, Affective State,
and Media Use, with Standardized Coefficients

78

Figure 6

Initial Model for School Context, Home Context, Affective State, and
Media Use with Standardized Coefficients

105

xii

CHAPTER 1
INTRODUCTION
During the past several decades, approaches to health promotion have increasingly strived
to align with ecological models designed to identify the set of variables that contribute to the
etiology of poor health choices (Crawford, Story, Wang, Ritchie, & Sabry, 2001). Ultimately, the
discovery of critical direct and indirect predictors of health-promoting behaviors (HPBs) will lead
to more effective preventive intervention programs (Callaghan, 2006). To date, evidence exists to
support four key domains where variables have been linked to dietary intake and physical activity
(HPBs): self-esteem, depression (affective state), peer social support and peer hostility and
bullying at school (school context), parental emotional support, closeness and involvement (home
context) and television, video games and computer (media use). The declining status of children’s
health in the United States has occurred over a generation. Poor eating habits together with
physical inactivity have been implicated as key risk factors for poor health during adolescence and
into adulthood.
Studies of the factors influencing HPBs in school-age children are underrepresented in the
literature, particularly studies involving longitudinal samples. Nursing interventions are needed to
promote HPBs among school-age children. Therefore, the purpose of the current study is to assess
predictors of HPBs in a longitudinal sample of children grade fourth to sixth (age 9-12 years) with
specific attention to the extent to which exposure to media and/or affective state mediate the
relationship between peer social support and peer hostility and bullying at school (school context)
and parental emotional support, closeness and involvement (home context) on two indicators of
dietary intake and physical activity (health-promoting behaviors).

1

Statement of the Problem
Even though children’s health status in the United States varies from state to state, the
health status of children is generally less than optimal (Fitzgerald, Lester, & Zuckerman, 2006;
Villarreul & Luster, 2006). Eight percent of all eighth graders reported heavy drinking within the
past 30 days, eight percent of all 12- to 17-year-olds have had a major depressive episode, 9
percent have asthma and incalculable others have behavioral disorders and are in less than good
health (Fitzgerald, Puttler, Refior, & Zucker, 2007). Longitudinal studies have revealed that risk
taking shows strong continuity from the pre-K and elementary school age into adolescence, and the
risk taking in this time span is intensified among children who are raised in high-risk families
(Fitzgerald et al., 2007). This is particularly true when children are exposed to neighborhood
violence and family aggression (Buu, et al., 2009). This means that multiple factors affect the riskresilience continuum, including those that influence health-promoting behaviors. In current studies
of parental emotional support, closeness and involvement, children reported their perception of
their parents’ response to their behaviors. The factors that influence health behaviors are
multidimensional and systemically related. Some variables have a direct impact on health
behavior; others are mediated or moderated by co-occurring factors.
Physical inactivity is increasing among school-age children, especially among those aged
11 to 14. The Youth Risk Behavior Surveillance System demonstrated that the percentage of
students who attend physical education class daily decreased from 49% in 2010 to 39% in 2012
(Centers for Disease Control, 2012), despite the continued promotion of guidelines recommending
that school-age children should participate in 60 minutes or more of moderate-to-vigorous physical
activity daily (USDHSS, 2012). The majority of youth do not meet such recommendations (Eaton

2

et al., 2012). Dieting and weight concerns, particularly among young girls (Gardin &
Hammarstrom, 2002), have also become public health problems because of their link to eating
disorders such as bulimia and anorexia.
Social support has been defined in numerous ways, generally referring to any behavior that
assists an individual in achieving a desired goal or outcome. In Pender’s Health Promotion Model
(HPM) (2002), social support is viewed as an interpersonal influence and is considered a protective
mechanism for health-promoting and -maintaining behaviors. Conceptually, social support can
create a growth-promoting environment, decrease stressful life events and provide feedback or
confirmation of actions. When individuals believe that they have adequate social support, the
resulting goals of health promotion and health maintenance are more likely to be achieved. Social
support is considered the most established determinant in the literature reinforcing physical
activity. Pender (1996) suggested that a number of possible sources of support for physical activity
should be examined, such as parents, older children or siblings, and friends. A few studies have
examined the impact of family and peer support on this influence (Higgins, Gaul, Gibbons, & Gyn,
2003; Sallis et al., 1992). As children age, they spend increasing amounts of time outside of the
home, and peers become an important source of social support for physical activity (Pender, 1996).
Media and health-promoting behaviors. For the past 20 years the impact of media on
children’s behavior has drawn increasing attention. Initially, researchers and health professionals
focused on the impact of television on children’s aggressive behavior (Eron, 1982) and weight
gain. Dietzand and Gortmaker (1985) concluded that there were positive and perhaps causal
connections. Increasingly, such concerns have shifted beyond children’s exposure to television to
include time spent playing video games or texting on mobile devices. Studies have reported

3

positive effects of computer-use on children’s academic achievement (Jackson et al., 2006; Kulik,
1994; Wenglinsky, 1998), expansion of social networks (Jackson, Fitzgerald, VonEye, Zhao, &
Whitt, 2010), or no effects (Jackson, 2008). Roschelle, Pea, Hoadley, Gordon, and Means (2000)
note that based on the extent to which Information Technology (IT) use interferes with the four
fundamentals of learning (active engagement, participation in groups, frequent interaction and
feedback and connections to real-world contexts), use of computer-based instruction will likely be
less effective. Indeed, investigators have shown that time on the Internet can be socially and
psychologically isolating, resulting in feelings of depression and loneliness (Amichai-Hamburger
& Ben-Artzi, 2003). With respect to video games, while they may enhance visual spatial skills
(Subrahmanyam, Krant, Greenfield, & Gross, 2000), other evidence suggests that levels of
aggression are higher in children who are heavy users of video games, particularly violent games,
and that teacher ratings of their behavior in school are more negative when compared with children
who use games less frequently (Anderson, Gentile, & Buckley, 2007).
Jackson, VonEye, Fitzgerald, Witt, & Zhao (2011) studied the effects of Internet use,
video game playing and cell phone use on children’s weight gain and self-esteem. Although ethnoracial group membership, age and income levels predicted grades in school and body mass index
(BMI) or body weight, none of the IT indicators did. Conversely, IT use did predict children’s
social self-esteem. Children who played video games more had lower social self-esteem than
children who played them less. On the other hand, children who used cell phones more had higher
social self-esteem than those who used them less. Similar findings held for general self-esteem.
Children are exposed to an extraordinary amount of information about energy dense foods
on television (Moore, 2008). Low levels of exercise, high consumption of energy dense foods and

4

low affect expression all contribute to children’s weight gain. With rare exceptions (e.g.,
Nintendo’s Wii), screen time is a time of high caloric consumption and low physical activity
(Jackson et al., 2010), contributing to weight gain and obesity (AOA, 2005). Finkelstein, Rhum,
and Kosa (2005) note that of 10 or more food commercials per hour, most were for fast foods and
soft drinks, adding to television’s impact on children’s weight gain. Adams (2006) specifically
targets poor eating habits, soft drinks, lack of exercise and high amounts of time watching
television or using a computer as major contributors to child obesity. Considering that 17% of 2-19
year olds and 10.4% of preschool-age children in the United States are obese (American Obesity
Association, 2005; Ogden et al., 2006), examining the role of media use on children’s weight gain
is important.
Affective state. Physical activity may offer protective, buffering effects on mental health
for youth prior to the onset of emotional problems (Dyck, 2000) by raising levels of self-esteem,
body image and self-concept (Higgins et al., 2003). Indeed, regular exercise participation has been
found to be associated with a decrease in state and trait anxiety, depression and stress (Dunn &
Madhukat, 2001). In Pender’s HPM, affective state reflects both self-esteem and depression, each
of which can influence HPBs independently or in combination. Chenchob, Barnes, Fitzgerald, Lee,
& Pandonu (2013) found that affective state mediated the effect of social support and
neighborhood safety on children’s HPBs. Many psychosocial models of health-promoting behavior
suggest that children develop higher levels of self-esteem when they set and attain goals (Crawford
et al., 2001), when they feel that others accept them and when they have positive loving
relationships with caregivers (Bee, 2000). Because studies of Information Technology use suggest
that high use of television and computers is linked to indictors of negative affective state, the

5

current study will examine whether media use and affective state also mediates the relationship
between school context, home context and HPBs in the sample of fourth to sixth graders.
More research is required regarding potential predictors for future health behavior, in
relation to both individual and family oriented factors, including factors related to the school
environment, such as teacher support (Gadin & Hammarstrom, 2002; Patnode et al., 2010).
Although some studies have reported linkages between variables such as self-perception and
health-promoting behaviors, none has assessed the inter-relationships among individual, school
and family context and HPBs (Crocker et al., 2003; Peason, Ball, & Crawford, 2011).
As stated in the introduction, the health status of children in the United States has been
below the average criteria, especially dietary intake and exercise. Crocker et al. (2003) suggested
that physical self-perceptions are stronger predictors of change in physical activity and dietary
restrain.
In summary, increasing HPBs must continue to have high priority. More research is
required regarding potential predictors for future health behavior, in relation to both individual and
family oriented factors, including factors related to the school environment. The current study
seeks to identify predictors of HPBs not only at the individual level but also in home context and
school context. The question of interest is whether affective state and media use mediate school or
home context effects on health-promoting behaviors.
Purpose of the Study
The purpose of the current research is to assess predictors of HPBs in a longitudinal sample
of children grade fourth to sixth (age 9-12 years) with specific attention to the extent to which
exposure to media and/or affective state mediates the relationship between school context, home

6

context and two indicators of health-promoting behaviors (dietary intake and physical activity).
The proposed study will focus on continuing to identify direct and indirect effects of factors
affecting HPBs, conceptualized within a systems framework and informed by Pender’s HPM.
Significance of the Study
The prevention of risky health behaviors among school-age children has high priority in
public health. It is important to find early predictors of poor health behavior because established
health behaviors in school-age children can be difficult to change later in life (Gadin &
Hammarstrom, 2002). Starting life in a suboptimal environment caused by poor diet, physical
inactivity or other factors in the built, social, or natural environment does not support growth.
These factors may also increase the risk of future chronic degenerative disease development
(Dwyer, 2006). The factors that influence health behaviors are multidimensional and systemically
related. Some variables have a direct impact on health behavior; others are mediated or moderated
by co-occurring factors. Whether their impact is direct or indirect, such factors cooperatively
support the processes that influence individuals to make decisions and participate in healthpromoting behaviors (Pender, Murdaugh, & Parsons, 2006; Wu & Pender, 2000). Early
identification of these interrelationships and an understanding of the dynamics that facilitate
health-specific behaviors can provide insight into health enhancing practices in early life,
especially in school-age children. Most studies of health behaviors among school-age children
have focused on individual factors rather than on the home and school contexts that are associated
with HPBs (Trost et al., 2003). It is reasonable to posit that health promotion programs will be
more effective if they are founded on evidence-based predictors that include the important personenvironment relationships that influence health behavior decisions (Gadin & Hammarstrom, 2002)

7

as they are defined within particular developmental periods. Diet and exercise intervention can
have a beneficial effect on body weight, BMI, serum cholesterol and fitness in children.
For school-age children, the main environment is not only their home but also their school.
Both home and school are important arenas for HPBs among school-age children (Eccles, 1999;
Gadin & Hammarstrom, 2002; Trost et al., 2003). However, in research on school-age children,
school as a health-promoting setting has lagged far behind school as an environment for academic
achievement and cognitive functioning. More research regarding the predictors of health behavior
is required, particularly using approaches that allow for inclusion of variables from multiple
contexts. Such an approach could ascertain how multiple risk factors coalesce in relation to both
individual and family oriented factors and factors related to the school environment.
The study of factors that are reliable predictors of HPBs among school-age children is
important for several reasons. First, it will provide a foundation for the development of
intervention strategies for nurses, school nurses and other health care professionals to assist those
children who are distressed or incapable of managing their health. Second, success of intervention
efforts is dependent upon evidence that clearly implicates etiologic influences on desired outcomes
linked to developmental age (Gadin & Hammars, 2002). For example, the interventions planned
for early childhood would differ markedly from those planned for middle school because the
causal variables influencing dietary intake and physical activity are markedly different. Because
the proposed research focuses on the transition from elementary years to middle school years,
findings may lead to new insights into the causal risk patterns associated with the development of
poor health behavior. Third, understanding the relationship among influencing factors and HPBs
may be beneficial in improving academic achievement and cognitive functioning in school-age

8

children. Identifying sources of the predictors of HPBs will enhance the effectiveness of specific
health promotion efforts because of the empirical link between predictors and desired outcomes
(Gadin & Hammarstrom, 2002). Fourth, there is a need for empirical validation of models of
HPBs. Improved theory in this area will lead to a better understanding of and fruitful insights into
HPBs. Fifth, it is hoped that the findings in this study can be used as a basis for successful
planning and provision of appropriate health care services for school-age children. Furthermore,
these results can inform health care providers, teachers, parents and researchers interested in
enhancing factors for promoting health behaviors for future interventions. Finally, rapid changes in
food intake and lifestyle pattern demonstrate a significant impact on the shifting pattern of disease
burden on the population (Dwyer, 2006). These changes should be monitored carefully and must
be reversed through appropriate behavior modification and the promotion of appropriate eating
practices and physical activities. Chronic diseases such as obesity, heart disease, osteoporosis,
diabetes and cancer continue to be major health concerns that interfere with the quality of life in all
life spans.
The factors that influence health behaviors are multidimensional and systemically related
(Sallis, Owen, & Fisher, 2008; Spear & Kulbok, 2001). As Trost et al. (2003) stated, little is
known about the factors that promote parental support. Whether their impact is direct or indirect,
such factors cooperatively support the processes that influence individuals to make decisions and
participate in health-promoting behaviors. Identification of the interrelationships and an
understanding of the dynamics that facilitate health-specific behaviors can provide insight into
HPBs.

9

Few studies have combined media use, affective state, school context, home context to
determine their unique and combined influences on health-promoting behaviors (Lowry, Wechsler,
Galuska, Fulton, & Kann, 2002; Patnode et al., 2010). Few studies have been published that
specifically concentrate on identifying predictors of HPBs among the relationships between
individual, family, school and community contexts (Patnode et al., 2010). The research regarding
school-age children and school as a health-promoting setting and an environment for development
has seldom been targeted for HPBs (Gadin & Hammarstrom, 2002). More research is required to
identify potential predictors of health behavior in relation to both individual and family oriented
factors, including factors related to school environment, such as teacher support, etc. Some studies
have reported on other variables such as self-perception and HPBs, but not specifically in regards
to individual, school and family context (Gadin & Hammarstrom, 2002).
As stated in the introduction, health status in children in the United States has been below
the average criteria, especially dietary intake and exercise. Crocker et al. (2003) suggested that
physical self-perceptions are stronger predictors of change in physical activity and dietary restrain.
This suggests that they have investigated health-related behaviors associated with dieting and
physical activity. The decline in physical activity and increase in BMI are consecutive concerns.
These findings strongly suggest that dietary intake and exercise are linked to body self-perceptions
and dietary restrain. In summary, what is known to date is that researchers have not yet
investigated the variables mentioned above.
The findings of this study will increase understanding about the predictors among media
use, affective state, school context, home context, and health-promoting behaviors. Moreover, the
findings from this study should provide essential knowledge concerning individual, school and

10

community aspects affecting school-age children that can predict health-promoting behaviors.
Future research is needed to identify potential social factors that might contribute to
changes in HPBs, especially given their role in predicting increases in healthy dietary intake and
exercise. It has been reported that peers, parents and the media are potential socializing agents that
could impact these observed changes in physical self-perceptions, and should therefore be
investigated (Crocker et al., 2003).
In summary, in health promotion, the importance of early intervention should be addressed.
This study will identify predictors not only on the individual level but also in home and school
contexts. A more structural perspective on predictor analysis will predict future children’s HPBs.
Further study is required to investigate whether the above findings correlate with these variables,
especially the potential individual, social, physical environment and mental factors as potential
determinants of change in a range of school-age children (Pearson, Rall, & Crawford, 2010). The
question remains whether or not media use mediates the relationship between school context,
home context and health-promoting behaviors (dietary intake and physical activity), and if
affective state mediates the relationship between media use and health-promoting behaviors. The
proposed study will focus on filling the gap of knowledge to explain the causes and effects among
these variables.
Significance to Nursing Science
The results of this study suggest that efforts designed to instill HPBs in school-age children
must be based on public health models grounded in systemic approaches rather than one-solution
causal models. Whether individual children will be responsive to attempts to instill HPBs will
depend on the extent to which prevention efforts are capable of improving affective state. Nursing

11

intervention approaches, therefore, should be multifaceted, focusing on enhancing self-esteem and
reducing depression, while simultaneously working to build support systems that effectively
enhance HPBs in children. These results can inform health care providers and researchers
interested in enhancing factors for promoting health behaviors for future interventions. Nursing
interventions should emphasize the importance of HPBs among school-age children. Strengthening
the delivery of health knowledge pertinent to self-esteem and social support could enrich HPBs in
school-age children. Nurses and other health care professionals can encourage and assist schoolage children to practice healthy behaviors, emphasizing the impact they have upon their own
health. The findings related to the importance of social support suggest nurses and other health
care providers can positively influence health behavior.
Significance of School Context
Findings will increase understanding of the relationship between factors that may impact
children’s HPBs. Teachers and staff who need to understand effective strategies for enhancing
HPBs can learn from these findings, allowing them to provide support and encouragement to
school-age children. The results of the study may suggest that counseling and health promotion
programs can be tailored to suit school-age children.
Such understandings provide a basis for reconstructing models of health promotion, such as
Pender’s HPM, to incorporate more dynamic frameworks for generating multiple pathways to
HPBs. New community practice guidelines offer additional evidence-based recommendations for a
wide array of school-, worksite-, and community-based programs and public policies to improve
dietary intake and physical activity levels for children and reduce harmful behaviors such as smoke
exposure, drug use and gun violence (Glanz, Rimer, & Viswanath, 2008). The current study may

12

also have implications for educational workshops and counseling for individuals who are at risk
for developing health problems, or those simply desiring more information about health and
wellness. Interventions are needed to promote HPBs among school-age children. Researchers
should consider using the significant relationships identified in this study to guide their future
intervention development and promote HPBs for school-age children.
New strategies for promoting health efforts have shifted away from individual-level approaches
toward systemic approaches that address socio-cultural, political, economic and physical
environmental factors (Yancey, Olsen, Guyton, Bakst, & Westman, 2004). This demonstrates the
importance of leadership within schools and communities to set priorities and direct local
resources toward disease risk reduction.
Significance of Home Context
Researchers required to routinely assessing participation in health promotion lifestyle
behaviors through familial and cultural values and beliefs can facilitate this process. The
identification of health resources such as health education classes and sources of accessible, lowcost health screening services would be useful for increasing health. Moreover, researchers should
determine which health-promoting behaviors to concentrate on for successful health promote on
intervention.
Parents could help shape a positive behavioral history for the future by providing
interventions that focus on the benefits and remove barriers to engaging in health-promoting
behaviors (Smith & Bashore, 2006). Smith and Bashore (2006) suggest that providing care from a
health promotion perspective may be useful in helping parents’ survivors normalize and increase
control over their lives. Parents should direct their conversation to adolescents or young adults,

13

clarify any misinterpretations and encourage them to actively participate in decisions that will
ultimately affect their health. These factors cooperatively support the processes that influence
individuals to make decisions and participate in health-promoting behaviors. Identification of the
interrelationships and an understanding of the dynamics facilitate health-specific behaviors.
Moreover, parents can provide specific information to enhance social support for health-promoting
behaviors within the context of the family. The importance of social support suggests that family
members, peers and friends can influence positive health behavior. This finding can inform parents
and children’s family interested in enhancing factors for promoting health behaviors in their
children.
Significant for Thailand
The findings from the current study may have potential importance for research on HPBs in
Thailand, where obesity and other non-communicable diseases such as asthma, diabetes, and
hypertension are significantly increasing in school-age children. Similarities and differences
between Thailand and Western culturals s have not been fully explored with respect to factors that
may influence health promoting behaviors among children and families. Nevertheless, the results
of the current study in combination with those reported by (Chenchob et al., 2013), provide a
starting point for additional research designed to understand the complex relationships that will
lead to culturally appropriate evidence-based programs designed to enhance the health and
wellness of children and families in Thailand. .
Research Questions
Research Question 1: Is there a direct relationship between events that occur in school and home
contexts for school-age children and health-promoting behaviors?

14

Research Question 2: To what extent does children’s affective state and use of media mediate the
relationship between school and home influences on health-promoting behaviors?
Research Question 3: Is there a relationship between media use and affective state in school-age
children?
Research Question 4: Do mediation effects vary as a function of sex?
Summary
The rationale for this chapter was to explore the statement of the problem, purpose,
background, significance, research questions and specific aims of the current study. The proposed
study is to examine and identify the predictive variables among media use, affective state, school
and home contexts as predictors of HPBs among school-age children.

15

CHAPTER 2
CONCEPTUAL FRAMEWORK
Among various individual health approaches to enhancing health-promoting behaviors,
Pender’s health-promotion model (HPM) has been widely studied, particularly within nursing
practice. According to Pender, Murdaugh, & Parsons (2002), individual characteristics and
experiences (prior related behavior and personal characteristics) have direct effects on healthpromoting behavior, but also have mediated effects through behavior-specific cognitions and
affects. Prior related behavior refers to an individual’s presenting state characteristics, and their
cumulative life-course experiences. Pender's HPM asserts that health-promoting behaviors are
mediated by perceived benefits of action, perceived barriers to action, perceived self-efficacy and
activity-related affect. The impact of personal factors is mediated by a set of interpersonal
influences (family, peers, providers, norms, etc.) and situational influences (options, demand
characteristics) (Pender et al., 2002).
Pender’s Health Promotion Model
The revised version of Pender’s HPM (2002) is depicted in Figure 1. The model addresses
three major constructs: 1) individual characteristics and experiences, 2) behavior-specific
cognitions and affect and 3) behavioral outcome. Pender’s comprehensive model of health
promotion is one of the explanatory nursing models that predict health behavior. A revised health
promotion model (Pender et al., 2002), based upon social learning theory, was modified to identify
the factors associated with exercise behavior. According to the revised health promotion model,
exercise as a health-promoting behavior is influenced by personal and behavior-specific cognitions
and affect (Pender et al., 2002). Behavior-specific cognitions and affect are the categories of major

16

motivational significance and provide a useful perspective in explanation of the phenomena. These
factors are critical for intervention, as they are subject to modification through nursing actions.
Individual Characteristics and Experiences
According to Pender, Murdaugh, & Parsons, et al. (2011), people have unique and
individual characteristics. These unique characteristics will influence the way in which people
engage in various behaviors. The individual characteristics or aspects of past experience allow the
researcher to include variables that may be important for a particular health behavior rather than
include all possible variables for all possible populations (Pender et al., 2002). All variables can be
described as follows.
Prior related behavior. Pender et al. (2011) stated that prior experiences influence current
behavior. Prior behavior is proposed as having both direct and indirect effects on engagement in
health-promoting behaviors. Prior behavior is proposed to also have an indirect influence on
health-promoting behavior because of self-efficacy, current actions, benefits and barriers. Activityrelated affect may also influence current actions (Pender et al., 2011).
Personal factors. In the revised (Pender et al., 2002), personal factors include biological,
psychological and sociocultural factors. Biological factors include variables such as age, body
mass index, pubertal status, menopausal status, aerobic capacity, strength, agility or balance.
Psychological factors may include self-esteem, self-motivation and perceived health status (Pender
et al., 2002; 2011). Examples of sociocultural factors include race, ethnicity, acculturation,
education and socioeconomic status.
Behavior-Specific Cognitions and Affect
Behavior-specific cognitions and affect are considered to be of major motivational

17

significance (Pender et al., 2011) Thus, perceived benefits of action, perceived barriers to action,
perceived self-efficacy, activity-related affect, interpersonal influences and situational influences
serve a vital role in interventions, because they are modifiable through nursing actions. Six
components in behavior-specific cognitions and affect have been identified and are discussed as
follows.
Perceived benefits of action. These perceived benefits directly motivate behavior.
Indirectly, perceived benefits influence behavior through commitment to the plan of action. One’s
plan to engage in a particular behavior is based on benefits or positive outcomes that will occur
(Pender et al., 2006). Individuals tend to take action if that action is likely to bring about positive
and desired outcomes (Pender et al., 2002; 2011).
Perceived barriers to action. Barriers consist of perception of Unavailability,
inconvenience, cost, difficulty and time constraints are frequently viewed as barriers to action,
which affect health-promoting behaviors (Pender et al., 2011). Perceived barriers to action affect
health-promoting behavior directly by hindering action and indirectly by decreasing one’s
commitment to a plan of action. Barriers can be real or imagined (Pender et al., 2002). The more a
person perceives barriers in practicing a particular behavior, the more likely that person will not be
committed to engage in the behavior (Pender et al., 2002).
Perceived self-efficacy. Perceived self-efficacy is a judgment of one’s abilities to perform
specific behaviors in certain situations (Bandura, 1997; Pender et al., 2011). This perception is not
necessarily based on the person’s actual skills but their perceptions of their abilities (Pender et al.,
2011). Self-efficacy is influenced by prior experiences or behavior. A person with high selfefficacy is more likely to adopt health-promoting behaviors than a person with low self-efficacy.

18

According to Pender’s HPM, “perceived self-efficacy motivates health-promoting behavior
directly by efficacy expectation and indirectly by affecting perceived barriers and determining the
level of commitment or persistence in pursuing a plan of action” (Pender et al., 2011, p. 47).
Activity-related affect. Three components of activity-related affect are affects related to the
activity itself, affects related to the self and affects related to the environment or context in which
the behavior takes place (Pender et al., 2011). Activity-related affect, which is one’s subjective
feeling state, may change as a result of stimulus properties of the behavior itself. This change may
occur at three different times: before, during and after a behavior (Pender et al., 2002). These
emotional reactions may be either positive or negative. This resulting feeling is likely to influence
whether a person will repeat a behavior or maintain the behavior long term (Pender et al., 2011).
Interpersonal influences. Interpersonal influences are cognitions or perceptions
concerning the behaviors, beliefs or attitudes of others, including expectations of significant others
(social norms), instrumental and emotional support (social support), and vicarious learning
experiences (modeling) (Pender et al., 2002; 2006; 2011). These sources of interpersonal
influences may affect health-promoting behaviors directly or indirectly through social pressure or
encouragement to commit to a plan of action (Pender et al., 2011). Social support networks
including family, peers and health care providers are primary sources of interpersonal influence.
Situational influences. Situational influences include perceptions of available options,
demand characteristics and pleasing features of the environment in which a given behavior is
proposed to take place (Pender et al., 2011). This variable may impede or facilitate health
behaviors depending on how the person reacts to the environment or situation (Pender et al., 2011).
Individuals are more likely to adopt in health-promoting behaviors in an environment in which

19

they feel compatible, reassured, safe, and related (Pender et al., 2011). Situational factors have
both direct and indirect influences on health behavior.
Commitment to a Plan of Action
Commitment to a plan of action initiates a behavioral event (Pender et al., 2011). “This
commitment will propel the individual into and through the behavior unless a competing demand
that the individual cannot avoid, or a competing preference that the individual does not resist,
intervenes” (Pender et al., 2011, p. 49). Commitment to a plan of action in the HPM implies the
following underlying cognitive processes: 1) commitment to carry out a specific action at a given
time and place and with specified persons or alone, irrespective of competing preferences and 2)
identification of definitive strategies for eliciting, carrying out and reinforcing the behavior
(Pender et al., 2011).
Immediate competing demands and preferences. Immediate competing demands and
preferences represent “alternative behaviors that intrude into consciousness as possible courses of
action immediately prior to the intended occurrence of behavior” (Pender et al., 2011, p. 49).
Competing demands are alternative or unanticipated behaviors, over which a person has little or no
control, but he or she must respond to, such as work and family care responsibilities (Pender et al.,
2011).
Behavioral outcome. Health-promoting behavior is the outcome of the HPM (Pender et al.,
2002; 2006; 2011). The behavior is directed towards gaining positive health outcomes, including
improved health (Pender et al., 2011).

20

Behavior-Specific
Cognitions
and Affect

Behavioral
Outcome

Perceived
benefits of
action

Prior
related
behavior

Immediate
competing demands
(low control) and
preferences (high
control)

Perceived
barriers to
action

Perceived
self-efficacy

Commitment to a
plan of action

Activityrelated affect

Personal
factors;
biological
psychological
sociocultural

Healthpromoting
behaviors

Interpersonal
influences
(family, peers,
providers);
norms,
support,
models
Situational
influences;
options
demand
characteristics
aesthetics

Figure 1. Pender’s Health Promotion. Model Source: Pender, N. J., Murdaugh, C. L., & Parsons,
th
M. A. (2002). Health promotion in nursing practice (4 ed.). Prentice-Hall: New Jersey.

21

A modified framework based on Pender’s HPM is proposed for this study. A schematic is
shown in Figure 2. The modified model is comprised of three constructs: 1) individual
characteristics and experiences, 2) behavior-specific cognitions and affect (depression, self-esteem,
support from parents, peers at school) and 3) behavioral outcome (health-promoting behavior:
dietary intake and physical activity) (Pender et al., 2006). Pender (1996) postulated that personal
factors such as gender and race affect HPBs directly, as well as indirectly, through interpersonal
influences. Because of the large number of potential personal factors, Pender (1996) suggested that
factors be limited to the few that are theoretically relevant to the explanation or prediction of a
given target behavior. According to the HPM, inherited and acquired characteristics influence
beliefs, affect and enactment of health-promoting behaviors (Pender et al., 2002; Chen, 1995;
Garcia et al., 1995; Gillis, 1993; Martinelli, 1999; Stutts, 1997). Although self-efficacy is one of
the major motivational factors in Pender’s HPM, the data set analyzed for the current study does
not contain a measure of that important variable. Instead, the current study focuses on the extent to
which internalized aspects of affective state, as measured by self-esteem and depressive symptoms,
mediate the relationship between contextual factors and children’s dietary intact and physical
activity.
In Pender’s health-promotion model, social support is viewed as peer social support and
parental emotional support and is considered a protective mechanism for health-promoting and
health-maintaining behaviors. Conceptually, social support can create a growth-promoting
environment, decrease stressful life events and provide feedback or confirmation of actions. When
individuals believe that they have adequate social support, the resulting goals of health promotion
and health maintenance are more likely to be achieved. Interpersonal influences, including social

22

support, are also posited to have a direct influence on the individual’s plan of action and healthpromoting behaviors. Health-promoting efforts have the potential long-term benefits of enhancing
the quality of life from childhood through the adult years, increasing longevity, reducing health
care costs and increasing national productivity through reduced illness and absenteeism. School
may be an important context for health promotion practices to the extent that it provides students
with feelings of safety and positive affect.
Health promotion is defined as specific behaviors that individuals perform on a daily basis
to improve or preserve their health and well-being (Pender, 1996). Pender (1996) proposed that a
person’s behavioral outcome does not rely only on the person’s perceptions as a primary source of
health motivation, but also on individual characteristics and experiences, as well as on cognitive
and affective factors. Furthermore, the model addresses three major constructs: 1) individual
characteristics and experiences, 2) behavior-specific cognitions and affect and 3) behavioral
outcome.
Study Framework
This modified model proposes a process consisting of concepts that influence healthpromoting behavior. The factors that influence health behaviors are multidimensional. All factors
are interrelated and therefore produce results that exert both direct and indirect influences on
health-promoting behaviors. These factors cooperatively support the processes that influence
individuals to make decisions and participate in health-promoting behaviors. Their proposed
relationships are described in detail below and hypothesized functional relationships between each
concept are schematically illustrated in Figure 1.
Pender (1996) developed the Health Promotion Model (HPM) in 1982 and revised it in

23

1987, then again in 1996. The HPM is an attempt to depict the multifactorial dimensions of people
interacting with the environment as they pursue health. The model incorporates concepts from
expectancy value theory and social learning theory.
In the early 1980s, the initial version of the HPM first appeared in nursing literature.
However, since the late 1980s, when public attention focused more readily on health promotion,
the demand for information to explain the factors that motivate people to seek their health potential
has risen (Pender, 1982). The model is organized similarly to the Health Belief Model and the
modified Health Belief Model (Becker, 1974), which were categorized into individual perceptions,
modifying factors and variables affecting likelihood of action. The HPM is also similar to healthprotecting behavior since it consists of a decision-making phase and an action phase.
The health promotion model was revised by removing some concepts including selfawareness and self-esteem because the problems associated with measuring such concepts are
formidable. For example, self-awareness is a general and rather ambiguous personal characteristic
that is not well operationalized. Furthermore, there is only limited empirical evidence that selfesteem affects level of participation in health-promoting behavior (Pender, 1987). According to
this model (Figure 1), health promotion entails activities directed toward developing resources that
maintain or enhance a person’s well-being. The HPM encompasses two phases: a decision making
phase and an activity phase. The model emphasizes seven cognitive/perceptual factors that
compose motivational mechanisms for acquiring and maintaining health-promoting behaviors,
including importance of health, perceived control of health, perceived self-efficacy, definition of
health, perceived health status, perceived benefits of health-promoting behaviors and perceived
barriers to health-promoting behaviors, and five modifying factors, including demographic

24

characteristics, biologic characteristics, interpersonal influences, situational factors and behavioral
factor that indirectly influence patterns of health behavior. In the action phase, barriers and cues to
action trigger activity in health-promoting behavior.
The current research tests a model derived from Pender et al. (2002) (see Figure 2). The
importance of health, perceived control of health and definition of health were deleted from the
revised model due to lack of sufficient empirical evidence of explanatory or predictive power,
particularly in studies of specific health behaviors.
Pender defined health promotion as specific behaviors that individuals perform on a daily
basis to improve or preserve their health and well-being and actualize human health potential
(Pender et al., 2002; 2011). She proposed that a person’s behavioral outcome does not rely only on
the person’s perceptions as a primary source of health motivation, but also on individual
characteristics and experiences, as well as on cognitive and affective factors (Pender, 1996).
Furthermore, Pender (1987) suggested that it is necessary to give attention to the environmental,
cultural and social conditions that affect health and HPBs of individuals and groups, because the
factors that influence healthful behaviors are interrelated and multidimensional.
Health-promoting behavior is the continuing behavioral, cognitive and emotional efforts of
school-age children to sustain and improve health and well-being. Health-promoting behavior is
mediating the effects of the antecedent variables such as interpersonal influences. The model
proposes that a combination of interpersonal influences, situational influences and affective state
for health behaviors directly influence the frequency of health-promoting behaviors. Therefore, it
is important to identify factors that predict HPBs, and may directly or indirectly influence HPBs
among school-age children, in order to help policy makers and public health professionals

25

develop and implement policies and programs that enhance HPBs.
The main reasons for modifying Pender’s HPM include the following: 1) the relationship
between situational/interpersonal influences on self-efficacy is not clearly understood and 2) the
theoretical model fails to account for a relationship between health-promoting behavior and health
outcomes. In the current healthcare arena, with emphasis on evidence-based practice, the
relationship between behavior and outcomes should be made precise (Srof & Velsor-Fried, 2002).
The model is based on a synthesis of research findings from studies of health promotion and
wellness behavior and serves three important functions: 1) the model introduces order among
concepts that may explain the occurrence of health-promoting behavior, 2) the model provides for
the generation of hypotheses to be tested empirically and 3) the model integrates disconnected
research findings into a coherent pattern (Pender, 1987). To develop more effective HPBs
interventions for school-age children, it is important for interventions to be based on theoretical
models that adequately explain and predict HPBs. A greater understanding of the mediating
mechanisms underlying HPBs may help researchers to develop and implement more effective
interventions.
When Pender’s model was developed, children’s use of digital devices was of little interest
in respect to HPBs. A great deal of attention was directed toward the effects of television violence
on aggression in children (Bandura, 1977) and the promise of developments of literacy and social
behavior from children’s television shows. But the effects of children’s use of media on healthpromoting behaviors have surfaced within the past decade with the surge in mobile devices,
games, available computers, iPods and the like and the possibility that such use was related to
reduction in physical activity and exposure to advertising. Therefore, the current study will

26

examine the direct and indirect influences on health-promoting behaviors via school context,
home context, media use and affective state.
Wave 1
School Context
-Peer hostility and bullying
-Peer social support

RQ 1

Wave 2
Media Use
-Television
-Video games
-Computer
RQ 2

RQ 3
Wave 2
-Computer Use
Affective State
-Self-esteem
-Depression

Wave 3
HPBs
-Dietary intake
-Physical activity

RQ 1

Wave 1
Home Context
-Parent emotional support
and closeness
-Parent involvement

Figure 2. Model of health-promoting behaviors in school-age children.
As shown in the model, school context and home context were measured in wave one of the
study. Affective state and media use were measure in wave two. Health-promoting behaviors were
assessed from students’ report of their dietary intake and physical activity in wave three of the
study.
27

In the current study involving a school-age longitudinal sample, the relative impacts of
school-based and home-based factors that influence HPBs are explored. The school-based domain
focuses on peer social support (Chenchob, Barnes, Fitzgerald, Lee, & Pandonu, 2011; 2012) but
adds school safety; whereas the home context domain focuses exclusively on parent variables
(parent emotional support, closeness and involvement). For example, the result of a longitudinal
sample of fourth- to sixth- grade students (Chenchob et al., 2013) provided evidence that affective
state (self-esteem, depression) mediated the relationship between social support (teachers, parents)
and neighborhood safety on two indicators of health-promoting behaviors (children’s dietary
intake and exercise). In the current study involving the same school-age longitudinal sample, the
researcher will be unpacking the social support domain to more directly assess the relative impacts
of school-based and home-based influences on HPBs. The current study will examine the extent to
which children’s involvement with media (television viewing, video games and computer use)
mediates school-based and home-based predictors of HPBs that were identified by Chenchob et al.
(2013) and many others in the HPM literature (e.g., Callaghan, 2006; Garcia, Pender, Antonakos,
& Ronis, 1998; Garcia et al., 1995; Robbins, Pis, Pender, & Kazanis, 2004). However, these
results have not been linked to HPBs, especially dietary intake and physical activity. Studying
children’s use of media in relation to HPBs has received strong support from extensive studies that
have implicated exposure to television advertising and similar Web-based messages as detrimental
to health in children (Moore, 2008).
In summary, the specific aims in the proposed research were influenced by Pender et al.
(2002) multivariate and social learning approach. Moreover, researchers represent a continuation
of this investigation of the factors that influence health-promoting behaviors during the transition

28

from elementary to middle school. A modified version of Pender’s HPM (2002) guides the models
to be tested in this study, a continuation of our investigation of the factors that influence healthpromoting behaviors during the transition from elementary to middle school. Pender’s HPM helps
healthcare professionals understand health behaviors from a broader perspective by examining the
factors that affect the choices people make about their health behavior rather than relying solely on
factors within the individual. Indeed, the HPM posits that health behavior is influenced by multiple
factors that are both internal and external to the individual. The HPM also identifies relationships
among those factors, and provokes awareness of those relationships to enhance our understanding
of the complex influences on HPBs in school-age children. Finally, Pender’s HPM has been
effectively used to guide several cross-cultural studies of health-promoting behavior in school-age
children (Robbins, Pis, Pender, & Kazanis, 2004; Wu & Pender, 2002), providing evidence of the
usefulness of this model. It has the potential to influence nursing practice, education and
administration. In nursing practice, nurses can use the health promotion model to provide a
coherent and organized framework for intervening with clients to increase health-promoting
behaviors.
Definitions of the Variables under Study
For the purpose of this study, the following definitions are used:
School-age children are fourth to sixth grade boys and girls from a Midwestern state.
School context was defined by children’s reports of the extent to which they perceived their
peers as supportive and by children’s exposure to hostile peer behaviors, such as bullying.
Home context is defined as children’s perceptions of the emotional support they receive
from their parents as well as the degree to which parents are involved in child-rearing activities.

29

Media use is defined as how often school-age children use television, video
games and computers.
Affective state is defined as self-esteem (a stable sense of personal worth or worthiness as
measured by self-report) and depression (feeling sad, blue, unhappy, miserable, or down in the
dumps).
Health-promoting behaviors are assessed as dietary intake and physical activity. These
behaviors serve to help school-age children sustain and improve health and well-being.
Summary
The purpose of this study is to examine the extent to which media use and affective state
mediates the relationships between selected variables from the school and home context and HPBs
in children. Moreover, based on the prior work (Chenchob et al, 2011; 2012; 2013), the researcher
will assess the extent to which children’s affective state mediates the relationship between media
use and HPBs.
Research regarding HPBs has typically been more focused on specific health-related
supports and influences. For example, when examining interpersonal influences, researchers
generally examine how interpersonal relationships are supportive of or discouraging for an
individual’s engagement in healthy behavior. One’s friends might encourage eating either healthy
or unhealthy food. The current study is examining how the more broadly defined school and home
contexts are related to health behaviors. The findings of this study may assist nurses and other
health care professionals in learning more about factors contributing to HPBs in school-age
children. The current study will enhance knowledge to guide effective nursing interventions aimed
at helping school-age children improve their HPBs.

30

CHAPTER 3
LITERATURE REVIEW
The purpose of this chapter is to synthesize the literature relevant to the ideas, theories, and
research related to health-promoting behaviors in school-age children and factors influencing
health-promoting behaviors, in relation to contextual and personal predictors. The chapter presents
relevant literature research findings that used Pender’s HPM as a theoretical perspective to explain
health behaviors and factors that may be predictors of an individual’s decision to participate in
health-promoting practices.
Health-Promoting Behavior and School-Age Children
The increasing prevalence of being overweight among children and adolescents highlights
the need for physical activity and effective dietary strategies for weight management (Davies,
Fitzgerald, & Mousouli, 2008; Fitzgerald & Mousouli, 2008). The prevalence of overweight and
obesity among children and adolescents has increased dramatically during the past 20 years, and is
more pronounced in children from private schools and urban communities than in those from
public schools or rural areas (Kosulwat, 2002). Contextual or social factors such as access to
recreational facilities or safety of neighborhood environments influence how individuals engage in
leisure physical activity.
Physical activity is an important component in overall approaches to the prevention of
chronic illness in children. Considering current evidence on the etiologic and epidemiologic
characteristics of illness and disease in the Western world, the most important problems in the
United States are non-communicable chronic diseases, such as heart disease, obesity, osteoporosis,
diabetes, and cancer. Many studies find that regular physical activity is a contributing health care

31

practice that promotes healthy behaviors. Physical activity benefits chronically ill patients by
improving their symptoms or their health conditions, and their quality of life. It also is beneficial to
psychological well-being, decreasing stress, anxiety, and anger and improving children’s academic
performance (Filed, Diego, & Sanders, 2001; Sallis, Prochaaka, & Taylor, 2000; Strong et al.,
2005)
Research has shown that there is a high dropout rate from programs designed to increase
and maintain physical activity. Structured exercise programs have reported dropout rates that range
from 9% to 87%. Excluding active participation in sports, girls tend to be less active in physical
exercise than boys (Berg et. al., 2009; Biddle, Gorely, & Stensel, 2004; Charoneying, 2005).
Therefore, investigators have examined programs designed to enhance efforts to involve more girls
in active exercise. Programs have demonstrated that when girls are involved in such programs their
physical activity and fitness is associated with perceived self-efficacy (Teerarungsikul et al., 2009).
However, the positive effects of the health promotion program were not sustained after 12 weeks
in the program (Teerarungsikul et al., 2009).
The challenge for physical activity planning; therefore, is to translate the beliefs and
intentions of adolescents into action, and to provide the support and encouragement needed to
create physically-active communities. However, most studies of health behaviors in children have
focused primarily on individual factors with less attention given to contextual factors that may
alter individual predictors of health promotion (Gardin & Hammarstrom, 2002).
It is important to find early predictors for two main reasons. First, identified risk factors can
be changed to create a more supportive environment for school-age children and individuals at risk
so that they can have special support and help. Second, home and school environment can produce

32

both risk and resilience factors that influence individual personal characteristics.
Pender et al. (2011) defined health promoting behaviors as those “activities motivated by the desire
to increase well-being and actualize human potential” (p. 5). Rather than stabilization and
avoidance of disease, health promotion behaviors develop the tension necessary to increase
sustained practices that improve health (Pender et al., 2006).
Health-promoting behaviors, especially dietary intake and physical activity, when instilled
during childhood and adolescence, tend to continue into adulthood. Physical activity, nutrition,
social support, life appreciation, health responsibility, and stress management are all important
components of models used in health promotion for adolescents (11 – 20 years). Physical activity
and nutrition are recognized as health behaviors associated with home and school-context
(Goodman, Dolan, Morrison, & Daniels, 2005; Goodman, Daniels, & Dolan, 2007). Although
adopting healthy behavior is recognized as one of the best approaches to avoid obesity, strategies
to promote HPBs in relation to obesity reduction have not been adequately addressed in the
literature (Ogilvie et al., 2007).
Emphasis on health promotion has been an integral part of nursing practices for so long that
it is foundational to the discipline of nursing. Evidence has shown that HPBs have been successful
in helping people improve their health practices and health conditions (Stuifbergen & Rogers,
1997; Stuifbergen & Timmerman, 2003). Pender (1996) posited that by engaging in healthpromoting lifestyles, individuals could maintain and enhance their well-being and prevent the early
onset of disabling health conditions. Adoption of healthy lifestyles could slow physical decline
from a chronic health problem and even improve general physical and mental well-being in
children. Additionally, there are indications in the health care literature that when identified early,

33

risky factors could be postponed through lifestyle changes and those health-promoting behaviors in
particular could benefit children. Although there is awareness of the need for health promotion in
school-age children, it is essential to better understand the factors that may contribute to decisions
of school-aged children to actually practice healthy lifestyles.
Some researchers have examined how diet, physical activity, and sedentary behaviors relate
to overweight status in young children. Studies in adolescents have shown a significant association
between physical activity and weight status (Patrick et al., 2004; Sallis, Prochaska, & Taylor,
2000). However, the evidence linking eating and physical activity pattern with overweight in youth
is contradictory and inconclusive. Most previous studies have examined nutrition or physical
activity factor, but few have examined both. Although most research on children’s dietary and
physical activity behaviors involve self report, more experimental studies including self-report data
can continue to help inform etiologic concepts about the system influences that affect children’s
practices related to health promotion (Patrick et al., 2004).
Mediating Influences on Health-Promoting Behaviors
One of the under-studied questions flowing from Pender’s HPM concerns the extent to
which predictors have direct or indirect influences on children’s health-promoting behavior.
Chenchob et al. (2011; 2012; 2013) derived a mediational model based on Pender et al. (2006)
approach in order to examine the extent to which affective state mediated the relationship between
indicators of social support and neighborhood context on diet and exercise. Neighborhood safety
and social support were both significant predictors of affective state but only when mediated
through the child’s affective state. The results of Chenchob et al. (2011; 2012; 2013) found that
children’s affective state mediated the relationship between indicators of social support and

34

perceived neighborhood safety and children’s HPBs. Efforts designed to instill HPB in school-age
children must be based on public health models grounded in systemic approaches rather than onesolution causal models.
School Context and Health-Promoting Behaviors
School context is defined as children’s reports of their parents’ emotional support and
involvement in the subjects’ everyday lives (sometimes referred to as parental monitoring). This
variable facilitates health behaviors depending on how reacts to the environment or situation
(Pender et al., 2002).
Situational influences included perceptions of available options, demand characteristics,
and pleasing features of the environment in which a given behavior is proposed to take place
(Pender et al., 2002). Situational factors have both direct and indirect influences on health
behavior. Individuals may want to behave in the ways that promote health, but the environmental
constraints may prevent healthy actions (Pender et al., 2011). Social support looms as one source
of resilience, particularly when such support occurs in the early years of social and neurobiological
development.
Social support has been defined in numerous ways, and generally refers to any
behavior that assists an individual in achieving a desired goal or outcome. Pender et al.(1994)
suggested that a number of possible sources of support for physical activity should be
examined, such as parents, older children or siblings, and friends. A few studies have examined the
impact of family and friend support for this influence (Sallis et al.,1992; Sallis, & Keating, 1994).
In particular, peers are potentially important source of social support for adolescent physical
activity (Pender et al., 1994). Conceptually, social support could promote environment, decrease

35

stressful life events, and provide feedback or confirmation of actions (Pender et al., 2002). When
individuals believe that they have adequate social support, the resulting goals of health promotion
and health maintenance are more likely to be achieved. Social support is considered the most
established determinant reinforcing physical activity and the most established determinant
reinforcing physical activity (Tinsley, 2007; Duncan, Duncan & Strycker, 2005). Both parents and
peers play a crucial role in supporting youths’ physical activity (Higgins, Gaul, Gibbons, & Gyn,
2003; Duncan, Duncan & Strycker, 2005; Voorhees et al., 2005).
Perceived importance of school environment is a key environmental component in
explaining physical activity behavior among youth (Fein, Plotnikoff, Wild, & Spence, 2004).
Fein et al. (2004) examined physical environments to explain and promote physical activity, with
home, neighborhood, and school as significant domains. The results showed that perceived
importance of the school environment was the only environmental variable significantly associated
with physical activity (β = .14; p<.05). This finding needs to provide and support school physical
environments related to physical activity. Gadin and Hammarstrom (2002) conducted a three-year
prospective study in grade six and grade nine. The objective of this study was to find predictors
among young children for later risky heath behaviors with special focus on school-related factors.
The results indicated that school-related factors could predict future health behaviors, especially in
relation to low physical activity among girls. Low physical activity among girls could best be
predicted by school-related factors such as classmate problems, teasing others, rowdiness in the
class.
Empirical evidence supports the observation that health-promoting behaviors in schoolage children is affected by valuing peer friendship and other relationships, especially peer social

36

support, and peer hostility and bulling at school. Studies have demonstrated that physical activity
environmental characteristics in the school setting have the ability to influence students’ activity
level (Baranowski, 2004; Sallis,Conway, Prochaaka, Mckenzie, Marshall, & Brown, 2001;
Sallis,Conway, Elder, Prochaaka, Brown, Zive, Marshall, & Alcaraz, 2003; Verstraete, Cardon, De
Clercq, & De Bourdeaudhuij, 2006). These findings suggest that the increased likelihood of being
active in school with rich facilitated environments could be attributed to several factors which are
related to children health-promoting behaviors (Salmon et al., 2007).
To our knowledge, no studies have combined affective state (self-esteem, depression),
social support (peer social support, parents support (parents emotional support and closeness,
parents involvement), situational influence (school safety) to determine their unique and
combined contributions to understanding health-promoting behaviors. Few researchers specifically
concentrate on predictors in the relationship among individual, family, school, and community
context (Higgins, Gaul, Gibbons, & Gyn, 2003). However, the research regarding school-age
children, school as a health-promoting setting and as an environment for children’s development
and behavior has seldom been focused on HPBs. Some studies have reported regarding other
variables such as self-perception and health-promoting behaviors (dietary intake and exercise), but
not specified in individual, school and family context (Sallis et al., 2001). The relationship
between school context and health-promoting behaviors from this study cannot be generalized to
the population at large. However, the majority of evidence favors a relationship between school
context and health-promoting behaviors. Less peer social support may have a significant direct
negative effect on school-age children. Hence, it is essential to explore how school context plays a
role in health-promoting behaviors among school-age children.

37

Home Context and Health-Promoting Behaviors
Home context is defined as children’s perceptions of the emotional support they receive
from their parents as well as the degree to which parents are involved in child rearing activities.
Parents managed their children’s health behaviors (Srof, & Velsor-Friedrich, 2006). Parents
also shape the lives of their children from birth through adulthood (Bayers, Bates, Pettit, & Dodge,
2009). Among children, the influence of friends and peers takes on greater importance, but the
research clearly demonstrates the continued significance of parents in shaping behaviors and
choices of teens as they faced the challenges of growing up and health-promoting behaviors
(Borkowsky, Ramey, & Bristol-Power, 2002). There is also clear evidence from longitudinal
studies that parenting practices are intergenerational (Kerr, Capaldi, Pears, & Owens, 2009).
Interpersonal influences are cognitions or perceptions concerning the behaviors beliefs or
attitudes of others including expectations of significant others (social norms), instrumental and
emotional support (social support), and vicarious learning experiences (modeling) (Pender et al.,
2002). Interpersonal influences may affect health-promoting behaviors directly or indirectly
through social pressure or encouragement to commit to a plan of action (Pender et al., 2002; 2011).
Social support networks including family are primary sources of interpersonal influence.
According to the HPM, these influences could be sources of encouragement for compliance with
health-promoting behaviors and/or the cessation of negative health behaviors.
Parents and children relationships, shared family activities, good parenting skills and
positive parental role modeling all have well-documented effects on children’s health behaviors
and development. Parenting has been shown to play a substantial role in socialization, and more
specifically, in the physical and emotional development of youth (Bayers et al., 2009). In some

38

areas, parents could make choices to make positive changes for their children (Bayers et al., 2009).
Close relationships between parents and their children including healthy open communication, and
perceived parental support are especially important during childhood, as they experience many
physical and emotional transitions from childhood to adolescence (Aufseeser et al., 2006).
Children are also less likely to report symptoms of depression and more likely to report high levels
of perceived well-being when children have parent support and parent involvement (Mounts,
2001). However, it is difficult to interpret the direction or the causal ordering of the association
between support from parent, children’s affective state and health-promoting behaviors because of
lack of longitudinal studies.
Parental monitoring involves knowing the children’s situation after school, as well as
knowing children’s friends, peers and activities (Bayers et al., 2009). Parental monitoring
combined with parental support has demonstrated a positive relationship to children’s higher selfesteem, and academic accomplishment (Mounts, 2001). In addition, parental monitoring has been
associated with fewer internalizing behaviors such as depression and withdrawal, and externalizing
behavior problems such as disturbing others and fighting (Barber et al.,1994; Brody, et al., 2002)
as well as a lower chance of drinking (Stephenson, Quick, & Atkinson, 2005), and engaging in
other risky behaviors. Low parental monitoring, low positive parental involvement consistently,
and lack of adult supervision, predict greater levels of conduct problems and delinquency in youth
(Bayers et al., 2009). The majority of evidence favors parental monitoring as a predictor of healthpromoting behaviors.
Although it is significant, parental involvement is not the only context that can affect
influence on children’s physical and emotional development. Parents may be considerably

39

affected by specific characteristics of the neighborhoods in which their families reside (Bayers et
al., 2009). Parental intake and children’s fat, fruit and vegetable intake have a consistent
relationship to their children’s health-promoting behaviors. A positive association was found for
the relationship between these behaviors and the availability and accessibility of fruit and
vegetables. All other associations studied between dietary behaviors and potential environmental
factors were inconsistent, appeared non-existent or were not replicated (Host et al., 2006). Parents
should support healthy food environments for their children (Jenkins, & Hormer, 2005,; Patrict, &
Nicklas, 2005; Ritchie et al., 2005; Story, Neumark-Sztainer, & French, 2004). Research needs to
warrant the direction of home context on health-promoting behaviors.
Family meals serve as an important time for children to communicate with and spend time
with their parents, which has been associated with fewer depressive symptoms and suicide
attempts, and better academic performance (Eisenberg, Neumark-Sztainer, & Bearinger, 2004).
Children who eat meals regularly with their parents are also more likely to eat fruits, vegetables,
and dairy foods and less likely to skip breakfast (Bayers et al., 2009). More frequent family meals,
a more structured family meal environment and a positive atmosphere at family meals are
associated with a lower likelihood of disordered eating (Neumark-Sztainer, Wall, Story, &
Fulkerson, 2004).
Parents’ health-related behaviors could influence children’s health-promoting behaviors in
several ways including providing positive (or negative) role models and by contributing to healthy
or unhealthy physical and social environments. Parental behavior could also shape children’s risky
behaviors by increasing simple access to cigarettes or alcohol in the home, or, on the positive side,
increasing access to healthy foods (Aufseeser et al., 2006). Moore and Harre (2007) examined the

40

eating behaviors, physical exercise and television viewing of secondary school students and
investigated their relationship with parental monitoring and family cohesion. The results showed
that parental monitoring and family cohesion were significant positively related to participants’
report of eating breakfast and eating healthy food, and significant negatively related to reports of
buying one’s own food and eating unhealthy food. Exercise rates were moderately correlated with
eating healthy food and weakly related to parental monitoring and family cohesion. Television
viewing was related to eating unhealthy food.
The family environment could be a strong source of support for developing adolescents,
providing close relationships, strong parenting skills, good communication, and modeling
positive behaviors models (Borkowsky, Ramey, & Bristol-Power, 2002). For example, research
shows teens who have positive relationships with their parents are less likely to engage in various
risk behaviors, including smoking, fighting, and drinking (Aufseeser et al., 2006). Both parents and
peers play an essential role in supporting youths’ physical activity experience (Kremarik, 2000;
Sallis et al., 2000; Okun, Karoly, & Lutz, 2002).
In summation, home context and and health-promoting behaviors have been investigated in
the literature regarding school-age children and supports the linkage among the current study
variables. For school-age children, parents play an important role in managing their children’s
health behaviors. Research on the influence of parents on health-promoting behaviors has been
consistent (Aufseeser et al., 2006; Kremarik, 2000; Sallis et al., 2000; Okun et al., 2002). Positive
parental involvement may also decrease risk for deviance by promoting competence and the
internalization of parental values.

41

In conclusion, the studies discussed in this literature review reflect the variables included in
the proposed study. In addition, the research supports the contention that these variables have an
impact on the practice of health-promoting behaviors among school-age children. The findings are
consistent; however, regarding the relationships between home context (parent emotional support
and closeness, and parent involvement) and health-promoting behaviors.
Media Use and Health-Promoting Behaviors
Media use is defined by how often school-age children use television, video games, and
computers. During the past 40 years, the impact of media on children’s behavior has garnered
increased attention. Initially, researchers and health professionals focused on the impact of
television on children’s aggressive behavior (Eron, 1982) and weight gain. Dietz and Gortmaker
(1985) concluded that there were positive and perhaps causal connections. Increasingly, such
concerns have shifted beyond children’s exposure to television to include time spent playing video
games or texting on mobile devices. Studies have reported positive effects of computer-use on
children’s academic achievement (Jackson et al., 2006; Kulik, 1994; Wenglinsky, 1998), and
expansion of social networks (Jackson et al., 2010), or no effects (Jackson, 2008). Roschelle, Pea,
Hoadley, Gordon, and Means (2000) noted that due to the extent to which IT use interferes with
the four fundamentals of learning (active engagement, participation in groups, frequent interaction
and feedback, connections to real-world contexts), use of computer-based instruction will likely be
less effective. Computer use can be socially and psychologically isolating, resulting in feelings of
depression and loneliness (Amichai-Hamburger, & Ben-Artzi, 2003). With respect to video games,
while video games may enhance visual spatial skills (Subrahmanyam et al., 2000), other evidence
suggests that levels of aggression are higher in children who are the heavy users of video games,

42

particularly violent games (Anderson, Gentile, & Buckley, 2007). Teacher ratings of children’s
behavior in school are more negative when compared with children who play games less
frequently (Anderson et al., 2007).
Jackson et al. (2011) studied the effects of internet, video game, and cell phone usage on
children’s weight gain and self-esteem. Although ethno-racial group membership and age and
income levels predicted grades in school and body mass index (BMI) or body weight, none of the
IT indicators did. Conversely, information technology use did predict children’s social self-esteem
(Jackson et al., 2011). Children who played video games more had lower social self-esteem than
children who played them less. On the other hand, children who used cell phones more had higher
social self-esteem than those who used them less.
Children are exposed to an extraordinary amount of information about energy dense foods
on television (Moore, 2008). Low levels of exercise, high consumption of energy-dense foods, and
low affect entirely contribute to children’s weight gain. With rare exception (e.g., Nintendo’s
Weii), screen time is the time of high caloric consumption and low physical activity (Jackson et al.,
2010), contributing to weight gain and obesity (AOA, 2005). Finkelstein, Rhum and Kosa (2005)
note that there are 10 or more food commercials per hour, most of which are for fast foods and soft
drinks, adding to televisions impact on children’s weight gain. Adams (2006) specifically targets
poor eating habits, soft drinks, lack of exercise, and high amounts of time spent watching
television or on a computer as major contributors to child obesity. In the United States, 17% of 219 year olds and 10.4% of pre-school age children are obese (American Obesity Association,
2005; Ogden et al., 2006). Therefore, examining the role of media use on children’s weight gain is
a vital issue.

43

Media use should include in assessments of children, and reinforce efforts of parents to
monitor and restrain TV viewing. Schools and community organizations could offer youth
comprehensive media education programs and promote local TV turnoff week project (Lowry,
Wechsler, Galuska, Fulton, & Kann, 2002). Decreasing the time spent on watching TV creates
chances for activity and many shift dietary intake away from high-calorie, high-fat foods incidence
advertised on television (Dietz & Gortmaker, 2001).
Assessments of children should include media use, and should reinforce efforts of parents
to monitor and restrain TV viewing. Schools and community organizations could offer youth
comprehensive media education programs and promote local TV turnoff week projects (Lowry et
al., 2002). Decreasing time spent watching TV creates increased chances for activity and may shift
dietary intake away from high-calorie, high-fat foods advertised on television (Dietz, 2001).
Previous research has shown a complex interaction between time spent watching
television, physical activity level and body fat level (Proctor et al., 2003; Biddle et al., 2004).
Even if television does not reduce exercise time, it may still be problematic by increasing and
promoting snack advertisements. To help develop effective interventions to reduce television
viewing and increase physical activity, a consistent correlation between TV viewing and negative
health behaviors needs to be identified (Gorely, Simon, Marshall, Stuart, & Biddle, 2004). TV
viewing appears to increase between-meal snacking and was inconsistently related to children
asking for and choosing TV-advertised foods, and actual dietary fat intake (Gorely et al., 2004).
Also, TV viewing in children is positively associated with parental viewing habits. Some studies
reported that no relationship was found between TV viewing and physical activity (Gorely et al.,
2004; Sallis et al., 2000). In summary, children’s media use may influence their health-promoting

44

behaviors. It is a factor encouraging expose to risky behaviors and healthy behaviors. The media
influences children’s perspectives depending on the media quality.
In conclusion, the studies discussed in this literature review reflect the variables included in
the proposed study. In addition, the research supports the contention that these variables have an
impact on the practice of HPBs among school-age children. The findings are not consistent;
however, regarding the relationships between children’s media use and HPBs, the studies conflict
or the findings are weak, therefore, further research is warranted in these areas.
Affective State and Health-Promoting Behaviors
Affective state is defined as self-esteem (a stable sense of personal worth or worthiness as
measured by self-report), and depression (feeling sad, blue, unhappy, miserable, or down in the
dumps).
Multiple factors including psychological attributes, HPBs and biological factors influence
on an individual’s ability to make required lifestyle changes. An absence of depression could be
influenced by physical activity because physical activity may offer protective, buffering effects on
mental health for youth prior to the onset of emotional problems (Dyck, 2000) by raising levels of
self-esteem, body image and self-concept (Higgins et al., 2003). Regular exercise participation has
been found to be associated with a decrease in state and trait anxiety, depression, and stress (Dunn
& Madhukat, 2001). Chenchob et al., (2011; 2012; 2013) found that affective state mediated the
effect both the relationship of social support and neighborhood safety on children’s healthpromoting behavior. Social support and neighborhood safety are indirectly affected on HPB.
Health-promoting behavior models suggest that children develop higher levels of self-esteem when
they set and attain goals (Crawford, et al., 2001), when they feel that others accept them, and when

45

they have positive loving relationships with caregivers (Bee, 2000). Children’s affective state
characteristics predict their high risk behavior from early childhood into the elementary years.
Because studies of information technology usage suggest that high use of television and computers
is linked to indictors of loss of affective state (depression, loneliness) (Jackson et al., 2006), the
current study examines whether media use and affective state also mediates the relationship
between school, home context and HPBs.
Sex differences in self–esteem have been noted with higher self–esteem commonly
identified in males and decreasing self–esteem in females as they enter adolescence (Hendricks et
al., 2001). Across culture and gender, abilities were valued as a promoting self –esteem (Guinn,
Vincent, Semper, & Jorgensen, 2000; Hendrickset et al., 2000). Studies revealed that self –esteem
might be influenced by a variety of factors including environment, culture and social support
(Mahat et al., 2002; Schmitz, 2006; Swaim & Wayman, 2004). Regardless of the antecedents to
self–esteem, the direct relationship to HPBs remains.
Positive Self–esteem promoting positive behavior is generally viewed in adolescent studies
(11 – 20 years) as global self–esteem rather than self–esteem garnered from past accomplishments
(Rosenberg et al., 1995). Self–esteem directly kindles HPBs decreasing the likelihood of health–
risk behavior (Hendricks et al., 2001; Riesch, Anderson, & Krueger, 2006; Torres & Fernández,
1995). Self–esteem was not significantly associated with nutritional health behavior (Torres &
Fernández, 1995). Nelson and Gordon–Larsen (2006) investigated relationships among self–
esteem and HPBs in almost 12,000 adolescents (11% Hispanic) using data from a national
longitudinal study with mean age of 15. Cluster analysis of this data revealed patterns of HPBs

46

associated with higher self–esteem. This finding raises a question about the influence of selfesteem on health-promoting behaviors two years later in school-age children.
In the current study, researcher hypothesize that school-age children who have greater selfesteem are more likely to practice the HPBs. Self-esteem represents a motivational force that
influences perceptions and coping behavior (Mann, Hosman, Schalma, & Vries, 2004). Therefore,
school-age children with high self-esteem are likely to select appropriate health activities, and have
greater commitment to pursue anticipated outcomes.
Mann et al. (2004) found that a high self-esteem score was a significant predictor of
engaging in HPBs in school-age children, and schools need to aim at helping children develop a
healthy sense of self-esteem as part of the development of their intra-personal intelligence. In those
studies, findings showed that self-esteem had a significant, direct effect on HPBs. Many
psychosocial models of health behaviors also suggest that children develop higher levels of selfesteem when they set and attain goals to succeed (Crawford et al., 2001). Children, who feel that
others accept them, and they are unconditionally loved and respected, will report a higher selfesteem (Bee, 2000). Also, self-esteem is critical to school-age children in making positive health
decisions (Davis et al., 1999). In the current study, it is possible that school-age children having
high level of self-esteem may have tendency to engage in the activities needed to enhance their
health and well-being, which may explain why school-age children who had greater self-esteem
are more likely to practice HPBs. Megel et al. (1994) examined relationships among self-esteem,
health promotion, nutrition, and weight in 57 female college freshmen (mean age 18.5 years). The
results indicated that positive correlations were identified among health promotion behaviors, self-

47

esteem, and nutrition. Self–esteem, self–efficacy and hope, have been demonstrated to be strong
predictors of HPBs in adolescents (Yarcheski et al., 2004).
Garcia et al. (1995) examined gender and developmental differences in exercise-related
beliefs and exercise behaviors of 286 racially diverse youth and found that female reported less
prior and current exercise, lower self-esteem, poorer health status, and lower exercise self-schema
than men. More active parents had more active children, indicating the importance of positive
parental role models. Ease of access to exercise facilities was related to the level of physical
activity in adults and also may affect exercise patterns of youth.
Research has shown significant relationships between depression and health behaviors
among adolescents, particularly regarding the use of tobacco, alcohol, and other drugs use among
adolescents. Less is known about the relationships between depression or depressive symptoms
and other health behaviors such as eating, nutrition, and physical activity among adolescents in
nonclinical settings. Depressive symptoms can be associated with lower rates of “healthpromoting” behaviors and may inhibit full-health potential across multiple domains of health
(Fulkerson, Sherwood, Perry, & Neumark-Sztainer, 2003).
In population-based male and female adolescent samples, depression or negative mood has
also been linked with health-compromising attitudes and behaviors such as weight dissatisfaction
(Tomori & Rus-Makovec, 2000), negative body image (Stice, Presnell, & Spangler, 2002),
disordered eating (Neumark-Sztainer et al., 2000). Samples limited to female adolescents have also
shown higher rates of binging (Ackard, Neumark-Sztain, Story, & Perry, 2003) purging, and
dieting (Pesa, 1999) among girls who reported depressed mood. One study illustrated a significant
negative relationship between depressed mood and health-promoting eating behaviors such as

48

eating breakfast and lunch (Kaplan, Landa, Weinhold, & Shenker, 1984). Skipping meals,
particularly breakfast, may affect concentration and learning. Fulkerson et al. (2004) described
associations between depression symptoms and nutritional intake or physical activity among
adolescents in a nonclinical trial. The results showed that depressive symptoms were positively
associated with health-compromising attitudes such as perceived barriers to healthy eating and
weight concerns, such as unhealthy weight-control behaviors. However, depressive symptoms
were negatively associated with health-promoting behaviors such as eating breakfast, lunch, and
dinner.
An absence of depression could influence physical activity because physical activity may
offer protective, buffering effects on mental health for youth prior to the onset of emotional
problems (Dyck, 2000) by raising levels of self-esteem, body image and self-concept (Higgins,
et al., 2003). Indeed, regular exercise participation has been found to be associated with a decrease
in state and trail anxiety, depression, and stress (Dunn, & Madhukat, 2001). Higgins et al. (2003)
reported that female youth were found to be less physically active and more concerned about being
overweight, and more depressed. However, females were more likely to report greater social
support than males and to be more socially involved.
A review of the literature reveals that self-esteem is consistently identified as a significant
predictor of health-promoting behaviors within a variety of groups, including children, and
adolescents (Bee, 2000; Davis et al., 1999; Mann, Hosman, Schalma, & Vries, 2004; Megel et al,
2004; Swaim & Wayman, 2004; Nelson & Gordon–Larsen, 2006; Yarcheski, 2004). Only one
study was demonstrated that self-esteem was not significantly associated with nutrition health
behavior (Torres & Fernández, 1995). Research on the influence of gender on health-promoting

49

behavior found that males reported higher self-esteem and higher exercise level than females
(Garcia et al., 1995). The gender difference in school-age children specifically requires further
exploration. Although self-esteem has been examined and its potential impact on health-promoting
behaviors in several health promotion studies has been consistently reported, no studies have
included in school-age children. Self-esteem among school-age children needs to be studied,
because the results may prove useful in enhancing the health of this population. Self-esteem and
health-promoting behaviors have been investigated in literature on school-age children that
supports the linkage among the current study variable.
Depressive mood were negatively associated with health-promoting behaviors, especially
in health-promoting eating behaviors such as binge eating (Stice, Presnell, & Spangler, 2002),
disordered eating (Neumark-Sztainer et al., 2000), and bulimic symptoms (Roew, Pickles,
Simonoff, Bulik, & Silber, 2002). Studies revealed that regular exercise has been shown to
decrease in depression (Dunn, & Madhukat, 2001; Higgins et al., 2003). There is insufficient
knowledge to identify whether depressed mood continued to influence health-promoting behaviors
two years later in school-age children.
Previous cross-sectional studies revealed that predictors of health-promoting behaviors are
associated more with family and school. There is a substantial literature base examining factors
associated with physical activity (PA) among youth (Crocker et al., 2003). Such factors could be
organized according to Pender’s HPM (2002) which suggest that affective state (self-esteem,
depression), social support (peer support, parents support (parents emotional support and
closeness, parent involvement), situational influence (school safety and neighborhood safety) are
crucial considerations when attempting to understand health-promoting behaviors. Studies

50

examined the relationship between the home and neighborhood environment and PA (Ferrial et al.,
2007; Kligemam, Sallis, Ryan, Frank, & Nader, 2007; Norman et al., 2006).
Srof and Velsor-Friedrich (2006) used Pender’s HPM to investigate in relation to
adolescent health. The HPM variables that did not demonstrate significance on the regression
model were entered into an exploratory path analysis equation in order to examine the indirect
effects of self-efficacy, grades, perceived health, social support, and exercise norms as modulated
by the overall perceived benefits-barriers differential. This study implied that home context in
which person live their lives can either sustain and expand their health potential or inhibit the
emergence of health and well-being. Cartland and Ruch-Ross (2006) found that older children
scored higher than younger ones on factors generally associated with improved health behaviors
(such as health knowledge and refusal skills). Older children scored lower than younger children
on healthy behaviors, especially risk behaviors. As health knowledge, refusal skills and other
protective factors increased, health behaviors improved slightly. Especially for gender, Garcia et
al. (1995) demonstrated that background characteristics, specifically gender, played a significant
role in determining exercise behavior. The study also supported the HPM as a useful model for
explaining physical activity and health-promoting behavior among teens. This study demonstrated
an indirect effect of self-efficacy and social support on physical activity behavior that may
contribute to adolescent health-promoting behavior. Also, girls scored significantly lower than
boys in terms of exercise, self-esteem, perceived health status, previous exercise experience, and
self-schema for exercise. However, there were no significant gender differences for self-efficacy,
social support, the benefits-barriers differential, or access to exercise facilities.
The results of previous studies (Garcia et al.,1995, 1998; Wu & Pender, 2002) supported

51

the theoretical propositions of the HPM. The studies provided evidence that age/grade and gender
are an important background characteristics influencing physical activity behavior. Frenn and
Malin (2003) used a combined the HPM and transtheoretical model (TM) approach to examine
diet and physical activity behaviors across stages of change in middle school students. The
findings demonstrated a significant difference in behavior across stages of change. The percentage
of dietary fat content and the temptation to consume high fat foods declined, and access to low fat
food increased. Physical activity behavior also increased overall across the stages of change. There
is insufficient knowledge to identify whether school context and home context influence HPBs in
school-age children. The study is particularly interested in individual context, including selfesteem and depression, which are believed to have a potential influence on HPBs. Both school and
home context may be an especially crucial factor for certain groups of school-age children, and
indeed physical limitations and the lack of accessibility, time, and resources have recently been
found in the literature on Western subjects to be barriers to HPBs.
Those variables may be barriers to engaging HPBs among school-age children as well. For
that reason, the proposed study investigates the relationship between variables as previous
mentioned among school-age children and HPBs. In addition, numerous studies have shown selfesteem to be a significant predictor of HPBs; therefore, this study addresses the linkage between
self-esteem, and media use as a determinant of HPBs among school-age children. Finally, a
number of studies have shown that interpersonal influences, defined as social support from parents
and peers, could affect individuals’ predispositions to engage in HPBs, and this variable is
included in the proposed study.

52

Summary
In summary, all of the studies cited above suggest the usefulness of the HPM in explaining
the occurrence of health-promoting behaviors. A review of the research literature indicates that the
HPM has been used in relation to various populations to study variables that could impact healthpromoting behaviors in school-age children. The results from previous studies could not be
generalized to school-age children, however, because of the different nature of school context,
home context, individual context, and exposure to media. In addition, some studies have been
limited by small sample sizes, convenience of sample selections, and the use of cross-sectional
designs.
The contributions of health-promoting behaviors and the factors influencing these healthpromoting behaviors in school-age children have few reports in the literature. In addition, there are
no current studies assessing the influence of those factors as mentioned above in school-age
children. To broaden our knowledge of health-promoting behavior and its related factors, research
is required to apply Pender’s HPM (2002) to a sample of school-age children. Such research could
provide data regarding the benefits of the HPM in understanding health-promoting behaviors. In
addition, the research could provide a foundation for the development of appropriate intervention
techniques to help school-age children increase their health-promoting behaviors and guide
research related to health-promotion activities.

53

CHAPTER 4
METHODS
Chapter four focuses on the research methods, including research design, sample, data
collection procedures, instruments, protection of human subjects, and data analysis strategy.
Design of the Original Coordinated Community Assessment Study
The Coordinated Community Assessment (CCA) was a longitudinal study designed to
conduct community needs assessment and evaluate student outcomes based on school program
participation in youth development programs. The guiding questions underlying CCA are: 1) What
are student needs and strengths, and do these needs and strengths differ across communities, 2)
What are the critical times to begin prevention activities across different domains of student
outcomes, 3) What are the differences in needs and strengths for boys and girls, 4) How do home,
school, and neighborhood environments influence student outcomes, and how do these influences
change over time and, 5) How does participation in school-based support services influence
student outcomes. The primary assessment instrument used in the study, the Community Student
2 2

Survey (C S ) is a survey designed to assess student outcomes related to health, safety, and
th

nutrition programs and services provided by schools to students from 4 to 12

th

grade throughout

2 2

Genesee County. The C S was designed in collaboration with multiple community partners.
2 2

The C S survey was administered in the spring of 2006, 2007, 2008, 2009, and 2010. A
total of 32,210 students completed the survey at least one time during the five years of the study.
Of these students, 12,450 students completed the survey in 2006. In 2007, 10,145 students
completed the survey. In 2008, 11,597 students completed the survey. In 2009, 12,175 students
54

completed the survey, and in 2010, 8,107 students completed the survey.
Procedure for Data Collection. Students completed the Coordinated Community Student
2 2

Survey (C S )-208 items, a 45-minute self-report questionnaire comprised of 208 items and
designed to assess student’s attitudes, beliefs, and behaviors about physical, social and mental
health as well as school, home, and neighborhood environments. Consent forms were distributed to
the parents of all students in participating school via student backpacks and home mailings.
Approximately 40% of all parents provided consent for students to participate in the study. Signed
parental consent forms and student assent forms were collected prior to data collection. University
research staff administered surveys at each participating school during school hours. Personal
identifying data were obtained for the purpose of connecting survey data across the years to build a
longitudinal database. Each student was assigned an identifying code in the database so that
researchers using the data set would have de-identified data.
Design of Current Study.
The present study uses a secondary data analysis using a de-identified longitudinal dataset
from the CCA. The current study involved one cohort (three waves) of elementary students in a
Midwest area comprised of over 100 urban, suburban, and rural schools (Barnes, Almerigi, & Hsu,
2010).
A cross sequential design was used to track children on a period (grades 4-6; grades 5-7,
and grades 6-8) (see Table 1).

55

Table 1
The Cross-Sequential Design of the Study by Grade and Wave
Child Grade

Wave 1 2006 (n)

Wave 2 2007 (n)

Wave 3 2008 (n)

th

424

th

472

424

th

474

472

424

474

472

4
5
6

th

7

th

474

8

Total N

1370

1370

1370

Note: Wave 2 included eight students who repeated a grade between wave 1 and wave 2
<1%; Wave 3 included 11 students who repeated a grade between wave 2 and wave 3
< 1%. 560 males, 810 females
Note, Students at each beginning grade were assessed at each of the next two grades.
Sample and Sampling Procedures
th

th

The sample consisted of all 1,370 students who were in the 4 to 6 grade in the first year
of the study (2005), and who also completed the survey in 2006 and 2007. The mean number of
students per school was 16.70 (SD= 22.28, Range= 1-156). Data from the full study indicate that
of the 32,210 who completed the survey, three or more consecutive waves of data were obtained
for 4,048 students (12.6%).

56

Table 2
The Percentage of Sexual and Racial of the Sample (n = 1370)
Demographic characteristics
Sex
Race

N
560
810
223
24
7
23
100
990
3

Boys
Girls
African American
American Indian
Asian
Hispanic
Multiracial
White
Not reported

Percentage
41%
59%
16%
2%
Less than 1%
2%
7%
72%
Less than 1%

Approximately 59% of the sample were girls (n=810) and 41% were boys (n=560).
Seventy-two percent (n=990) of the students self-identified as White, and 16% (n=223) of students
self-identified as African American. A small percentage identified as another ethnicity (2%
American Indian, n=24; <1% Asian, n=7; 2% Hispanic, n=23; and 7% multiracial, n=100).
Sampling Procedures
The sample selection from the original data set for this study were obtained by the onestep procedure that selected all of the school-age children in the original data set.
Inclusion criteria. Selection criteria are based on all 4

th

th

grade to 6 grade children and no

missing data at any of the three waves of data collection.
Exclusion criteria. Only children who did not have parental consent or those with parental
consent who refused to sign an assent form were excluded from the original study.

57

Instrument
The demographic questionnaire consisted of five questions. The data collector asked
common demographic questions such as name, birth date, gender, and race.
Scale Descriptions
Data from four broad domains and ten constructs were utilized for this study:
1. School context: Peer social support at school (4 items), peer hostility and bullying at school (9
items),
2. Home context: Parent involvement (9 items), parent emotional support and closeness (8 items)
3. Affective state: Depression (13 items) and self-esteem (8 items)
4. Health-promoting behavior: Dietary intake (6 items), physical activity (2 items), Media use (3
items).
School context and home context were measured in wave 1 of the study. School context
was comprised of three variables: peer social support at school, peer hostility and bullying at
school. Students self-reported their experience of peer social support at school on six items scored
on a Likert scale ranging from “Strongly Disagree,” “Disagree,” “Agree” to “Strongly Agree”
(α=.75). Students self-reported their experience of peer hostility and bullying at school on 9 items
scored on a Likert scale ranging from “Never,” “Not much,”“Sometimes” to A lot (α=.87). A
standardized factor score was computed for school context using at least square regression
approach. All scale scores loaded strongly onto the single factor with evidence of correlational
relationships between the factor scores and factors. A high score in school context indicates a
student perceives greater risk in his or her peer in school.
Home context was comprised of two variables: parent emotional support and closeness, and

58

parent involvement. Students self-reported their experience of parent emotional support and
closeness on nine items scored on a Likert scale ranging from “Strongly Disagree,” “Disagree,”
“Agree” to “Strongly Agree” (α=.90). Students self-reported their experience of parent
involvement on 9 items scored on a Likert scale ranging from “Never,” “Not much,” “Sometimes”
to A lot (α=.79). A standardized factor score was computed for school context using at least square
regression approach. All scale scores loaded strongly onto the single factor with evidence of
correlational relationships between the factor scores and factors. A high score in home context
indicates greater levels of support from parents in the child’s life.
Affective state was measured at wave 2 of the study, approximately one year after Wave 1
data were collected. Students self-reported their experience of depression on 13 items scored on a
Likert scale ranging from “Not at all like me,” “Not much like me,” “Kind of like me” to “A lot
like me” (α=.90). The depression scale score was recoded to represent a lack of depression
symptoms. Students self-reported their experience of self-esteem on seven items scored on a Likert
scale ranging from “Not at all like me,” “Not much like me,” “Kind of like me” to “A lot like me”
(α=.78). A standardized factor score was computed for affective state using a least squares
regression approach with Varimax rotation. Both scale scores loaded strongly onto the single
factor with evidence of correlational relationships between the factor scores and factors. A high
score in affective state indicates a positive emotional state.
Media use was measured at wave 2 of the study, approximately one year after Wave 1 data
were collected. Student self-reported their experience of exposed to media on three items scored on
a Likert scale ranging from “Less than 1 hour a week,” “1-3 hours a week,” “4-6 hours a week,”
“7-9 hours a week,” “10-12 hours a week,” “13-15 hours a week” to 16+ hours a week”.

59

Item responses were dichotomized into two categories: “healthy” and “at risk”. Item responses of
less than 1 hour a week, to 7-9 hours a week, were categorized as healthy. Item responses of 10 or
more hours a week were categorized as “at risk”. The dichotomized variables were then summed
to create a total health-promoting behavior score reflecting cumulative risk (ranging from 0 =
healthy response to 3 = at risk on all 3 items). A high score in media use indicates a student
perceives greater risk in his or her the exposed media.
Health-promoting behaviors at Wave 3 were assessed from students’ report of their dietary
intake (six items) and physical activity (two items). Dietary intake items included self-report of
frequency of intake of healthy and unhealthy foods in the past year on a scale of “Never,” “Not
much,” “Sometimes” to “A lot”. Item responses were dichotomized into two categories: “healthy”
and “at risk”. A “Never” or “Not much” response to healthy food items was categorized as
“healthy” for healthy food items and “at risk” for unhealthy items. Physical activity items included
self-reported engagement in physical activity “during and average week” on a scale of “Never,” “1
or 2 days a week,” “3 or 4 days a week” and “5 to 7 days a week”. Item responses were
dichotomized into two categories: “healthy” and “at risk”. Item responses of “Never” or “1 or 2
days a week” were categorized as “healthy”. The dichotomized variables were then summed to
create a total health- promoting behavior score reflecting cumulative risk (range from 0 = at risk
response on all 8 items to 8 = healthy on all eight items). A high score in HPBs indicates a student
2 2

perceives more involvement in his or her HPBs. The C S subscales and their alpha reliability
indices are presented in Table 3.

60

Table 3
2 2

The C S Subscales Constructs, Items, Rating Scale and alpha reliability indices

Construct

Items

Rating Scale

Reliability

1.Affective
State
-Depression

During the PAST YEAR, how often did the
following things happen?
a. I felt good.
b. I felt upset.
c. I felt that I could not stop being sad.
d. I felt good about the future.
e. I had a hard time sleeping.
f. I was happy.
g. I talked less than usual.
h. I felt lonely.
i. I felt sad.
j. I was bothered by things that usually
don’t bother me.
k. I felt down and unhappy.
l. I felt like I was too tired to do things.
m. I felt like crying

-Not at all
-A little
-Some
-A lot

.89

-Self-esteem

How well do the following statements
describe you?
a. I think I am good looking.
b. I think I’m pretty good at figuring out
problems at school.
c. I think I am very smart.
d. I have lots of friends.
e. I get along well with other kids.
f. I like who I am.
g. I am good at sports.
h. I think I am a good person.

-Not at all like me
-Not much like me
-Kind of like me
-A lot like me

.75

2.School
environment
- Peer
hostility and
bullying at
school

In the PAST YEAR, how often did the
following things happen
a. I felt safe at my school.
b. A kid at my school hit or pushed me
when they were not playing around.
c. A kid at my school said he or she was
going to hurt me.

-Never
-Not much
-Sometimes
-A lot

61

Table 3 (cont’d)
Construct

Items

Rating Scale

Reliability

d. A kid at my school told lies or false
rumors about me.
e. I skipped class or school without
permission because I was afraid of being
hurt by another student.
f. I have been left out or ignored by kids at
my school.
g. My money was taken away by a kid at
my school.
h. My things were taken away or broken by
a kid at my school.
i. I saw violence in my school (for example,
bullying, hitting, kicking, punching)
-Peer social
support at
school,

How much do you agree or disagree with
the following statements?
a. There are students at my school who
really care about me.
b. I feel alone when I’m at my school.
c. It is hard to make friends at my school.
d. I usually spend lunch and/or recess time
alone at my school.

-Strongly disagree
-Disagree
-Agree
-Strongly agree

.75

3.Home
environment
- Parent
emotional
support and
closeness

How much do you agree or disagree with
the following statements? Your parents /
caregivers…
a. enjoy spending time with you.
b. listen to you.
c. spend time reading with you.
d. know your friends.
e. do fun things with you.
How much do you agree or disagree with
the following statements?
a. I like to talk to my parents / caregivers.
b. I like to spend time with my parents /
caregivers.
c. I try to make my parents / caregivers
happy

Strongly disagree
-Disagree
-Agree
-Strongly agree

.90

62

Table 3 (cont’d)
Construct

Items

Rating Scale

- Parent
involvement

How often do your parents / caregivers
(whoever takes care of you)…
a. ask about what you’ve been doing in
school?
b. check whether you have done your
homework?
c. see if your homework is correct?
d. come to your school’s activities or help in
your classroom?
e. talk to you about your problems?
f. limit the amount of time you can watch
TV?
g. limit the amount of time you can play
video
games?
h. tell you that you shouldn’t drink alcohol or
use drugs?
i. let you stay home alone with no adults
around?

-Never
-Not much
-Sometimes
-A lot

4. Health
behaviors
- Dietary
Intake

In the PAST YEAR, how often did you drink
or eat…
a. soda pop?
b. milk, cheese, or yogurt?
c. 100% fruit juice? (such as orange juice,?
apple juice, grape juice)
d. French fries, potato chips, or other fried
potatoes
e. fruit?
f. vegetables?

-Never
-Not much
-Sometimes

- Physical
activity

During an AVERAGE WEEK, how often do
you…
a. exercise or play for 30 or more minutes
where you were sweating and breathing
hard?
b. lift weights or do strength training?,

-A lot
-Never
-1to2 day a week
-3-4 days a week

63

-5to7days a week

Reliability
.79

Table 3 (cont’d)
Construct
-Media use

Items

Rating Scale

During an AVERAGE WEEK, how often do
you…
a. watch TV?
b. play video games?
c. surf the internet (not for homework)?

Reliability

-Less than 1 hour
a week
-1-3 hours a week
-4-6 hours a week
-7-9 hours a week
-10-12 hours a
week
-13-15 hours a
week
-16+ hours a week

Note. C2S2 = Genesee County Coordinated Community Student Survey, 2004. © Jessica BarnesNajor. Questions reproduced by permission. For access to the questionnaire contact Dr. Jessica
Barnes-Najor (barnes33@msu.edu).
Data from four broad domains and eight constructs were utilized for this study:
In its final form, the survey assessed 24 constructs, each comprised of a minimum of four items.
Data from four broad domains and nine constructs were utilized for this study.
Data Management
Data provided for this study were fully cleaned and no missing data were present.
Plan for Data Analysis
Research questions were identified and addressed through data analysis using the Statistical
Package for Social Sciences (SPSS) and Mplus. Data analysis strategies included computation of
descriptive statistics, Cronbach’s alpha coefficient, and structural equation modeling (SEM). SEM
was used to examine the hypothesized relationships between HPBs in school-age children that
could be explained by school context, home context and media use, and the extent to which media
use mediated other effects. Maximum likelihood (ML) estimation is a normal theory and fullinformation method set and is the default in most SEM programs.
64

All of the estimates of model parameters could be tested statistically in a simultaneous
analysis of the entire system of variables, maximizing the likelihood (the continuous
generalization) that the data (the observed covariance) were drawn from the population (Kline,
2011). To perform the ML estimation method, the following assumptions need to be met:
independence of the scores, multivariate normality of the endogenous variables, and independence
of the exogenous variables and error terms.
In order to evaluate the model fit, multiple fit indicators should be used to examine the
2

model fit, including the Chi-square (χ ), the root mean square error of approximation (RMSEA),
the Bentler’s Comparative Fit Index (CFI), the Tucker Lewis Index (TLI), the standardized root
2

mean square residual (SRMR). The Chi-square (χ ) and RMSEA were all essentially used to report
2

the model fit. A Chi-square (χ ) test, which results in the indication that something was not
statistically significant, is preferable to determine the acceptability of model fit (Kline, 2011).
The RMSEA values less than .06 are typically indicative of a good fit (Hu, & Bentler,
1999). Values between .08 and .10 are generally indicative of a mediocre fit (MacCallum,
Browne, & Sugawara, 1996) and values greater than .10 are usually considered to be indicative of
a poor fit (Siedlecki, et al., 2010). The CFI and TLI were also used to identify the fit statistics;
values closer to 1.0 indicate a better fit (Hu, & Bentler, 1999). The value of these indicators greater
than .95 is considered to signify a good fit (Hu, & Bentler, 1999). However, sometimes a cut-off of
greater than .90 is also used (Bentler, 1992). Finally, the SRMR was also used to determine the
model fit. The combination threshold for concluding acceptable fit is based on the following
indicators: (a) the CFI is greater than or equal to .95 and (b) the SRMR is less than or equal to .08
65

(Hu, & Bentler, 1999).
Analytic procedures are described below for each research specific research questions:
Research Questions 1: Is there a direct relationship between events that occur in school and home
contexts for school-age children and health-promoting behaviors? The SEM analysis was
conducted to determine the contributions of each individual variable to the total variance of the
HPB’s and the mediation effects of media.
Research Questions 2: To what extent does children’s affective state and use of media mediate the
relationship between school and home influences on health-promoting behaviors? A SEM analysis
was used to assess the mediation effects.
Research Question 3: Is there a relationship between media use and affective state in school-age
children? A SEM analysis was used to assess the relation between these two variables.
Research Question 4: Do mediation effects vary as a function of sex? A Stacked SEM analysis
was used to examining gender difference in the model and model paths.
Protection of Human Subjects
This study received approval from the Michigan State University Institutional Review
Board overseeing protection of human subjects. Dr. Jessica V. Barnes-Najor, the principal
investigator of the research project, approved access to the database. The data set obtained from
the parent study was de-identified so there was no contact with the original participants. The
dataset was encrypted and saved electronically on a secure server and the password-protected.

66

CHAPTER 5
RESULTS
Descriptive Statistics. Descriptive statistics were used to describe the variables by the
range, means and standard deviations. A SEM) analysis was used to describe the relationships
between the variables.
Table 4 presents the mean scores, standard deviations and range of scores for the study
variables. Examination of individual scale scores provided evidence that the range of scores was
acceptable for every variable measured in this study. Adequate variance existed in student
responses to survey items.
As shown in Table 5, statistical comparisons for sex differences indicate that boys perceive
greater risk related to peer relationships than girls, as indicated by peer social support (t = 2.44,
p=.014) and peer hostility and bulling (t= 4.021, p=.000). Girls report greater support from
parents than did boys as indicated by parent emotional support (t = -.3389, p=.001) and parent
involvement (t= -2.018, p=.044). Consistent with similar studies, girls report more depressive
symptoms than boys (t= -3.953, p=.000). Finally, girls report lower use of media than boys
(t=4.413, p=.000). There are no significant differences between boys and girls in the measurement
of self-esteem (t= -1.269, p=.205), and HPBs (t=-1.54, p=.122).
Pearson’s correlation analyses revealed parent emotional support and involvement and selfesteem were all positively correlated with HPBs. Peer social support, peer hostility and bulling at
school, media use and depression were all negatively correlated with HPBs. Self-esteem was found
to be negatively correlated with media use (p<.001). Depression was found to be positively
correlated with media use (p<.001).

67

Table 4
The Mean, Range, Standards Deviation, Construct, Variables and Scale Score Characteristics
by Sex (N=1,370
Sex

School context

Parent emotional support
Home context

2

Parent involvement

Media use

3

Media use

Depression
Affective state

4

Self-esteem

5

HPB Sum

HPBs

1.70
1.62
1.65
1.78
1.66
1.70

1 to 4

3.23

0.65

1 to 4

3.35

0.57

1 to 4
1 to 4
1 to 4
1 to 4

3.30
2.95
3.02
2.99

0.60
0.55
0.53
0.54

0 to 3
0 to 3
0 to 3
1 to 4
1 to 4
1 to 4
1 to 4
1 to 4
1 to 4

0.70
0.80
0.81
1.89
2.04
1.97
3.34
3.38
3.36

0.91
0.98
0.98
0.64
0.68
0.67
0.53
0.49
0.52

Boy
Girl
Total

Peer hostility and bulling

1 to 4
1 to 4
1 to 4
1 to 4
1 to 4
1 to 4

Boy
Girl
Total
Boy
Girl
Total
Boy
Girl
Total

1

Boy
Girl
Total
Boy
Girl
Total

Standard
Deviation
0.65
0.62
0.63
0.60
0.53
0.56

Total
Boy
Girl
Total

Peer social support at school

Mean

Boy

Variables

Range

Girl

Constructs

0 to 8
0 to 8
0 to 8

2.98
4.90
4.94

1.44
1.28
1.35

1

Note, A high score in school context indicates a student perceives greater risk in his or her peer in
2

school, A high score in home context represents greater levels of support from parents in the
3

child’s life, A high score in media use indicates a student perceives greater risk in his or her
4

5

exposure to media, A high score in affective state indicates a positive emotional state, A high
score in HPBs indicates a student perceives more involvement in his or her HPBs.
68

Table 5
Sex Difference Comparisons Mean Difference on Variables Used in the Study
Variables

95% Confidence Interval
of the Difference

t-test for Equality of Means
Sig.

T
Peer social
support
Peer hostility
and bulling
Parent
emotional
support
Parent
involvement
Self-esteem

Df

Mean
(2-tailed) Difference

Std. Error
Difference

Lower

Upper

2.449

1364

.014

.08592

.03509

.01710

.15475

4.021

1363

.000

.12442

.03094

.06372

.18511

-3.389

1368

.001

-.10166

.03000

-.16051

-.04281

-2.018

1367

.044

-.06023

.02985

-.11879

-.00167

-1.269

1368

.205

-.03562

.02807

-.09069

.01944

Depression

-3.953

1366

.000

-.14464

.03659

-.21642

-.07286

Media use

4.413

1357

.000

.23800

.05394

.13220

.34381

-1.547

1368

.122

-.11510

.07438

-.26101

.03081

HPBs
Note. p<.05

69

Table 6
The Correlation Matrix for All Study Variables.
Parent
Peer
Peer
emotional
Parent
hostility
social
SelfMedia
support involvement and bulling support esteem Depression use HPB
1

.485**

-.348**

-.368**

.347**

-.269**

-.073** .218**

Parent
involvement

.485**

1

-.173**

-.169**

.247**

-.156**

-.126** .224**

Peer hostility
and bulling

-.348**

-.173**

1

.491**

-.296**

.420**

.110** -.083**

Peer social
support

-.368**

-.169**

.491**

1

-.320**

.337**

.022 -.196**

Self-esteem

.347**

.247**

-.296**

-.320**

1

-.377**

-.087** .229**

Depression

-.269**

-.156**

.420**

.337**

-.377**

1

.135** -.131**

Media use

-.073**

-.126**

.110**

.022

-.087**

.135**

1

-.134**

HPB

.218**

.224**

-.083**

-.196**

.229**

-.131**

-.134**

1

Parent
emotional
support

Note. **p< 0.01; *p<.05
Result and Analysis
The SEM equation models specifying the relationships between variables in this study were
estimated using Mplus Version 7. SEM represents a synthesis of path analysis, involving the
comparison of hypothesized model covariance with observed covariance (see Appendix A).
Advantages of this procedure include the generality and flexibility of model specification and the
ability to assess fit of the hypothesized model to the observed data (Kline, 2011)
A SEM analysis was computed to determine the amount of variability of HPBs in schoolage children that could be explained by school context (peer social support, peer hostility and
70

bulling at school), home context (parent emotional support and parent involvement), media use
(television, video games and computers), and affective state (self-esteem and depression). The
SEM analysis was conducted to determine the contributions of each individual variable to the total
variance of the HPBs.
The initial model was specified based upon the relationships among variables that were
hypothesized to exist or not to exist (see Appendix B). This distinction is important because any
unspecified relationships among variables were assumed to be equal to zero. After specification of
the initial model was completed, the alternative model was tested. In testing the alternative model,
the following steps were followed: 1) Evaluation of paths by examining significance of parameter
estimates, 2) consideration of the change in explained variance for health-promoting behaviors,
and 3) testing significant improvement in model fit by fixed parameters. Upon completion of these
steps, it was found that the paths between school context and media use and school context and
HPBs should be set to 0. As presented in Table 7, when these paths were set to 0 the model
provided a good fit as indicated by the fact that the chi-square became insignificant (χ2(df=2)=
2.318, p=0.313). Additional indices indicated that the alternative model fit the data (Tucker-Lewis
Index=0.997, Comparative Fit Index=0.999, Root Mean Square Error of Approximation=0.011
and Standardized Root Mean Square Residual=0.009). The alternative model restricted the paths
from school context to HPBs to 0, thereby providing a better fit than the initial model which
assumed these relationships existed.

71

Table 7
The Comparison of the Model Fit Indices between the Initial and the Alternative Model
Fit indices
2

Chi-square (χ )

Models
CMIN
Initial model
Alternative model

DF

0.000
2.318

0
2

0.000
0.313

Δ

P

72

2

ΔDF=2; Δχ =2.318

Wave 1
School Context
=Peer hostility and bulling
=Peer social support
0.000

0.000
0.403***

Wave 2
Media Use
-Television
-Video Games
-Computer

-0.070**
Wave 3
HPBs
-Dietary intake
-Physical activity

0.091**
-Computer Use

-0.116***

Wave 2
Affective
State
-Self-esteem
-Depression

-0.213***

-0.104***

0.179***

Wave 1
Home Context
=Parent Emotional support
and closeness
=Parent involvement

Figure 3. Path Model 1 for School Context, Home Context, Affective State and Media Use with
Standardized Coefficients
*p<0.05, **p<.01, ***p<.001; solid line represent = p<.0.05.

Research Question 1: Is there a direct relationship between events that occur in school and home
contexts for school-age children and health-promoting behaviors?
As described above, in order to obtain a model that fit the data the paths leading from
school context to media use and to HPBs were restricted to 0. This indicates no direct effect of
73

school context (β =0.000, p=0.999) on HPBs. Alternately, home context ((β =0.179, p=0.000) had
a significant direct effect on HPBs.
Table 8
Mediation, Effect Size, Direct Effect and Indirect Effects of School Context on HPBs
Mediation

β

Direct effect
Indirect effects
Through media use
Through affective state
Total effect
Percentage of indirect effects

0.000
-0.042
0.000
-0.042
-0.042
-0.042/-0.042=100%

Table 9
Mediation, Effect Size, Direct Effect and Indirect Effects of Home Context on HPBs
Mediation

β

Direct effect
Indirect effects
Through media use
Through affective state
Total effect
Percentage of indirect effects

0.179
0.030
0.008
0.022
0.209
0.030/0.209= 14%

Research Question 2: To what extent does children’s affective state and use of media
mediate the relationship between school and home influences on health-promoting behaviors?
The SEM analysis indicated that only children’s affective state mediated the effects of
school context on children’s HBPs (Table 8; 100% mediation with 4.2% effect size). Children’s
media use (.8% effect size) and affective state (2.2% effect size) mediated the effects of home
context on children’s HPBs (Table 9; 14% mediation with 3% total indirect effect), although home
context also independently contributed to HPBs (17.9% effect size).
74

Research Question 3: Is there a relationship between media use and affective state in
school-age children? As shown in Figure 3, the result in this current study indicate that affective
state is significantly positively related with media use (β <.091, p=000); the covariance among the
matrices for media use and affective state, within the larger model, were significantly positively
related.
Research Question 4: Do mediation effects vary as a function of sex? Following the same
procedure used to test the hypothesized model to examine the direct and indirect effects of the
study variables on HPBs, a stacked SEM analysis, using the alternative model was conducted to
examine sex differences in the model and model paths.
The test of the stacked models indicated that the overall model fit was good. The ChiSquare test of model fit was not statistically significant, indicating a good-fitting model (χ2 (df=4)
=5.432, p=0.254). The Tucker-Lewis Index (TLI=0.989), Comparative Fit Index (CFI=0.998),
Root Mean Square Error of Approximation (RMSEA=0.022) and Standardized Root Mean Square
Residual (SRMR=0.013) all indicated a good-fitting model.
The results from the stacked SEM analysis are shown in Figures 4 and 5. For both boys and
girls, home context independently contributed to HPBs (effect size for boys: 18.6%, effect size for
girls: 18.4%) while school context did not have a direct effect on HPBs. Media use was a
significant predictor of HPBs for boys only. As shown in Tables 11 and 12, affective state
mediated the relationship between home and school contexts and HPBs for both boys (effect size
for home mediation: 1.7%; effect size for school mediation: -4.4%) and for girls (effect size for
home mediation: 2.5%; effect size for school mediation: -3.8%). Alternatively, media use was a

75

mediator in the relationship between home context and HPB for boys only, although this effect
was small at 1.1%.
Table 10
The Comparison of the Model Fit Indices between the Initial Model and the Alternative
Model for Sex Differences
Fit indices
2

Chi-square (χ )

Model
Initial model
Alternative model

CMIN
0.000
5.342

DF
0
4

P
0.000
0.254

Δ
2

ΔDF=4Δ χ =5.342

Table 11
Mediation, Effect Size, Direct Effect and Indirect Effects of School Context on HPBs by
Sex Differences
Mediation
Direct effect (1)
Indirect effects
Through Media Use
Through Affective State
Total effect
Percentage of indirect effects

Boys
β
0.000
-0.044
0.000
-0.044
-0.044
-0.044/-0.044=1=100%

Girls
β
0.000
-0.038
0.000
-0.038
-0.038
-0.038/-0.038=1=100%

Table 12
Mediation, Effect Size, Direct Effect and Indirect Effects of Home Context on HPBs by
Sex Differences
Mediation
Direct effect (1)
Indirect effects
Through Media Use
Through Affective State
Total effect
Percentage of indirect effects

Boys
β
0.186
0.028
0.011
0.017
0.214
0.028/0.214=0.130=13.0%
76

Girls
β
0.184
0.031
0.006
0.025
0.147
0.031/0.147=0.210=21.0%

Wave 1
School Context
=Peer hostility and bulling
=Peer social support
0.000

0.000
0.453***

Wave 2
Media Use
-Television
-Video Games
-Computer

-0.094**
Wave 3
HPBs
-Dietary intake
-Physical activity

0.093**
-0.114**

-Computer2
Wave Use
Affective State
-self-esteem
-depression
-0.175***

-0.097**

0.193***

Wave 1
Home Context
=Parent Emotional support
and closeness
=Parent involvement

Figure 4. Path Model 2 in Boys for School Context, Home Context, Affective State and Media
Use with Standardized Coefficients
*p<0.05; **p<.01; ***p<.001; solid line represent =p<.0.05.

77

Wave 1
School Context
=Peer hostility and bulling
=Peer social support
0.000
0.383***

-0.103**

0.000

Wave 2
Media Use
-Television
-Video games
-Computer

-0.058

0.117**
*
-Computer2
Wave
Use
Affective State
-Self-esteem
-0.100**
-Depression
-0.247***

Wave 3
HPBs
-Dietary intake
-Physical activity

0.178***

Wave 1
Home Context
=Parent emotional support
and closeness
=Parent involvement

Figure 5. Path Model 3 in Girls for School Context, Home Context, Affective State and Media
Use with Standardized Coefficients
*p<0.05; **p<.01; ***p<.001; solid line represent =p<.0.05.

78

Summary
This chapter presented the results from quantitative analysis. There is no direct effect of
school context on HPBs two years after school context was assessed. Alternately, home context
had a significant direct effect on HPBs. Only children’s affective state mediated the effects of
school context on children’s HPBs. Children’s media use and affective state mediated the effects
of home context on children’s HPBs, although home context also independently contributed to
HPBs. Affective state is significantly positively related with media use; the covariance among the
matrices for media use and affective state, within the larger model, were significantly positively
related. For both boys and girls, home context independently contributed to HPBs while school
context did not have a direct effect on HPBs. Media use was a significant predictor of HPBs ’s for
boys only. Affective state mediated the relationship between home and school contexts and HPBs
for boys and for girls. Alternatively, media use was a mediator in the relationship between home
context and HPBs for boys only.

79

CHAPTER 6
DISCUSSIONS AND IMPLICATIONS
This study has been designed to assess predictors of HPBs in fourth to sixth grade children
using a cross-sequential longitudinal design to track predictors over a three-year period. Specific
attention has been given to the extent to which children’s use of media and/or their affective state
mediate the relationship between peer (school context) and parent (home context) predictors of
dietary intake and physical activity (HPBs). The study has been designed as an addition to existing
literature investigating direct and indirect effects of factors influencing HPBs, conceptualized
within the historical context of Pender’s HPM (2002).
Descriptive analyses for sex differences across all variables produced effects that are
consistent with similar studies. The boys in this study reported greater risk in peer relationships,
primarily in the reporting of hostile and aggressive behaviors, as well as greater use of media than
the participating girls. The girls in this study reported receiving greater emotional support from
parents and reported more depressive symptoms than the boys. There were no significant sex
differences regarding measures of self-esteem.
Analyses based on the study’s research questions generally support mediational effects.
For each of the research questions, SEM analyses indicated:
Research Question1: Only parent variables of home context had direct effects on HPBs.
Research Question 2: Only children’s affective state mediated the effects of school context (peers)
on HPBs. Although media use and affective state mediated the effects of home context, home
context also independently contributed to predicting HPBs.
Research Question 3: Affective state and media use were positively related.

80

Research Question 4: For both boys and girls, affective state completely mediated the relationship
between school context and HPBs, though affective state and media use only partially mediated the
effects of home context and HPBs.
Discussion of Research Question 1
The first question asked whether there was a direct relationship between school and home
context and children’s health-promoting behaviors. Results indicated that only parent variables
(home context) had direct effects on HPBs.
Home Context and HPBs. Only parent variables of home context had direct effects on
HPBs. This observation is consistent with those found in studies of parental influence on HPBs in
school-age children (Borkowsky et al., 2002; Mounts, 2001). In these studies, parent emotional
support and parent closeness demonstrated significant, direct effects on the HPBs of their children.
In the current study, it is possible that children experiencing risky health behaviors and having
high levels of parental emotional support, involvement and closeness may be more likely to engage
in activities which enhance their HPBs. If so, it may explain why children who have reported
perceived emotional support they receive from their parents, as well as the degree to which parents
are involved in child rearing activities, are more likely to engage in HPBs. Children’s perception
of support from their parents significantly predicted their engagement in HPBs.
Studies of parental influences on HPBs clearly point to the importance of parental
monitoring for promoting healthy behavior in children (Aufseeser et al., 2006; Kremarik, 2000;
Sallis et al., 2000; Okun et al.,2002). Such studies suggest that positive parent involvement in
children’s HPBs decreases risk for deviance by promoting competence and the internalization of
parental values.

81

Moore and Harre (2007) examined the eating behaviors, physical exercise and television
viewing of secondary school students and investigated their relationship with parental monitoring
and family cohesion. The results of their studies showed that parental monitoring and family
cohesion were positively related to participants’ reports of eating breakfast and healthy foods.
These results also showed a negative relationship in the participants’ reports of buying their own
food and eating unhealthy foods. Exercise rates were moderately correlated with eating healthy
foods and weakly related to parental monitoring and family cohesion. Television viewing was
related to eating unhealthy foods.
Findings demonstrated that parents play a critical role in shaping the behaviors of teens as
they face the challenges of growing up and engaging in HPBs (Borkowsky et al., 2002). Close
relationships between parents and their children, such as healthy open communication and
perceived parental support, are especially important during childhood, as children experience many
physical and emotional transitions from childhood to adolescence (Aufseeser et al., 2006).
Children are less likely to report symptoms of depression and are more likely to report high levels
of well-being when they have parent support and parent involvement in their activities (Mounts,
2001). Parental monitoring, combined with parental support, has been demonstrated to be
positively related to higher self-esteem, academic accomplishment and greater academic success in
school (Mounts, 2001). Parental monitoring has also been associated with fewer internalizing
behaviors, such as depression and withdrawal, as well as with externalizing behaviors such as
disturbing others and fighting (Barber al., 1994; Brody et al., 2002).
School Context and HPBs. The current study studied one aspect of school context.
Contrary to expectations, school context did not predict HPBs. This contrasts with numerous

82

studies (Baranowski, 2004; Sallis et al., 2001; Sallis et al., 2003; Verstraete et al., 2006). As school
context in these studies was conceptually defined as physical activity environment characteristics
(see Table 13).
Two key factors may account for these differences. As indicated previously, school context
was defined by children’s reports of the extent to which they perceived their peers as supportive
and by children’s exposure to hostile peer behaviors, such as bullying.
Table 13
Defined School Context
Researcher

Defined School Context

Baranowski, 2004;

Physical activity environmental characteristics in the school
setting have the ability to influence students’ activity level
Perceived importance of school environment is a key
environmental component in explaining physical activity
behavior among youth
School-related factors could predict future health behaviors,
Physical activity environmental characteristics
School with rich facilitated environments
School with rich facilitated environments could be attributed to
several factors which are related to children health-promoting
behaviors
Physical activity environmental characteristics

Fein et al., 2004

Gadin & Hammarstrom, 2002
Sallis et al., 2001
Sallis et al., 2003
Salmon et al., 2007

Verstraete et al., 2006

Peer relations and close friendships play extremely important roles in children’s emotional
development. Children spend most of their daytime hours engaged in academic and leisure
activities with peers and close friends, and these relationships provide a critical source of
emotional support (Prochaaka et al., 2001). Peers influence children’s and adolescents’ diet and
exercise in many positive and negative ways. Positive peer relationships may promote physical
activity by boosting adolescents’ mood and physical self-esteem (Sallis, et al., 2003). Also, close
83

friends are a primary source of companionship for the exercise aspects of a chronic disease
regimen and can support healthy eating habits in children (Verstraete et al., 2006), as perceived
support by classmates and close friends (Sallis, et al., 2003). According to Verstraete et al. (2006),
children who spend time with a close friend in the school context may also prevent feeling
different or left out by peers. HPBs in school-age children are affected by valuing peer friendship
and other relationships; especially peer social support at school (Greca, Bearman, & More, 2002) .
Fein et al. (2004) examined physical environments to explain and promote physical activity, with
home, neighborhood and school as significant domains. Home context was defined by researchers
as how children perceived neighborhood safety. School context was defined as the importance of
the school environment that supports school physical environment.
Gadin and Hammarstrom (2002) conducted a three-year prospective study of children
enrolled in grade six through grade nine. The objective of this study was to identify predictors
among young children for later risky heath behaviors with special focus on factors unique to the
school environment. The investigators assessed the importance of following school related factors,
such as classmate problems (having many friends as wanted, feeling left out of the peer group and
being called rude words), low teacher support, teasing classmates, difficulties sitting still in class
and students’ relation to physical education at school. The results of this study identified numerous
negative health behavioral outcomes such as depression, anxiety, difficulty falling asleep and
feeling sad at school.
Parenting process, neighborhood structure and the development of youths’ externalizing
behaviors were investigated in a longitudinal sample (Beyers et al., 2003). Neighborhood structure
was defined as adolescent perceived neighborhood safety. Parenting process was defined as

84

adolescent perceived their positive involvement with their parents and parental monitoring.
The prevention of unfavorable health behaviors among children become a high priority in
public health. It is important to identify early predictors of HPBs, as established health behaviors
in childhood can be difficult to change later in life. Most of the studies performed on health
behaviors in children have focused only on individual factors without any connection to subjects’
environments. Health promotion will be more effective if it is based on a program that changes the
environment in which people make decisions that affect their HPBs choices.
The second factor affecting this study’s conclusion that school context did not predict
HPBs is the difference in research design between this and similar studies. The current study used
a longitudinal approach to assess the extent to which events that occurred in the 4th, 5th, or 6th grade
predicted HPBs over a three year period. In short, it examined the extent to which distal events in
children’s lives predicted proximal health behaviors, either directly or as mediated by factors distal
to the predictors (media use or affective state).
Children who reported experiencing an unsafe school environment were less likely to
experience support from parents and positive affect. Correlations between HPBs and direct and
indirect influences were as expected: children who experience safe school context, have more
parental support and experience more positive affect were shown to be at lower risk for not
engaging in healthy behaviors.
Parental emotional involvement, parent support and self-esteem were all positively
correlated with health-promoting behaviors. Peer social support, peer hostility and bulling at
school, media use and depression were all negatively correlated with health-promoting behaviors.

85

Although home, school context and health-promoting behaviors have been examined in
several studies, almost all such studies are cross-sectional and involve self-reported data
(Aufseeser et al., 2006; Kremarik, 2000; Sallis et al., 2000; Okun et al., 2002; Greca et al., 2002;
Lohaus et al., 2004).
Cross-sectional design limits the interpretation of study findings to the relationships among
selected variables that occur in the same time context. As all measures are proximal, no change can
be assessed; that is, the descriptive nature of the design does not allow for a strong causal
relationship between study measures.
A longitudinal study found that substance use established at ages 14 to 15 could be
predicted by educational performance and social behavior at school displayed at ages seven to nine
(Hops, Davies, & Lewin, 2000). A study of neighborhood effects on children that includes
repeated measurement of both parenting process and youth outcomes could investigate change
across time in the effects of neighborhood on parenting, as well as change across time independent
effects of parenting (Beyyers et al., 2003). It is difficult to determine whether school and home
context have changed HPBs due to the lack of longitudinal studies. More research is needed about
possible predictors for future health behaviors, in relation to both individual and family oriented
factors and factors related to the school environment, such as classmate relations, teacher support.
The relationship between school context and health-promoting behaviors from this study cannot be
generalized to the population at large. However, the majority of evidence favors a relationship
between school context and health-promoting behaviors. Less peer social support may have a
significant direct negative effect on school-age children. Hence, it is essential to explore how
school context plays a role in health-promoting behaviors among school-age children.

86

In conclusion, only parent variables of home context had direct effects on HPBs. As
mentioned before, most studies of HPBs among school-age children have focused on individual
factors rather than on home and school contexts that are associated with HPBs. Parents could help
shape a positive their children’s HPBs. For school-age children, the main environment is not only
their home but also their school. This current study studied one aspect of school context (peer).
Such an approach could ascertain how multiple factors coalesce in relation to both individual and
family oriented factors and factors related to school context. Researchers should determine which
HPBs to concentrate on for successful HPBs intervention.
Discussion of Research Question 2
Mediational Effects of School and Home. The analysis of question 2 provided evidence
that school contextual influences on HPBs are fully mediated by children’s affective states.
Although both affective state and media use mediated home context influences on children’s
HPBs, home context also has direct effects. These findings extend prior work investigating
mediational influences on children’s HPBs (Chenchob et al., 2013) which indicate that affective
state mediates the relationship between social support and neighborhood context on dietary intake
and physical activity. Neighborhood safety and social support are both significant predictors of
HPBs, but only when mediated through the child’s affective state. Accordingly, individuals are
more likely to engage in HPBs when they are exposed to multiple interacting influences.
Individuals may want to behave in the ways that promote health, but the environmental constraints
may prevent healthy decision choices (Pender et al., 2002).
In the present study, children’s media use influenced HPBs. Media use is a factor
encouraging exposure to both risky behaviors and healthy behaviors. Parents who are likely to also

87

provide poor emotional support in addition to the content of show networks.
Initially, researchers and health professionals focused on the impact of television on
children’s aggressive behavior (Eron, 1982) and weight gain. Dietz and Gortmaker (1985)
concluded that there were positive and perhaps causal connections. Increasingly, such concerns
have shifted beyond children’s exposure to television to include time spent playing video games or
texting on mobile devices. Studies have reported positive effects of computer use on children’s
academic achievement (Jackson et al., 2006; Kulik, 1994; Wenglinsky, 1998) and expansion of
social networks (Jackson et al., 2010), though others have demonstrated no effect at all (Jackson,
2008). Computer use can be socially and psychologically isolating, resulting in feelings of
depression and loneliness (Amichai-Hamburger & Ben-Artzi, 2003). While video games may
enhance visual spatial skills (Subrahmanyam et al., 2000), other evidence suggests that levels of
aggression are higher in children who are heavy users of video games, particularly violent games
(Anderson, Gentile & Buckley, 2009). Teacher ratings of these children’s behavior in school are
more negative when compared with children who play games less frequently (Anderson et al.,
2007).
Children are exposed to an extraordinary amount of information about energy-dense foods
on television (Moore, 2008). Low levels of exercise, high consumption of energy-dense foods and
low affect contribute to children’s weight gain. With rare exceptions, screen time is a time of high
caloric consumption and low physical activity (Jackson et al., 2010), contributing to weight gain
and obesity (AOA, 2005). Finkelstein et al.(2005) note that children watch ten or more food
commercials per hour, most of which are for fast foods and soft drinks, adding to television’s
impact on children’s weight gain. Adams (2006) specifically targets poor eating habits, soft drinks,

88

lack of exercise and high amounts of time spent watching television or using a computer as major
contributors to child obesity. Therefore, examining the role of media use on children’s weight gain
is important.
Previous research has shown complex interactions between the time children spend
watching television, physical activity level and body fat level (Proctor et al., 2003; Biddle et al.,
2004). Even if television does not reduce exercise time, it may still be problematic by increasing
and promoting snack advertisements. To help develop effective interventions to reduce television
viewing and increase physical activity, a consistent correlation between TV viewing and negative
health behaviors needs to be identified (Gorely et al., 2004). TV viewing appears to increase
between-meal snacking and was inconsistently related to children asking for and choosing TVadvertised foods and actual dietary fat intake (Gorely et al., 2004). Also, TV viewing in children is
associated with parental viewing habits.
Study findings indicated that affective state and media use are important mediators of
HPBs in school-age children. A better understanding of factors influencing their HPBs may lead to
more effective health promotion programs aimed at maximizing the health potential of school-age
children. The results of the current study, in combination with those reported by Chenchob et al.
(2013), provide a starting point for additional research designed to understand the complex
relationships that link children’s affective states and media use, as well as peer, parent and
neighborhood influences on children’s HPBs. These factors cooperatively support the processes
that influence individuals to make decisions regarding HPBs. Identification of the
interrelationships and an understanding of the dynamics can facilitate health specific and health
enhancing behaviors.

89

In conclusion, research supports the contention that these variables (meditational effects of
school and home) have an impact on the practice of HPBs among school-age children. It is
interesting to note that this current study is handful that examined internal indicators (affective
state) of well being as a predictor of HPBs. Affective state was both a direct predictor of
engagement HPBs and a significant mediator between external influential factors and children’s
HPBs. Moreover, one such variable is children’s media use. Media use contributed to children’s
HPBs. Studying children’s use of media in relation to HPBs receives strong support from the
extensive studies that implicate exposure to television advertising and similar Web-based message
as detrimental to health in children. The findings are not consistent howeverthe regarding ,
relationships betweenchildren’s media use and sBPH. Further research is warranted in these
areas.
Discussion of Research Question 3
Relationship of Affective State and Media Use. Affective state and media use are
positively related.
In this study, frequent exposuree to media use was associated with higher depression and
lower self-esteem. This finding is consistent with a longitudinal study of youth aged 11 to16 years
of age conducted by Witt et al. (2011) which indicated that self-esteem was negatively associated
with playing videogames and positively associated with general computer use. Children who play
video games more have lower social self-esteem than children who play them less frequently.
Conversely, children who use cell phones frequently have higher social self-esteem than those who
use them less often. Similar findings hold for general self-esteem (Means, 2000).
Jackson et al. (2011) studied the effects of internet use, video game playing and cell phone

90

on children’s weight gain and self-esteem. Although ethno-racial group membership, age and
income levels predict academic achievement and body mass index (BMI), none of the information
technology (IT) indicators do. Conversely, information technology accurately predicts children’s
social self-esteem.
While video games may enhance visual spatial skills (Subrahmanyam et al., 2000), other
evidence suggests that levels of aggression are higher in children who are heavy users of video
games, particularly violent games (Anderson et al., 2009). Teacher ratings of children’s behavior
in school are more negative when compared with children who play games less frequently
(Anderson et al., 2009). An absence of depression could influence physical activity because
physical activity may offer protective, buffering effects on mental health for youth prior to the
onset of emotional problems (Dyck, 2000) by raising levels of self-esteem, body image and selfconcept (Higgins et al., 2003). Indeed, regular exercise participation has been found to be
associated with decreases in state and trait anxiety, depression, and stress (Dunn & Madhukat,
2001).
The results of this study suggest that whether individual children will be responsive to
attempts to instill HPBs will depend on the extent to which prevention efforts are capable of
improving affective state (self-esteem and depression) so that children can be receptive to HPBs
messages. Children who get their health behaviors and live in supportive environments that enable
proactive health behaviors tend to be more active.
Discussion of Research Question 4
Sex Differences. The fourth aim was to determine whether there are sex differences in the
relationships among variables.

91

Sex differences in self–esteem have been noted with higher self–esteem commonly
identified in boys and decreasing self–esteem in females as they enter adolescence (Hendricks et
al., 2001). Across culture and gender, abilities were valued as promoting self–esteem (Guinn et al.,
2000; Hendrickset et al., 2000). Gadin and Hammarstrom (2002) conducted a three-year
prospective study in grade six and grade nine. The objective of their study was to find predictors
among young children for later risky heath behaviors with special focus on school-related factors.
The results indicated that school-related factors could predict future health behaviors, especially in
relation to low physical activity among girls. Low physical activity among girls could best be
predicted by school-related factors such as classmate problems, teasing others, rowdiness in the
class and having difficulties sitting still.
Sex differences are often discussed relative to media use. Research indicates that males use
computers to search the internet for information and to play videogames, while females use
computers for communication such as e-mail (Jackson, Ervin, Gardner, & Schmitt, 2001).
However, Ohannessian (2009) found no gender differences in how adolescents use computers, but
did find that anxiety levels between males and females differed and also that males used
technology as an escape. Pierce (2009) found that girls tend to be more self-conscious in social
settings, often relying on technology such as texting, instant messaging and social websites to
communicate with their peers. Using technology for social communication may make individuals
less self-conscious than in face-to-face communication. Willoughby (2008) used a two-wave study
to discover changes in technology use during high school; boys maintained the same general
computer use whereas girls’ use declined from over the same period.
Consistent with previous research (Gross et al., 2002), gender differences in technology use

92

were observed in this research. Girls reported less videogame playing and more communication
technology use than boys (Papacharissi & Rubin, 2000). These gender differences are relatively
stable over time at least between the ages of 12 and 15.
In conclusion, this current study examined media use and affective state as mediators
between internal, external factors and HPBs. For both boys and girls, affective state completely
mediated the relationship between school context and HPBs, though affective state and media use
only partially mediated the effects of home context and HPBs. The finding from the current study
provided a foundation for the development of appropriate intervention strategies for boys and girls.
Success of intervention efforts is dependent upon evidence that linked to sex differences. The
causal variables (media use) influencing HPBs are markedly sex differences.
Pender’s Conceptual Model of HPM Revisited
HPBs are vital to maintaining and improving people’s health. Pender’s HPM (2006) is a
helpful model to enhance understanding of human behaviors because the model incorporates
internal and external factors that influence human behaviors. The findings of the current study
support Pender’s HPM (2006) that individual characteristics and experiences, behavior-specific
cognitions and affect are related to explain HPBs among school-age children.
The findings of this study support some aspects of Pender’s HPM model. For example, data
showed that home context, including parent emotional support, parent involvement and closeness
directly influenced children’s HPBs. However, the data did not support relationships between
school context and HPBs. Further research is required to clarify and understand these associations.
Such understandings provide a basis for re-constructing Pender’s HPM and to incorporate
more dynamic systems frameworks for generating multiple pathways to health-promoting

93

behaviors. For example, when Pender’s HPM was developed, children had little exposure to
media, including the internet, with respect to HPBs. Since then, children’s exposure to and use of
such media has increased. Media use cooperatively supports the processes that influence
individuals to make decisions and participate in health-promoting behaviors. Another example, the
prior study involving the longitudinal sample, Chenchob et al. (2013) found that children’s
affective state mediated the relationship between indicators of social support and perceived
neighborhood safety and children’s HPBs. Identification of the interrelationships and an
understanding of the dynamics that facilitate health specific behaviors provide insight into health
enhancing behaviors, making the model useful to researchers to further study the improvement of
health care in school-age children.
Study Limitations
The study was limited by the particular sample characteristics. Although the sample was
representative of the population, it was not representative of the various ethnic groups that
comprised the minority population of the United States. Most participants were Caucasian
American boys and girls from one geographical region; factors that limit the applicability of the
results in other populations.
The major disadvantage of the secondary data analysis was that the data was originally
obtained for another purpose, so the researcher using secondary analysis infrequently knows about
the data collection procedures. As the researcher has worked closely with the original investigators
and has completed one study from the current data set, the effect of this limitation was minimized.
An additional disadvantage to secondary analysis is that data may be out-of-date, or the quality of
the data may be so large that statistical results may be significant but the results are less

94

meaningful (Caslte, 2003). Calculating effect sizes and adjusting for sample size by increasing
significance levels for accepting findings as significant are often used to counter this issue.
This study is limited by the method used to measure engagement in HPBs. Using selfreports of children’s engagement in physical activity and dietary intake does not provide the most
accurate assessment of children’s actual behaviors. However, self-report is appropriate to utilize
evaluation an internal self-perception on HPBs.
These findings may reflect a response bias that parents who are interested in health
promotion issues may be more likely to grant permission for their children to participate in such
studies.
Strength of the Study
Strengths of the study include its longitudinal design, the large region of school-age
children, a range of theoretically-derived predictors from Pender’s HPM and the fact that one
attempt at model testing involving Pender’s model was successful (Chenchob et al., 2013).
Implication for Nursing Practice
Nursing intervention approaches should be multifaceted, focusing on enhancing selfesteem and reducing depression, while simultaneously working to build support systems that
effectively enhance HPBs in children. These results can inform health care providers and
researchers interested in enhancing factors for promoting health behaviors for future interventions.
Nursing interventions should emphasize the importance of HPBs among school–age children.
Enhancing the teaching and strengthening the delivery of health knowledge pertinent to selfesteem and social support could enrich the HPBs for school-age children. Nurses and other health
care professionals can encourage and assist school-age children to practice healthy behaviors with

95

emphasis on the impact and control children have upon their own health.
Nurses should direct their conversation about HPBs directly to adolescents and young
adults, clarify any misinterpretations and encourage them to actively participate in decisions that
will ultimately affect their health. Studies are needed to examine interventions that will be most
effective in helping adolescent and young adults both maintain and engage in HPBs (Smith &
Bashore, 2006). Thus, nurses can use the Health Promotion Model (HPM) to intervene in
children’s HPBs, but the HPM needs to modify the appropriate contact with children’s context
both at home and school.
In nursing practice, nurses can use HPBs as a framework to provide a coherent and
organized framework for intervening with clients to increase HPBs. For example, nurses can
develop interventions that address school context, home context, affective state and media use
relevant to a particular health behavior. In addition, understanding participants’ knowledge of risk
factors and health promoting strategies is crucial because school-age children may be prevented or
delayed from risky health behaviors if they practice appropriate risk factor modifications.
Implication for Research
HPBs in school-age children require long-term management and can result in changes over
time Thus, findings from this cross-sequential study could serve as a baseline for further
longitudinal studies of HPBs and factors that influence HPBs in school-age children. The use of
longitudinal study designs would be helpful in identifying predictors of HPBs over time and
ascertaining relationships among factors influencing HPBs. Future research examining HPBs in
children should identify changes over time in media use and HPBs by utilizing a growth curve
model.

96

This study examined the relationships between school context, home context, affective
state, media use and HPBs in school-age children. In light of the epidemic rates of obesity and
obesity related health problems in children, research in children’s health is growing. Future
research examining health-promoting behaviors in children should identify additional external and
internal factors related to how children make decisions about dietary intake and physical activity.
Also, examining these processes in populations of children who are at particularly high risk for
obesity is also recommended for nurses, other health care providers and researchers.
Researchers should consider using the significant variables from this study to guide their
development of future interventions to promote HPBs for school-age children. For instance,
researchers may develop an intervention which promotes physical activity and social support from
parents and peers for school-age children. Additionally, research to develop interventions which
enhance self-esteem and decreases depression is highly recommended.
Future qualitative research studies are needed to explore children’s subjective perception of
HPBs, as well as internal and external factors influencing HPBs in school-age children. Published
research may be useful in identifying variables that enhance HPBs of children who have specific
risky health behaviors.
Implication for Policy
The emphasis in the health care system is shifting from the treatment of illness to the
prevention of disease, and ultimately to the promotion of health. New strategies for promoting
health efforts have shifted away from individual-level approaches toward population-based
approaches that address socio-cultural, political, economic and physical environmental factors
(Yancey et al., 2004). This demonstrates the important of leadership within local government

97

agencies and communities to set priorities and direct local resources toward reducing risky health
behaviors.
Public policies addressing health promotion must be worked in complex environments of
influence in children’s lives (Mcgininis, Russo, & Knickman, 2002). Initiatives for healthier
lifestyles, environments and social conditions should be built into these policies. Further, public
policy interventions are available to improve population health. These can take many forms, from
advertising the identity of restaurants in violation of food-safety protocols to grants that encourage
communities to build bike paths. The surface transportation policy project to set aside highway
trust fund resources for community initiatives for side walking trails, and bike paths could be one
of our most important steps to better health (Pender et al., 2002).
A research agenda should focus on the relationships between social factors associated with
poor health outcomes and the mechanisms that lead to poor health. In HPBs, the highest research
priority may be to better understand how social marketing and behavior changing interventions can
be designed and implemented to work at the population level. Ultimately, success at engaging
HPBs opportunities in school depend on the reliability of the ways these complex interactions. The
policy plans encouraged walking, bicycling, nutrition information campaigns for children (Adler &
Newman, 2002; Mcgininis et al., 2002).
The main question facing policy makers is what if anything can be done to promote
children’s HPBs. Because school context and home context is changeable and provider support
appears to be a factor, incentivizing or holding providers accountable for children’s adoption of
HPBs is a possible policy direction. In particular, some models of health care are more amendable
to supporting children’s HPBs. For example, in the Pender’s HPM, where children behaviors is the

98

focus and where a functioning health care team provides coordinated care, customizing care to
support children’s HPBs is possible.
Implication for Nursing Practice in Thailand
The results of this study suggest that counseling and health promotion programs should be
tailored to suit school-age children. These findings will increase the understanding of the
relationship among factors that may impact children’s HPBs. Teachers and staffs need to
understand effective strategies for enhancing HPBs, allowing them to provide support and
encouragement to school-age children. However, most studies investigating health promotion to
date have been conducted in Western countries. Western cultural values are different from the
cultural values held by the Thai population. Diversity in health-promoting behavior across
cultures could assist nurses in providing more culturally appropriate interventions to Thai
children. Thus, cross-cultural studies of health-promoting behaviors would provide valuable
information about cultural differences that may lead to diverse variations in beliefs concerning the
benefits of HPBs. Further research is required to clarify such associations.
In addition, health promotion must move beyond individuals, to families and their
communities. Identification of the factors that predict both external and internal positive health
outcomes in children is valuable. Additionally, health-promoting behavior and related variables are
important for managing health.
There are potential sex differences with respect to influences of peers and parents on
children’s health behaviors. To promote physical activity and prevent overweight or obesity among
Thai children and pre-adolescents, providers must recognize that children’s contexts and
meditational effects of school and home are related to dietary intake and physical activity.

99

Achieving and maintaining one’s own health is the overall benefit of social marketing
communication between children, their parents, teachers, peers, community organization leaders,
community stakeholders and healthcare providers.
In Thailand, new community practice guidelines offer additional evidence-based
recommendations for a wide array of school-, worksite-, and community-based programs and
public policies to improve dietary intake and physical activity levels for children and reduce
children’ risky behaviors (Glanz, Rimer, & Viswanath, 2008). The present study may also have
implications for interventions such as educational workshops and counseling for individuals whom
are at risk for developing health problems, or those simply desiring more information about health
and wellness.
Interventions are needed to promote HPBs among school-age children. Researchers should
consider using the significant relationships identified in this study to guide their future intervention
development to promote HPBs in school-age children. New strategies for promoting health efforts
have shifted away from individual-level approaches toward systemic approaches that address
socio-cultural, political, economic and physical environmental factors (Yancey et al, 2004). This
demonstrates the need for leadership within school and communities to set priorities and direct
local resources toward disease risk reduction.
There is a dearth of studies exploring HPBs from the perspective of school-age children,
and, consequently, little is known about these phenomena in Thailand. Examining these processes
in populations of children who are at particularly high risk for obesity is also recommended.
Likewise, HPBs and factors that are related to HPBs in school-age children have not been clearly
identified. It is noted that there is a need for empirical studies of the HPBs and factors related to
100

HPBs in this age group. Factors influencing HPBs in this population should be further clarified.
Furthermore, increased use of objective measurement of some of variables such as physical
activity would help to explore the extent to which children engage in HPBs.
A study in Thailand showed obesity is increasing significantly in school- aged children in
Thailand (Thai Health Promotion Foundation, 2013).To improve this condition in school-aged
children, the Department of Health, in collaboration with the Thai Health Promotion Foundation,
has launched a project to reduce obesity in school-age children by promoting balanced diet and
lifestyle. However, this alone may not lower the rate of obesity in this population, as support from
parents, teachers and peers, along with mediational effects of school and home, are essential to
maintain the program.
To promote healthy lifestyles are recommended to use policy tools. In particular, groups
are focusing on the role of healthy food and exercise in reducing obesity and preventable chronic
diseases in children. Suggestions include providing healthier food at home and in schools;
improving the clarity of nutrition labels eliminating agriculture tax subsidies for unhealthy
products; taxing unhealthy products and promoting physical fitness. School lunches should be part
of the learning curriculum, and not sold for profit. Unhealthy foods should be removed from
institutions such as schools. Thailand needs to treat obesity as an epidemic taking over the nation.
Public policy can promote healthier lifestyles by promoting the practice of eating
unprocessed foods, healthy nutrition beyond the standard food pyramid, exercise in homes and
schools and the concept that children are responsible for their health. Financial incentives for
healthy behaviors and for the use of proven prevention methods should be provided to local
governments.

101

Summary
This chapter has presented a summary and discussion of the study findings. Additionally,
this chapter included the implications of this study and recommendations regarding three areas:
nursing practice, research and policy. Overall, the study has added to the knowledge of HPBs and
factors influencing HPBs in school-age children. The study findings support the existence of
relationships between school context, home context, affective state, media use and HPB in schoolage children. From those findings, interventions may be designed to help improve HPBs for this
population.

102

APPENDICES

103

APPENDIX A: COVARIANCE MATRIX
Table 14
Covariance Matrix

HPBs
HPBs
AFFECT
MEDIA
SCH_C
HOME_C

1.833
-0.238
-0.137
- 0.158
0.304

AFFECT

MEDIA

SCH_C

0.996
0.116
0.478
-0.357

0.967
0.041
-0.114

0.998
-0.359

104

HOME_C

0.998

APPENDIX B: INITIAL MODEL
Wave 1
School Context
=Peer hostility and bulling
=Lack of peer social support
-0.013

0.042
0.405***

Wave 2
Media Use
-Television
-Video Games
-Computer

-0.070**
Wave 3
HPBs
-Dietary intake
-Physical activity

0.091**
-0.101***

Wave Use
-Computer 2
Affective State
-self-esteem
-Lack of
depression
-0.212***

-0.099**

0.177***

Wave 1
Home Context
=Parent Emotional support
and closeness
=Parent involvement

Fiqure 6. Initial Model for school context, home context, affective state, and media use with
standardized coefficients
*p<0.05; **p<.01; ***p<.001; solid line represent =p<.0.05.

105

APPENDIX C: MSU IRB APPROVAL LETTER

106

REFERENCES

107

REFERENCES

Ackard, D. M., Neumark-Sztain, D., Story M., Perry, C. (2003). Overeating among adolescents:
prevalence and associate with weigh related characteristics and psychological health.
Pediatrics, 111, 67-74.
Adams, C. K. (2006). Childhood and adolescent obesity: A winning way to fight a “losing battle.”
Northeast Florida Medicine, 57, 47-48.
American Obesity Association (2005, May). Fact sheet: Obesity in youth. Retrieved from
http//www.obesity.org/subs/fastfacts/obesity_youth.shtml/
Amichai-Hamburger, T, & Ben-Artzi, E. (2003). Loneliness and internet use. Computers in
Human Behavior, 19,71-80.
Anderson, C. A., Douglas A. Gentile, D. A. & Buckley, K. E. (2009). Violent Video Game Effects
on Children and Adolescents: Theory, Research, and Policy. J Youth Adolescence, 38, 483–
485. doi 10.1007/s10964-008-9344-1
Aufseeser, D., Jekielek, S., & Brown, B. (2006). The family Environment and adolescent wellbeing: Exposure to positive and negative family influences. From Children Trends and the
National Adolescent Health Infromation Center. Retrieved from
http://www.childtrends.org/Files/Child_Trends-2006_06_01_FR_FamilyEnvironmen.pdf.
Bandura, A. (1977). Social Learning Theory. New York: General Learning Press.
Baranowski, T. (2004). Mediating variable framework in physical activity interventions. How we
Are doing? How might we do better. Am J Prev Med, 15, 266-297.
Bee, H. L. (2000). The Journey of Adulthood (4th ed.). Upper Saddle River, NJ: Prentice Hall.
Bentler, P. M. (1992). On the fit of models to covariances and methodology to the Bulletin.
Psychological Bulletin, 112(3), 400-404.
Berg, T.D. et at. (2009). The health and well being of adolescents in the United States. Journal of
Adolescent Health. 45(6), 2-3.
Beyyers, M. J., Bates, E. J., Pettit, S. G., & Dodge, A. K., (2003). Neighborhood Structure,
Parenting Process and the Development of Youths’ Externalizing Behaviors: A Multilevel
Analysis. American Journal of Community Psychology, 30 (1/2), 35-53.

108

Biddle, S. J., Gorely, T., Simon J Marshall, S. J., Ian Murdey, I., Noel Cameron, N.
(2004)Physical activity and sedentary behaviours in youth: issues and controversies,
Journal of the Royal Society for the Promotion of Health, 124 (1), 29-33.
Borkowsky, J., Ramey, S., & Bristol-Power, M. (Eds.). (2002). Parenting and the child’s world:
Influences on academic, intellectual, and social-emotional development. Mahwah, NJ:
Lawrence Erlbaum as cited in Hair, E., Moore, K., Garrett, S., Kinukawa, A., Lippman, L.
& Michelson, E. (2005). The parent-adolescent relationship scale. In K. Moore & L.
Lippman (Eds.) What do children need to flourish (pp. 183-202). New York: Springer
Science
Brody, G. H., V. M. Murry, et al. (2002). Longitudinal pathways to competence and psychological
adjustment among African American children living in rural single-parent households.
Child Development, 73, 1505-1516.
Buu, A., Mansour, M., Wang, J., Refior, S.K., Fitzgerald, H.E., & Zucker, R.A. (2007).
Alcoholism effects on social migration and neighborhood effects on alcoholism over the
course of 12 years. Alcoholism: Clinical and ExperimentalResearch, 31, 1545-1551.
Buu, A., DiPiazza, C., Wang, J., Puttler, L. I., Fitzgerald, H. E. & Zucker, R. A. (2009).
Parent, family, and neighborhood effects on the development of child substance use and
other psychopathology from preschool to the start of adulthood. Journal of Studies on
Alcohol and Drugs,70, 489-498.
Byrne, B. M. (2010). Structural equation modeling with AMOS: basic concepts, application, and
programming. (2nd ed.). New York: Taylor and Francis Group, LLC.
Callaghan, D. (2006). Basic conditioning factors’ influences on adolescents’ healthy behaviors,
self-efficacy, and self care. Issues in Comparative Pediatric Nursing, 29, 191-204.
Casper, R. C. (1998). Depression and eating disorders. Depression Anxiety, 8 (96-104).
Centers for Disease Control, (2012). The Youth Risk Behavior Surveillance System (YRBSS).
Retrieved from http://www.cdc.gov/healthyyouth/yrbs/index.htm.
Charoneying, W. (2005). Factors affecting exercise behavior of upper secondary level
school students in school under the Office of Prachinburi Educational Service
Area. Master's thesis, Burapha University, Thailand.
Chenchob, P., Barnes, J., Fitzgerald, H. E., KungSook, L., & Pandonu, G. (2011, November).
st
Health-Promoting Behaviors in School- Age Children. Paper presented at the 1
International Symposium on Health Research & Development and the 3
Regional Conference on Public Health, Bali Indonesia.

rd

Western Pacific

Chenchob, P., Barnes, J. , Fitzgerald, H. E., KungSook, L., & Pandonu, G. (2012, April).
109

Predictors of Health-Promoting Behaviors in School- Age Children. Paper presented at the
2012 Midwest Nursing Research (MNRS) Annual Research Conference in Dearborn,
Michigan.
Chenchob, P., Barnes-Najor, J. , Fitzgerald, H. E., KungSook, L., & Pandonu, G. (2013).
Affective State as a Mediator of Health-Promoting Behaviors in School- Age Children.
Manuscripts submitted for publication.
Collins, W. A., Maccoby, E. E., Steinberg, L., Hetherington, E. M., & Bornstein, M. (2000).
Contemporary research on parenting: The case for nature and nature. American
Psychologist, 55, 218-232.
Crawford, P. B., Story, M., Wang, M. C., Ritchie, L. D., & Sabry, Z. I. (2001). Ethnic issues in the
epidemiology of childhood obesity. Pediatric Clinics of North America, 48(4), 855–878.
Croker, P., Sabiston, C., Forrestor, S., Kowalski, N., & McDonough, M. (2003). Predicting
Change in Physical Activity, Dietary Restraint, and Physique Anxiety in Adolescent Girls.
Canadian Journal of Public Health. 94(5), 332-336.
Cranford, J.A., Zucker, R.A., Jester, J. M., Puttler, L.I., Fitzgerald, H. E., (2010). Parental alcohol
involvement and adolescent alcohol expectancies predict alcohol involvement in male
adolescents. Psychol Addict Behav. 24(3), 386-96.
Davies, H.D., Fitzgeral, H., & Mousouli, V. (2008). Obesity in Childhood and Adolescence:
Understanding development and prevention. Westport, CT: Praeger Publishers.
Dietz, W. H., & Gortmaker, S. L. (2001). Preventing Obesity In Children And Adolescents.
Annual Review of Public Health, 22, 337-353. doi: 10.1146/annurev.publhealth.22.1.337
Dunn, A.L., & Madhukar, H.T. (2001). O’Neal HA. Physical activity dose-response effects on
Outcomes of depression and anxiety. Med Sci Sport Exerc. 33(6), S587-97.
Dancan, S.C., Dancan, T. E., Strcker, L. A. (2005). Sources and types of social support in youth
physical activity. Health Psychology 2000;24, 3-10.
Dyck, L. (2000). Psychosocial health and well-being. In: Reid C., Dyck, L., McKay, H. Frisby, W.
(Eds.). The Health Benefits of Physical for Girls and Women. Vancouver, BC: Center for
Excellence in Women’s Health, 23-59.
Eaton, Kann, Kinchen, Shanklin, Ross, Hawkins, & Harris et. al., (2012). “Youth Risk Behavior
Surveillance -- United States in Centers for Disease Control and Prevention (CDC):
Retrieved from http://www.cdc.gov/healthyyouth/yrbs/index.htm
Gadin, K. G., & Hammarstrom, A. (2002). Can school-related factors predict future health
behavior among young adolescents? Public Health,116, 22-29.
110

Garcia, A., Pender, N. J., Antonakos, C. L., & Ronis, D. L. (1998). Changes in physical activity
beliefs and behaviors of boys and girls across the transition to junior high school. Journal
of Adolescent Health, 22, 394-402.
Garcia,A. W., Norton Broda,M. A., Frenn,M., Coviak, C., Pender, N. J., & Ronis, D. L. (1995).
Gender and developmental differences in exercise beliefs among youth and prediction of
their exercise behavior. Journal of School Health, 65, 213-219.
Goodman, E., McEwen, B. S., Dolan, L. M., Schafer-Kalkhoff, T., Adler, N. E. (2005). Social
disadvantage and adolescent stress. The Journal of Adolescent Health: Official Publication
of the Society for Adolescent Medicine, 37, 484-92
Goodman, E., Daniels, S. R,, Meigs, J. B., Dolan, L. M. (2007). Instability in the diagnosis of
metabolic syndrome in adolescents. Circulation, 115, 2316-22
Gorely, T., Marshall, S. J., Stuart J H Biddle S. J. (2004) Couch kids: correlates of television
viewing among youth., 152-63. International Journal of Behavioral Medicine, 11(3), 152163.
Guinn B, Vincent V, Semper T, & Jorgensen L, (2000). Activity involvement, goal perspective,
and self-esteem among Mexican American adolescents. Research Quarterly for Exercise
Sport (RQES), 71(3), 308 - 311.
Hendricks et al., (2011). Why Does Gender Matter? Counteracting Stereotypes with Young
Children Dimensions of Early Childhood, Dimensions of Early Childhood, 39(3), 1120.
Higgins, J. W., Gaul, C., Gibbons, S., & Van Gyn, G. (2003). Factors influencing physical activity
levels among Canadian youth. Canadian Journal of Public Health, 94(1), 45-51.
Host, K. V. et al. (2007). A systematic review of environmental correlates of obesity-related
dietary behaviors in youth. Health Educ Res, 22(2), 203-226. doi: 10.1093/her/cyl069.
Eisenberg, M. E., Olson, R. E., Neumark-Sztainer, D. (2004). Correlations between family meals
and psychosocial well-being among adolescents. Arch Pediatr Adolesc Med, 158 (8), 792–
796.
Fein, A. J., Ronald, C., Plotnikoff, T., Wild, C., & Spence J. C. (2004). Perceived Environment
and Physical Activity in Youth. International Journal of Behavioral Medicine, 11(2), 135142.
Field, T., Diego, M., & Sanders, C.E. (2001). Exercise is positively related to
adolescents’relationships and academics. Adolescent,36(141), 105-110.
Finkelstein, E. A., Rhum, C. J., & Kosa, K. M. (2005). Economic causes and consequences of
obesity. Annual Review of Public Health, 20, 239-257.
111

Fitzgerald, H. E., Puttler, L. I., Refior, S., & Zucker, R. A. (2007). Family response to children
and alcohol. Alcohol Treatment Quarterly, 25, 11-25.
Fitzgerald, H. E., Zucker, R. A., & Freeark, K. (Series Eds). (2006). Crisis in Youth Mental
Health. Westport, CT: Praeger Press.
Fitzgerald, H. E., Lester, B. M., & Zuckerman, B. (Eds) (2006). Crisis in Youth Mental Health Vol
1: Disorders of childhood. Westport, CT: Praeger Press.
Fitzgeral, H., & Mousouli,V. (2008). Childhood overweight and academic achievement. Westport,
CT: Praeger Publishers.
Frederick-Recascino, C.M. (2002). Self-Determination Theory and Participation Motivation
Research in the Sport and Exercise Domain. In Deci, E.L. & Ryan, R.M. (Eds). Handbook
of Self-Determination Research. Rochester, New York: The University of Rochester Press.
Fulkerson, .A. J, Sherwood, E. N., Perry, L. C., Neumark-Sztainer, D., & Story, M. (2003).
Depressive Symptoms and Adolescent Eating and Health Behaviors: a multifaceted view in
a population-based sample. Preventive Medicine. 38, 865-875.
Glanz K., Rimer B.K., Viswanath K. (2008). Health Behavior and Health Education: Theory,
Research, and Practice (4th ed). San Francisco: Jossey-Bass.
Goodman, E., Dolan, L. M., Morris, J. A., & Daniels. (2005). Factor Analysis of Clustered
Cardiovascular Risks in Adolescence : Obesity Is the Predominant Correlate of Risk
Among Youth. Journal of American Heart Assiciation. 111, 1970-1977. doi:
10.1161/01.CIR.0000161957.34198.2B.
Gorely, T., Simon, J., Marshall, Stuart, .J. H ., & Biddle (2004). Couch kids: correlates of
television viewing among youth. International Journal of Behavioral Medicine, 11(3), 152163.
Gross, E. (2004). Adolescent Internet use: What we expect, what teens report. Applied
Developmental Psychology, 25, 633–649.
Gross, E., Juvonen, J., & Gable, S. (2002). Internet use and well-being in adolescence.
Journal of Social Issues, 58, 75–90.
Higgins, W. J., Gaul, C., Gibbons, S., Gyn V.G., (2003). Factors Influencing Physical Activity
Levels Among Canadian Youth. Canadian Journal of Public Health,94, 45-51.
Haug, E., Torsheim, T., & Samdal, O., (2008). Physical environmental characteristics and
individual interests as correlates of physical activity in Norwegian secondary schools:
The health behavior in school-aged children study. International Journal of Behavioral
Nutrition and Physical Activity 2008, 5:47 doi:10.1186/1479-5868-5-47
112

Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55,
doi:10.1080/10705519909540118
Jackson, L.A. (2008). Adolescents and the Internet. In D. Romer & P. Jamieson (eds). The
changing portrayal of American youth in popular media. Annenberg Public Policy Zcenter
at the University of Pennsylvania. New York: Oxford University Press.
Jackson, L., Ervin, K., Gardner, P., & Schmitt, N. (2001). Gender and the Internet: Women
communicating and men searching. Sex Roles, 44, 363–380.
Jackson, J. A., Fitzgerald, H. E., von Eye, A., Zhao, Y., & Whitt, E. A. (2010). The digital divides
in the U.S.L Access, broadband, and nature of Internet use. In E. Ferro, Y. K. Dwivedi, J.
R., Gil-Garcia, & M. D. Williams (eds). Handbook of research on overcoming digital
divides: Constructing an equitable and competitive information society. (Vol. 1, pp 223238). Hershey, PA: IGI Global.
Jackson, L. A., von Eye, A., Biocca, F. A., Barbatsis, G., Zhao, Y. & Fitzgerald, H. E. (2006).
Does home Internet use influence the academic performance of low-income children?
Findings from the HomeNetToo project. Special Section on Children, Adolescents, and the
Internet, Developmental Psychology, 42, 429-435.
Jackson, L. A., von Eye, A., Fitzgerald, H. E., Witt, E. A., & Zhao, Y. (2011). Internet use,
videogame playing and cell phone use as predictors of children’s body mass index (BMI),
body weight, academic performance, and social and overall self-esteem. Computers in
Human Behavior, 27, 599-604.
Jenkins , S., & Horner, S. D. (2005). Barriers that influence eating behaviors in adolescents. J
Pediatr Nurs, 20(4), 258-67.
Kerr, D. C. R. & Capaldi, D. M., Owen, L. D., Wiesner, M., & Pears, K. C. (in press). Changes in
at risk American men's crime and substance use trajectories following fatherhood. Journal of
Marriage and Family.
Kline, R. B. (2011). Principles and practice of structural equation modeling. (3rd ed.). New York:
The Guilford Press.
Kosulwat V. (2002). The nutrition and health transition in Thailand. Institute of Nutrition, Mahidol
University, Phutthamonthon, Salaya, Nakhon Pathom, Thailand. Public Health Nutr. 183189.
Kremarik, F. A. (2000). Family affair: Children’s participation in sport. Can Social Trends.
Statistics Canada, 11, 20-24.
Kulik, J. A. (1994). Meta-analytic studies of findings on computer-based instruction. Technology
assessment in education and raining. Hillsdale, NJ: Erlbaum.
113

Lowry, R., Wechsler, H., Galuska, A. D., Fulton, E. J., & Kann, L. (2002). Television Viewing and
its Associations with Overweight, Sedentary Lifestyle, and Insufficient Consumption of
Fruits and Vegetables Among US high School Students: Difference by Race, Ethnicity, and
Gender. Journal of School Health. 72(10), 413-421.
Loukas, A, Roalson, L.A. (2006). Family environment, effortful control, and adjustment among
European American and Latino early adolescents. Journal of Early Adolescence, 26:432–
455.
MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and determination
of sample size for covariance structure modeling. Psychological Methods, 1(2), 130-149.
Mean, M.Clemens M. H. Hosman , M. H., Herman P. Schaalma , H. P. (2004) . Self-esteem in a
broad-spectrum approach for mental health promotion. Medicine Health Education
Research, 19 (4), 357-372. doi: 10.1093/her/cyg041
Megel, M. E., Wade, F., Hawkins, P., Norton, J. (1994). Health promotion, self-esteem, and weight
among female college freshmen. Health Values: The Journal of Health Behavior,
Education & Promotion, 18(4), 10-19.
Milligan, R. A. K., Burke, V., Beilin, L. J., et al. (1997). Health-related behaviors and psychosocial characteristics of 18 year-old Australians. Social Science Medication. 45(10),
1549-1562.
Moore, E. S. (2008). Food marketing goes online: A content analysis of web sites for children. In
H. E. Fitzgerald & V. Mousouli (eds). Obesity in childhood and adolescence (Vol 2):
Understanding development and prevention (pp 93-115). Westport, CT: Praeger.
Moore J, Harré N. (2007). Eating and activity: the importance of family and environment. Health
Promot J Austr, 18(2), 143-8.
Mounts, N. (2001). Young adolescents' perceptions of parental management of peer relationships.
Journal of Early Adolescence, 21(1), 92-122.
Melissa C. Nelson, M. C. & Penny Gordon-Larsen, P. (2006). Division of Epidemiology and
Community Health, Pediatrics . 117, 1281 -1290. doi: 10.1542/peds.2005-1692
Neumark-Sztainer, D., Story, M., Hannan, P. J., Beuhring, T., & Resnick, M. D. (2000).
Disordered eating among adolescents: associations with sexual/physical abuse and other
familial/psychosocial factors. Int J Eat Disorder, 51, 249-258.
Neumark-Sztainer, D., Wall, M., Story, M., & Fulkerson, J. A. (2004). “Are Family Meal Patterns
Associated with Disordered Eating Behaviors Among Adolescents?” Journal of Adolescent
Health, 35(5), 350-359.
114

Nies, M.A., Chruscial, H.L. & Hepwort, J.T. (2003). An Intervention to Promote Walking in
Sedentary Women in the Community. Am J Health behave, 27(5), 524-535.
Norman, G. J., MF Zabinski, M. F., Adams, M. A. Rosenberg, M. A., AL Yaroch, A. L., Atienza,
A. A. (2007). A review of eHealth interventions for physical activity and dietary behavior
change. American journal of preventive medicine, 33(4), 336–345.
Ogden, Cl. L., Carroll, M. D., Curtin, L. R., McDonnell, M. A., Tabak, C. J., & Flegal, K. M.
(2006). Prevalence of overweight and obesity in the United States, 1999-2004. Journal of
the American Medical Association, 295(13), 1549-1555.
Ogilvie et al. (2007). Interventions to promote walking: systematic review. BMJ, 334, 1-10. doi:
http://dx.doi.org/10.1136/bmj.39198.722720
Ohannessian, C. (2009). Media use and adolescent psychological adjustment: An examination of
gender differences. Journal of Child Family Studies, 18, 582–593.
Okun, M. A., Karoly, P., & Lutz, R. (2002). Clarifying the contribution of subjective norm to
predicting leisure time exercise. Am J Health Behav,26(4), 296-305.
Oliver. V. M., Sanchez, A. M., Suarez, P. E., Velez, A. H. & Arroyo, C. Y. (2002). Brest cancer
health promotion model for older Puerto Rican women: Result of a pilot program. Health
Promotion International, 17, 3-11.
Papacharissi, Z., & Rubin, A. M. (2000). Predictors of Internet use. Journal of Broadcasting and
Electronic Media, 44, 175–196.
Parker, J., & Benson, M. (2004). Parent-adolescent relations and adolescent functioning: Selfesteem, substance abuse, and delinquency. Adolescence, 39 (155), 519-530.
Patrick, K., Norman, G., Calfas, K.J., Sallis, J.F., Zabinski, M. F., Rupp, J., & Cella, J. (2004).
Diet, Physical Activity, and Sedentary Behaviors as Risk Factors for Overweight in
Adolescence. Arch Pediatr Adolesc Med, 158, 385-390.
Patrick, H., Nicklas, T. A. (2007). A Review of Family and Social Determinants of
Children’s Eating Patterns and Diet Quality. J Am Coll Nutr, 24(2), 83-92.
Pearson, N., Ball, K., & Crawford, D. (2011). Predictors of changes in adolescents’ consumtion of
fruits, vegetables and energy-dense snacks. British Journal of Nutrition, 105, 795-803.
nd

Pender, N. J. (1987). Health promotion in nursing practice (2
Lange.

rd

Pender, N. J. (1996). Health promotion in nursing practice (3
Lange.
115

ed.). Norwalk, CT: Appleton &

ed.). Stamford, CT: Appleton &

th

Pender, N.J., Murdaugh, C.L., & Parsons, M.A. (2002). Health Promotion in Nursing Practice (4
ed). Prentice-Hall: New Jersey.

th

Pender, N.J., Murdaugh, C.L., & Parsons, M.A. (2006). Health Promotion in Nursing Practice (5
ed). Prentice-Hall: New Jersey.

th

Pender, N.J., Murdaugh, C.L., & Parsons, M.A. (2011). Health Promotion in Nursing Practice (6
ed). Prentice-Hall: New Jersey.

Pesa, J. (1999). Psychological factors associated with dieting behaviors among female adolescents.
Journal of School Health, 69, 196-201.
Pierce, T. (2009). Social anxiety and technology: Face-to-face communication versus
technological communication among teens. Computers in Human Behavior, 25,
1367–1372.
Prochaska, J. J., Rodgers, V. M., & Sallis, F. J. (2002). Association of Parent and Peer Support
with Adolescent Physical Activity. Research Quarterly for Exercise and Sport, 73(2), 206210.
Proctor, M. H., Moore, L .L., Gao, D, Cupples, L. A, Bradlee, M. L., Hood, M. Y., Ellison. R. C.
(2003). Television viewing and change in body fat from preschool to early adolescence:
The Framingham Children's Study. Int J Obes Relat Metab Disord, 27, 827-833.
nd

Raykov, T., & Marcoulides, G. A. (2006). A First Course in Structural Equation Modeling (2
Lawrence Erlbaum Associates, London.

)

Ritchie, L. D., G. Welk, D. Styne, D. E. Gerstein, and P. B. Crawford. (2005). “Family
Environment and Pediatric Overweight: What is a Parent To Do?” Journal of the American
Dietetic Association, 105 (5), S70-9.
Riesch, S. K., Anderson, L.S., Krueger. H. A. (2006). Parent-child communication processes:
preventing children's health-risk behavior. J Spec Pediatr Nurs. 11(1), 41-56.
Robbins, L. B., Pender, N. J., Ronis, D. L., Kazanis, A. S., & Pis, M. B. (2004). Physical activity,
self-efficacy, and perceived exertion among adolescents. Research in Nursing & Health,
27, 435-446.
Robbins, L. B., Pis, M. B., Pender, N. J., & Kazanis, A. S. (2004). Exercise self-efficacy,
enjoyment, and feeling states among ado-lescents. Western Journal of Nursing Research,
26, 699-715.
Roew, R., Pickles, A., Simonoff, E., Bulik, C. M. & Silberg, J. L. (2002). Bulimic symptoms in
the Virginia Twin Study of Adolescent Behavioral Development: correlates, comorbility,
116

and genetics. Biol Psychiatry, 51, 172-182.
Roschelle, J. M., Pea, R. D., Hoadley, C. M., Gordon, D. N., & Means, B. M. (2000). Changing
how and what children learn in school with computer-based technologies. Children and
Computer Technology, 10(2), 76-101.
Willoughby, T. (2008). Longitudinal study of internet and computer game use by adolescent boys
and girls: Prevalence, frequency of use, and psychosocial predictors. Developmental
Psychology, 44, 195–204.
Witt E. A., Massman, A. J., & Jackson, L. A. (2011). Trends in youth’s videogame playing, overall
computer use, and communication technology use: The impact of self-esteem and the Big
Five personality factors. Computer in Human Behavior, 27(2), 763-769.
Wu, T., & Pender, N. (2002). Determinants of physical activity among Taiwanese adolescents:An
application of the health promotion model. Research in Nursing & Health, 25, 25-36.
Sallis, J. F., Conway, T. L., Prochaaka, J. J., Mckenzie, T. L., Marshall, S. J., Brown, M. (2001).
The association of school environments with youth physical activity. Am J Pub Health,
91, 618-620.
Sallis, J. F., Conway, T. L., Elder, J. P., Prochaaka, J. J., Brown, M., Zive, M. M., Marshall, S. J.,
Alcaraz, J. E. (2003). Environment interventions for eating and physical activity: A
Randomized control trail in middle schools. Am J Prev Med, 24, 209-217.
Sallis, J.F., Prochaska, J.J., & Taylor, W.C. (2000). A Review of Correlates of Physical Activity of
Children and Adolescents. Medicine & Science in Sports & Exercise, 32(5),963-975.
Salmon, J., Booth, M. L., Phongsavan, P., Murphy, N., Timperio, A. (2007). Promoting physical
activity participation among children and adolescents. Epidemiol Rev, 29, 144-59.
Siedlecki, K. L., Manly, J. J., Brickman, A. M., Schupf, N., Tang, M. X., & Stern, Y. (2010). Do
neuropsychological tests have the same meaning in Spanish speakers as they do in English
speakers? Neuropsychology, 24(3), 402-411.
Smith, A. B. & Bashore, L. (2006). The Effect of Clinic-Based Health Promotion Education on
Perceived Health Status and Health. Journal of Pediatric Oncology Nursing, 23(6), 326334. doi: 10.1177/1043454206293266.
Srof, J. B. & Velsor-Friedrich, B. (2006). Health Promotion in Adolescents: A Review of Pender’s
Health Promotion Model. Nursing Science Quarterly, 19(4), 366-373.
Stephenson, M., Quick, B., & Atkinson, J. (2005). Authoritative parenting and drug-prevention
practices: Implications for antidrug ads for parents. Health Communication, 17 (3), 301321.
117

Stice, E., Presnell, K., & Spangler, D. (2002). Risk factors for being eating on set in adolescent
girl: 2-year prospective investigation. Health Psycho, 21,131-138.
Strong, W.B. et al. (2005). Evidence based physical activity for school-age youth. Journal
Pediatar. 146, 732-737.
Story, D. Neumark-Sztainer, S. & French, J. A. (2004). Influences on adolescent eating patternsthe
importance of family meals. J Adolesc Health, 32 (5), 365–373.
Stuifbergen, A. K., & Rogers, S. (1997). Health promotion: An essential component of
rehabilitation for persons with chronic disabling conditions. Advances in Nursing Science,
19(4), 1-20.
Stuifbergen, A. K., & Timmerman, G. (2003). A randomized clinical trial of a wellness
intervention for women with multiples sclerosis. Archives of Physical Medicine and
Rehabilitation, 84, 467-476.
Subrahmanyam, K., Krant, R. E., Greenfield, P. M., & Gross, E. F. (2000). The impact of home
computer use on children’s activities and development. The Future of Children, 10, 123144.
Thai Health Promotion. (2013). Retrived from http://en.thaihealth.or.th/about/strategy
Teerarungsikul, N., Phuphaibul, R., Loveland-Cherry, C.J., Pookboonmee, Kijboonchoo, K.,
Nityasuddhi D. (2009). Effectiveness of a Physical Activity Promotion Program on
Perceived Self-Efficacy, Physical Activity and Physical Fitness among Thai Adolescent
Girls. Thai J Nurs Res.13(2), 81-94.
Tomori, M., Rus-Makovec, M. (2000). Eating behavior, depression, and self-esteem in high school
students. Journal of Adolescent Health, 26(36), 361-367.
Trost, G. S., Sallis, F. J., Patty, R. R., Freedson, P.S., Taylor, W. C., & Dowda, M. (2003).
Evaluating a Model of Parental Influence on Youth Physical Activity. American Journal of
Preventive Medicine. 25(4), 277-282. doi: 10.1177/1559827610387236
USDHSS (2012). How much physical activity do you need? Retrieved from
http://www.cdc.gov/physicalactivity/everyone/guidelines/children.html
Verstraete, S. J., Cardon, G. M., De Clercq, D. L., De Bourdeaudhuij, II. M. (2006). Increasing
children's physical activity levels during recess periods in elementary schools: the effects of
providing game equipment. Eur J Public Health, 16(4),415-419.
Villarruel, F., & Luster, T. (Eds) (2006). The crisis in youth mental health. Vol. 2: Issues for
Adolescence. Westport, CT: Preager Press.

118

Van Voorhees, B. W. et al. (2008). Integrative Internet-Based Depression Prevention for
Adolescents: A Randomized Clinical Trial in Primary Care for Vulnerability and Protective
Factors. J Can Acad Child Adolesc Psychiatry. 17(4): 184–196.
Wenglinsky, H. (1998). Does it compute? The relationship between educational technology and
student achievement in mathematics. Princeton, NJ: Educational Testing Service.
Rehabilitation, 69(8), 583-590.
Wu, T. Y., & Pender, N. J. (2005). A panel study of physical activity in Taiwanese youth: Testing
the revised Health-Promotion Model. Family & Community Health, 28, 113–124.
Yarcheski, A., Noreen E. Mahon, N. E., & Yarcheski, T. J. (2004)A Meta-Analysis of Predictors
of Positive Health Practices. Journal of Nursing Scholarship, 20, 102–108.
doi: 10.1111/j.1547-5069.2004.04021.x
Yancy, W.S., Olsen, M.K., Guyton, J.R., Bakst, R.P., Westman, E.C. (2004). A low-carbohydrate,
ketogenic diet vs. a low-fat diet to treat obesity and hyperlipidemia: a randomized
controlled trial. Ann Intern Med, 140, 769-777.

119