Vol.3, No.1, 42-50 (2013) Open Journal of Preventive Medicine
Mediation effects of social support on relationships
of perceived environment and self-efficacy with
school-based physical activity: A structural equation
model tailored for Japanese adolescent girls
Li He1*, Kaori Ishii2, Ai Shibata2, Minoru Adachi3, Keiko Nonoue4,5, Koichiro Oka2
1Graduate School of Sport Sciences, Waseda University, Tokorozawa, Japan; *Corresponding Author: meethl@moegi.waseda.jp
2Faculty of Sport Sciences, Waseda University, Tokorozawa, Japan
3Graduate School of Education, Okayama University, Okayama, Japan
4Sonan Junior High School, Okayama, Japan
5Waseda Institute for Sport Sciences, Tokorozawa, Japan
Received 5 December 2012; revised 7 January 2013; accepted 15 January 2013
Background: Identifying correlates of physical
activity that can be targeted as potential me-
diators is important for developing interven-
tions to promote physical activity in adolescent
girls. However, the mediated effects of multi-
level correlates of ph ysical activit y remain p oor ly
understood. Therefore, the present study aimed
to examine direct and mediated effects of per-
sonal, social and perceived school physical en-
vironmental factors on school-based physical
activity of Japanese adolescent girls. Methods:
In this cross-sectional survey of the Japanese
adolescent lifestyles, 344 junior high school
girls were invited to complete self-report meas-
ures of age, grade, weight, height, self-efficacy,
social support (family, friends and teachers),
perceived school physical environment (equip-
ment, facilities and safety) and physical activity
at school (min per week during lunch time and
after-school hours). Structural equation model-
ing analysis controlling for age was performed
to examine the effects of body mass index (BMI),
self-efficacy, social support and school physical
environmental variables on lunchtime and af-
ter-hours physical activity. Results: The final
structural model demonstrated an acceptable
fit for each context-specific physical activity.
During lunch recess, perceived equipment and
friend support exhibited direct effects on physi-
cal activity; perceived facilities, safety, and self-
efficacy were indirectly associated with physical
activity through friend support. During after-
school hours, both family and friend support
directly affected physical activity at school;
perceived safety, facilities and self-efficacy ex-
hibited indire ct effect s on physical acti vity through
family or friend support. How ever, there were no
significant associations between equipment and
after-school-hours physical activity. Regardless
of contexts, BMI had neither direct nor indirect
effects on physical activity. Conclusion: Social
support from family and friends was identified
as factors mediating the effects of perceived
environment and self-efficacy on school-based
physical activity among Japanese adolescent
girls. This finding encourages the future devel-
opment of effective interventions to promote
physical activity through family and friend sup-
port in the fut ur e.
Keywords: Public Health; Physical Activity;
Adolescents; Correlates; Ecological Model;
Regular physical activity during adolescence has been
well demonstrated to have short- and long-term benefits
for health [1,2]. However, the majority of adolescents are
not physically active. Throughout the world, participa-
tion in physical activity declines dramatically from ado-
lescence, particularly in girls [3]. In Japan, girls are less
active than boys at all ages. According to the latest na-
tional investigation, about 39% of adolescent girls (vs.
16% of adolescent boys) do not engage in 60 min of
physical activity daily [4]. Considering that physical ac-
tivity interventions should be designed for specific at-
risk groups, adolescent girls are one such group, which
Copyright © 2013 SciRes. OPEN ACCE SS
L. He et al. / Open Journal of Preventive Medicine 3 (2013) 42-50 43
underscores the interest in developing successful inter-
ventions to promote their physical activity.
Such interventions require identifying modifiable cor-
relates of physical activity that can be targeted as poten-
tial mediators [5]. However, most previous studies fo-
cused on the direct effects of correlates. There is little
research investigating indirect effects of factors on
physical activity.
From the social ecological perspective, physical ac-
tivity behavior is determined by an interaction of multi-
level influences across personal, social and environ-
mental factors, although previous studies mainly exam-
ine psychological and/or social environmental factors
[6-12]. Few studies have examined physical environ-
mental influences on physical activity, which are recog-
nized as having long-term effects on population-based
health behaviors. Even fewer studies have examined both
the direct and indirect effects of multilevel factors on
physical activity behavior [13-15].
Moreover, such ecological approach suggests that
examined behaviors should be specific to the relevant
environment (i.e., the school physical environment is
relevant to school-based physical activity). However,
existing studies of environmental influences on adoles-
cent physical activity behavior have focused on overall
physical activity level without distinguishing physical
activity in different contexts (i.e., in certain places and at
certain times) [16]. Currently, no studies have reported
direct and indirect effects of correlates on context-spe-
cific physical activity among adolescents, although a
comprehensive understanding of how environmental and
personal factors interact to influence individuals’ con-
text-specific physical activity behavior is necessary for
the development of effective interventions and health po-
For children and adolescents, schools are important
settings for physical activity. Students spend much of
their day at school and have many potential opportunities
for daily activity there, such as physical education
classes, break times (including lunch recess) and af-
ter-school hours. Physical education classes alone do not
provide them with enough opportunity to meet recom-
mended amounts of physical activity [17]. One review
suggested that involvement in physical activity during
non-curricular school time contributes 5% - 40% of the
recommended daily 60 minutes of physical activity [18].
It is therefore important to explore the direct and indirect
effects of personal, social and school environmental fac-
tors that may influence such non-curricular physical ac-
tivity at school.
As a particularity of social context among Japanese
adolescent girls, the strong desire to thinness or under-
weight can be listed. Contrary to other developed coun-
tries in which increase of obesity are major problems, the
prevalence of underweight increased among them [19-
22]. Society’s equating thinness with beauty and attract-
tiveness in women can be one of reason for this [23].
Thus, contribution of body weight on the relationship
between multilevel factors and physical activity may be
different between Japanese adolescents and those in
other western countries.
On the other hand, increase in autonomy and peer in-
fluence is one of relatively common characteristics in
adolescence regardless of nationality. Especially, gender
influences on these become more pronounced [24].
Compared with boys, adolescent girls, with the earlier
development of puberty and self-concept, has been found
to be more concerned about the self-appearance, popu-
larity in the class, the evaluation from other classmates
and teachers [25]. Thus, even in same school environ-
ments, the pattern and direction in relationship between
factors and physical activity would be difference be-
tween genders. Therefore, further understanding how the
multilevel factors interact with each other to develop the
tailored intervention strategies to promote physical ac-
tivity among adolescent girls.
In sum, the present study expanded current literature
by examining not only the direct but also the indirect
effects of all the selected individual, social and environ-
mental factors with school-based physical activity at dif-
ferent times of days in Japanese adolescent girls.
2.1. Participants and Data Collection
Data for the present study were collected from a cross-
sectional survey of the Japanese adolescent lifestyle.
Data collection was conducted between Oct. and Dec.
2010. A total of 761 public junior high school students
(aged 12 - 15 years old) in Japan participated in this sur-
vey, including 344 girls. All participants were invited to
complete a self-report questionnaire individually during
class time. Information on demographic such as age,
gender and grade were collected with this questionnaire.
Informed consent was obtained from all participants,
their guardian and the school. Participation was volun-
tary, and confidentiality was ensured. The study protocol
was approved by the Research Ethics Committee of Wa-
seda University.
2.2. Measures
Measures of weight and height were collected through
a height and weight measuring scale. Body mass index
(BMI) was calculated. Participants were grouped into
underweight, normal weight, overweight and obesity by
BMI ranges specific for age and gender using Centers for
Disease Control and Prevention criteria [26,27].
Copyright © 2013 SciRes. OPEN ACCE SS
L. He et al. / Open Journal of Preventive Medicine 3 (2013) 42-50
Participants were required to report the frequency
(day/week) and duration (min/day) of physical activity at
school during lunch recess and after hours in a usual
week. For analysis, minutes per week were calculated for
each setting (lunchtime physical activity and after-school
physical activity) [28].
Based on a previous instrument [29], 10 items were
used to assess three factors of school physical environ-
ment. The three factors were 1) “equipment” (3 items),
examining the accessibility or usability of physical equip-
ment (e.g., There is enough equipment for activities at
school); 2) “facilities” (4 items), measuring the accessi-
bility or usability of physical activity facilities (e.g., The
school grounds are big enough for activities); and 3)
“safety” (3 items), investigating perceived safety of phy-
sical activity equipment and facilities (e.g., It is safe to
engage in physical activity on the grounds and in the
gym at school). All items were rated on a four-point
scale from 1) strongly disagree to 4) strongly agree. The
factorial reliability (equipment: Cronbach α = 0.71; fa-
cilities: Cronbach α = 0.75; and safety: Cronbach α =
0.83) of this scale was confirmed by respondents.
In terms of social support for physical activity, par-
ticipants were asked to rate support from three sources
on a four-point scale from 1) not supportive at all to 4)
strongly supportive for the following question: “How do
you rate support for engaging in physical activity from 1)
family; 2) teachers at school and 3) friends at school?”.
The measure of self-efficacy related to physical activ-
ity (i.e., belief in one’s ability to be active relative to
peers [30]) contained 1 item with responses ranging from
1) strongly disagree to 4) strongly agree. The statement
was “I am able to do physical activities/exercises/sports
better than my friends.”
2.3. Data Analysis
Only girls with complete data (N = 280) were included
in the analyses. Means and standard deviations (SD)
were then calculated through descriptive statistics using
SPSS18.0 software. The size of the final sample was
adequate to estimate the models [31]. Finally, structural
equation modeling (SEM) analysis with maximum like-
lihood estimation in Amos 17.0 was performed to test the
fit of proposed models.
The original proposed model that led to a good model
fit of the final model is described below. The measure-
ment model included (a) three latent variables of physi-
cal environment: equipment (3 indicators), facilities (4
indicators) and safety (3 indicators); (b) relations be-
tween latent variables and their indicators and (c) corre-
lations between the three latent environmental factors.
The structural model included (a) paths from perceived
physical equipment, facilities and safety and BMI to
perceived self-efficacy and self-reported physical activity;
(b) path from self-efficacy to each source of social sup-
port and (c) paths from self-efficacy and three sources of
social support to physical activity.
Model fit was assessed using the goodness-of-fit index
(GFI), adjusted goodness-of-fit index (AGFI), root mean
square error of approximation (RMSEA) and Akaike
information criterion (AIC). GFI and AGFI are used to
measure how well the model fits the data, which varies
from 0 to 1, with .90 indicating an acceptable model fit
and 0.95 indicating a good model fit [32,33]. RMSEA is
a measure of the discrepancy between a population-based
model and a hypothesized model assessed per degree of
freedom. There is good model fit if the RMSEA is less
than or equal to 0.05, with the upper limit of confidence
interval less than 0.08 and the lower 90% confidence
limit including or close to 0 [33]. A lower AIC value for
a model reflects a better-fitting model compared with
competing models [34]. A model was considered to fit
the data when the following criteria were met: GFI >
0.90, AGFI > 0.90 (AGFI < GFI), RMSEA< 0.05 and a
lower AIC value than competing models. A p value less
than 0.05 was considered statistically significant.
To adjust the original specified model, new free pa-
rameters were added based on the modified indices be-
fore the Wald test that deleted all non-significant free
parameters to increase model fitness. Then only signifi-
cant causal paths with corresponding standardized re-
gression coefficients (β) were shown in the figures of
final structural models that demonstrated a good model
fit. With the standardized regression coefficients, the
magnitude of each factor could be directly compared
with other factors in the model.
3.1. Participant Characteristics
Adolescent girls (mean age = 13.44, SD = 0.93) with
complete data comprised the final sample. Mean height
and weight were 155.37 cm (SD = 5.32) and 46.98 kg
(SD = 8.72), respectively. The majority of adolescents
had normal weight (5 BMI < 85 percentile, n = 236,
84.3%). More information about characteristics of the
studied variables is provided in Table 1.
3.2. Structural Equation Model for
Lunch-Time Physical Activity
The final structural model for lunchtime physical ac-
tivity in Figure 1 demonstrated a good model fit (GFI =
0.95, AGFI = 0.93, RMSEA = 0.02 [90% confidence
interval = 0.00 - 0.04]). The recalculation of the model
after addition and deletion of free parameters reduced the
AIC value from 780.95 to 215.56. During lunch recess,
perceived friend support (β = 0.11) was found to have a
direct positive effect on girls’ physical activity. Self-
Copyright © 2013 SciRes. OPEN ACCE SS
L. He et al. / Open Journal of Preventive Medicine 3 (2013) 42-50 45
Table 1. Characteristics of participants and outcome physical
activity variables.
Participants (Na = 280)
Mean SDa
Age 13.44 0.93
Height 155.37 5.32
Weight 46.98 8.72
BMI 19.40 3.14
Lunch-time Physical Activity 8.46 26.74
After-school Physical Activity 138.89 259.05
Grade (N, %)
Grade 1 99 35.4%
Grade 2 96 34.3%
Grade 3 85 30.4%
Weight status (N, %)
Underweight 12 4.3%
Normalweight 236 84.3%
Overweight 22 7.9%
Obesity 10 3.6%
aN: number; SD: standard deviation.
Figure 1. Effects of personal, social and environmental factors
on lunch-time physical activity. Only statistically significant
paths are shown in the figure. The significance level was set at
p < 0.05. BMI: body mass index; Family: family support;
Teacher: teacher support; Friend: friend support.
efficacy (β = 0.04) indirectly influenced physical activity
through friend support. With respect to the influences of
school environmental factors, perceived equipment ex-
hibited a direct negative effect (β = 0.15) on physical
activity. The total effects of perceived facilities (β = 0.01)
and safety (β = 0.01) on physical activity were fully
mediated by self-efficacy and friend support. Equipment
was identified as the most influential environmental fac-
tor related to physical activity. There were no significant
associations of BMI, family support or teacher support
with physical activity.
3.3. Structural Equation Model for
After-School Physical Activity
The final structural equation model for after-school
physical activity in Figure 2 also demonstrated a good
model fit (GFI = 0.95, AGFI = 0.93, RMSEA = 0.03
[90% confidence interval = 0.00 - 0.04]). The AIC value
was reduced from 842.24 to 219.74 after the model
modifications. In the final structural model, perceived
equipment, teacher support and BMI failed to exhibit
direct or indirect effects on physical activity. Perceived
facilities (β = 0.02) and safety (β = 0.02) were found to
indirectly affect physical activity through self-efficacy
and family support or friend support. Their effect sizes
on physical activity were generally low. The standard-
ized indirect effect of self-efficacy on physical activity
through family and friend support was 0.09. Support
from friends (β = 0.16) and family (β = 0.13) were found
to directly affect physical activity. The final model iden-
tified friend support as the most influential factor directly
affecting physical activity during after-school hours.
The present cross-sectional study examined effects of
perceived school physical environment, social support,
self-efficacy and BMI on school-based physical activity
in adolescent girls.
Figure 2. Effects of personal, social and environmental factors
on after-school physical activity. Only statistically significant
paths are shown in the figure. The significance level was set at
p < 0.05. BMI: body mass index; Family: family support;
Teacher: teacher support; Friend: friend support.
Copyright © 2013 SciRes. OPEN ACCE SS
L. He et al. / Open Journal of Preventive Medicine 3 (2013) 42-50
The primary finding of this study was that social sup-
port directly affected school-based physical activity as a
possible mediator. Specifically, friend support was found
to be the source of support most highly related to physi-
cal activity in both lunchtime and after-school settings.
Previous studies have demonstrated similar findings
on moderate-to-vigorous physical activity during non-
school-time and mean daily physical activity in adoles-
cent girls [11,35-38]. Collectively, findings suggest that
developing strategies to encourage or assist with friends’
physical activity behaviors can be beneficial in promot-
ing physical activity in adolescent girls, regardless of
contexts. Compared with friend support, support from
family and teachers for physical activity were found to
be less important in both contexts. This finding is under-
standable because adolescent girls are becoming inde-
pendent from family and strengthening their identifica-
tion with peers [11,37]. Interestingly, the effect of family
support on physical activity was significant in the af-
ter-school context but not at lunch time. This suggests
that parents mainly act as a “gate keeper” by providing
instrumental support (e.g., transportation and providing
access to equipment) so that children can engage in or-
ganized physical activity during after-school hours [39,
40]. Similarly, the non-significant association of teacher
support with physical activity was understandable be-
cause physical activity becomes a free choice in leisure
time. Teachers’ influence may be more significant in
physical education courses rather than in free time be-
havioral choices. In Japan, approximately 71.2% of jun-
ior high school girls join in their school’s extracurricular
sports clubs during after-school hours. Only 0.7% of
girls are motivated by teachers. However, 26.4% of re-
port that they take part in extracurricular sports clubs
because of their friends and family [41]. Considering that
no comparable previous studies exist, to better under-
stand the context-specific difference in social support
related to physical activity, more in-depth research is
needed to examine preferred types of physical activity in
different contexts, and to identify types of support (e.g.,
encouragement or tangible assistance) in association with
context-specific physical activity. Findings in thepresent
study suggest considering family and friend support as
possible mediators in future longitudinal or intervention
studies to observe changes of physical activity in specific
contexts. Based on available evidence [35,42], when test-
ing the interventional effects of social support on direct
promotion of physical activity at school, the following
should be considered: increasing the frequency of parents
or friends watching or talking about physical activity,
parents or friends increasing their encouragement of girls
joining in extracurricular sport clubs started at school
and parental assistance with fees for equipment needed
for these clubs.
Self-efficacy has been demonstrated as a consistent
correlate and mediator of adolescents’ physical activity
[10,11,14,16,43,44]. However, in the current study, self-
efficacy did not directly affect physical activity of ado-
lescent girls. The inconsistency between previous studies
and the current study might be attributed to aspects of
self-efficacy measured. The present study focused on the
self-efficacy in performance of activities; previous stud-
ies primarily examined barriers self-efficacy. Ryan et al.
found that the impacts of different types of self-efficacy
(e.g., barriers self-efficacy, performance self-efficacy
and asking self-efficacy) on physical activity were dif-
ferent [30]. Additionally, our finding might further con-
firm the study of Dishman et al. suggesting that physical
activity interventions designed to enhance self-efficacy
might be especially needed during preadolescence [7].
Therefore, to gain a complete understanding of the re-
lationship between self-efficacy and context- specific
physical activity, future studies should include more as-
pects of self-efficacy that may directly account for the
variance in physical activity, and should follow changes
in self-efficacy throughout primary and junior high
This study provided additional evidence of associa-
tions between different features of school physical envi-
ronment and physical activity by context [45-48]. First,
perceived equipment had a direct negative effect on
girls’ lunchtime physical activity. It is possible that
girls who actually used equipment reported a lower
score for available equipment than girls who did not use
it because the latter were less aware of the real equip-
ment situation. Second, the negative effect of perceived
safety on school-based physical activity might be ac-
counted for by inaccurate measures of the physical envi-
ronment. Some previous studies have shown that the
agreement between perceived and objectively measured
environment is often poor, and the relationship between
objective and self-report measures of physical environ-
ment and physical activity is inconsistent among adoles-
cents [49,50]. Therefore, future research is needed to test
whether perceptions of equipment and safety match ob-
jective measurements to clarify the influences of equip-
ment and safety on physical activity. In addition, this
study indicated that increasing perceptions of accessible
facilities at school might promote physical activity par-
ticipation among adolescent girls through increasing
self-efficacy and social support. This information high-
lights the importance of increasing information channels
regarding available or accessible facilities at school
among students or parents. There are various ways to
accomplish this. For instance, school officials could re-
port to students and parents in timely manner updated
news about regular maintenance of facilities, or install a
sign on school grounds informing students about the
Copyright © 2013 SciRes. OPEN ACCE SS
L. He et al. / Open Journal of Preventive Medicine 3 (2013) 42-50 47
available facilities and policies about using them.
One more finding of this study was that BMI had no
significant influence on girls’ physical activity. BMI is
one of the most studied biological markers of body
weight and shape. Evidence from systematic reviews
have shown that there is no consistent association be-
tween BMI and adolescent girls’ physical activity level,
with the majority of studies reporting either a small
negative or no correlation [43,51,52]. The small or non-
significant effects of BMI may suggest that other poten-
tially body weight-and shape-related factors like body
image need to be assessed in future behavioral models
for girls. Different from BMI, the construct of body im-
age measured physical appearance attitudinally, consist-
ing of subjective feelings and beliefs on one’s own ap-
pearance (e.g., body dissatisfaction) [53] and perceptions
of how the body moves and functions, or what the body
can “do” [54]. Considering cultural influences and norms
on poor and ideal body image and that girls tend to be
more concerned about their physical appearance than
boys during adolescence, body image rather than BMI is
often thought to be more likely to reflect variances of
physical activity or other weight-related behavior in
adolescent girls [43,55]. Therefore, it would be worth-
while including the construct of body image in future
The collective evidence is sufficient to encourage the
use of support from friends and families as mediators in
future intervention studies designed to increase physical
activity among Japanese adolescent girls. Given that dif-
ferences between school systems and environments
may limit the generalizability of findings between coun-
tries, and many previous studies were conducted in
Western countries, research on Japanese adolescent girls
is needed. There are several strengths of the present re-
search worth noting. First, this study extended previous
research by measuring both direct and indirect effects
of multilevel contributing factors on context-specific
physical activity rather than overall physical activity
level. In addition, several sources of social support and
various school physical environmental attributes were
examined concurrently in the present study. Although
previous studies have compared the effects of family and
friend support on physical activity, there are no studies
that simultaneously investigated the relative importance
of family, friends and teachers on girls’ school-based
physical activity. Second, the present study contributed
to studies on adolescents by exploring the mechanisms
underlying the physical activity behavior of Asian ado-
lescent girls. Finally, the present study used SEM analy-
sis which was helpful in exploring potential mediators
that can possibly intervene and allowed examination of
relative contributions of factors explaining physical ac-
Despite the strengths, a number of limitations must be
noted. One is the use of a self-report measure of physical
activity, which is subject to error and bias. Further stud-
ies should combine existing objective and subjective
measures to investigate context-specific physical activity
more accurately. Furthermore, lack of objective meas-
urement of the physical environment characteristics may
account for the inaccurate perceptions of the physical
environment. Future research should use both self-report
and objective measures to accurately evaluate the envi-
ronment. Another limitation is the list-wise deletion
adopted in the present study, which may lead to bias in
data findings. A further limitation is that the generalize-
bility of findings beyond the study location may be lim-
ited by the data from a single school. However, the
prevalence of overweight and obesity in the present
study (11.5%) was slightly higher than the prevalence in
a national survey (7.4%) [22]. This implies that the
structural models in the present study might be reflective
of counterparts across the country. Still another limita-
tion is the cross-sectional data used in this study, which
only permits estimates of between-person relations among
variables. Therefore, a longitudinal or interventional de-
sign that permits estimates of the effects of changes of
mediators on the physical activity within participants is
warranted in future research.
In summary, the present study indicated that 1) social
support mediated the cross-sectional effect of physical
environment and self-efficacy on physical activity among
adolescent girls and 2) associations between examined
factors and school-based physical activity differed by
context. Results imply that it is possible to develop a
school-based intervention through collective supports
from families and friends to increase girls’ school-based
physical activity level. Furthermore, the mediated effect
of perceptions of self-efficacy and physical environment
that may be helpful for increasing perceptions of social
support should be tested in future studies.
The authors would like to thank the participating students and school
in the present study. This investigation was supported by the Grants-
in-Aid for Scientific Research (No. 22700680) from the Japan Society
for the Promotion of Science, Waseda University Grant for Special
Research Projects (2010A-095, 2011A-092), and the Global COE Pro-
gram “Sport Sciences for the Promotion of Active Life” from the Japan
Ministry of Education, Culture, Sports, Science, and Technology.
[1] Hallal, P.C., Victora, C.G., Azevedo, M.R. and Wells, J.C.
(2006) Adolescent physical activity and health: A system-
atic review. Sports Medicine, 36, 1019-1030.
Copyright © 2013 SciRes. OPEN ACCE SS
L. He et al. / Open Journal of Preventive Medicine 3 (2013) 42-50
[2] Janssen, I. and LeBlanc, A.G. (2010) Systematic review
of the health benefits of physical activity and fitness in
school-aged children and youth. International Journal of
Behavioral Nutrition and Physical Activity, 7, 40.
[3] World Health Organization (2012) Global recommenda-
tion on physical activity for health.
[4] Ministry of Education, Culture, Sports, Science and
Technology, Japan (2012) 2010’s national survey result of
physical and athletic capacity and exercise habits: (6)
sports and exercise habits.
[5] Baron, R. and Kenny, D. (1986) The moderator-mediator
variable distinction in social psychological research:
conceptual, strategic, and statistical considerations. Jour-
nal of Personality and Social Psychology, 51, 1173-1182.
[6] Dishman, R.K., Motl, R.W., Saunders, R., Felton, G.,
Ward, D.S. and Dowda, M., et al. (2004) Self-efficacy
partially mediates the effect of a school-based physical
activity intervention among adolescent girls. Preventive
Medicine, 38, 628-636. doi:10.1016/j.ypmed.2003.12.007
[7] Dishman, R.K., Saunders, R.P., Motl, R.W., Dowda, M.,
and Pate, R.R. (2009) Self-efficacy moderates the relation
between declines in physical activity and perceived social
support in high school girls. Journal of Pediatric Psy-
chology, 34, 441-451. doi:10.1093/jpepsy/jsn100
[8] Dishman, R.K., Dunn, A.L., Sallis, J.F., Vandenberg, R.J.
and Pratt, C.A. (2010) Social-cognitive correlates of
physical activity in a multi-ethnic cohort of middle-
school girls: Two-year prospective study. Journal of Pe-
diatric Psychology, 35, 188-198.
[9] Hsu, YW, Chou, C.P., Nguyen-Rodriguez, S.T., McClain,
A.D., Belcher, B.R. and Spruijt-Metz, D. (2011) Influ-
ences of social support, perceived barriers, and negative
meanings of physical activity on physical activity in mid-
dle school students. Journal of Physical Activity and
Health, 8, 210-219.
[10] Trost, S.G., Sallis, J.F., Pate, R.R., Freedson, P.S., Taylor,
W.C. and Dowda, M. (2003) Evaluating a model of pa-
rental influence on youth physical activity. American
Journal of Preventive Medicine, 25, 277-282.
[11] Wu, T., Pender, N. and Noureddine, S. (2003) Gender
differences in the psychosocial and cognitive correlates of
physical activity among Taiwanese adolescents: A struc-
tural equation modeling approach. International Journal
of Behavioral Medicine, 10, 93-105.
[12] Motl, R.W., Dishman, R.K., Ward, D.S., Saunders, R.P.,
Dowda, M. and Felton, G., et al. (2005) Comparison of
barriers self-efficacy and perceived behavioral control for
explaining physical activity across 1 year among adoles-
cent girls. Health Psychology, 24, 106-111.
[13] Motl, R.W., Dishman, R.K., Ward, D.S., Saunders, R.P.,
Dowda, M. and Felton, G., et al. (2005) Perceived physi-
cal environment and physical activity across one year
among adolescent girls: Self-efficacy as a possible me-
diator? Journal of Adolescent Health, 37, 403-408.
[14] Motl, R.W., Dishman, R.K., Saunders, R.P., Dowda, M.
and Pate, R.R. (2007) Perceptions of physical and social
environment variables and self-efficacy as correlates of
self-reported physical activity among adolescent girls.
Journal of Pediatric Psychology, 32, 6-12.
[15] Lubans, D.R., Okely, A.D., Morgan, P.J., Cotton, W.,
Puglisi, L. and Miller, J. (2012) Description and evalua-
tion of a social cognitive model of physical activity be-
haviour tailored for adolescent girls. Health Education
Research, 27, 115-128. doi:10.1093/her/cyr039
[16] Van der Horst, K., Paw, M.J., Twisk, J.W. and Van Mech-
elen, W. (2007) A brief review on correlates of physical
activity and sedentariness in youth. Medicine & Science
in Sports & Exercise, 39, 1241-1250.
[17] McKenzie, T.L., Marshall, S.J., Sallis, J.F. and Conway,
T.L. (2000) Leisure-time physical activity in school en-
vironments: An observational study using SOPLAY. Pre-
ventive Medicine, 30, 70-77.
[18] Ridgers, N., Stratton, G. and Fairclough, S. (2006) Physi-
cal activity levels of children during school playtime.
Sports Medicine, 36, 359-371.
[19] Ministry of Health, Labour, and Welfare, Japan (2007)
2007’s national survey results of Japanese health and nu-
trition status.
[20] Ministry of Health, Labour, and Welfare, Japan (2008)
2008’s national survey results of Japanese health and nu-
trition status.
[21] Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan (2009) 2009’s national survey result of
physical and athletic capacity and exercise habits: Physi-
cal fitness and BMI.
[22] Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan (2010) 2010’s national survey result of
physical and athletic capacity and exercise habits: Physi-
cal fitness and BMI.
[23] Sano, A., Le, D.S., Tran, M.H., Pham, H.T., Kaneda, M.
and Murai E., et al. (2008) Study on factors of body im-
age in Japanese and Vietnamese adolescents. Journal of
Nutritional Science and Vitaminology, 54, 169-175.
[24] Berk, L.E. (2005) Infants, children and adolescents. 5th
Copyright © 2013 SciRes. OPEN ACCE SS
L. He et al. / Open Journal of Preventive Medicine 3 (2013) 42-50 49
Edition, Allyn & Bacon Inc., Bosten.
[25] Koizumi K. (1997) Adaptation process during transition
from elementary school to junior high school. (Shō-
chūgakkō de no kankyō ikō jitai ni okeru jidō, seito no
tekiō katei chūgakkō nyūgaku, tenkō o chūshin to shite,
in Japanese). Kasama Shobo, Tokyo.
[26] Cole, T., Bellizzi, M., Flegal, K. and Dietz, W. (2000)
Establishing a standard definition for child overweight
and obesity worldwide: International survey. Brtish Medi-
cal Journal, 320, 1240-1243.
[27] Centers for Disease Control and Prevention (2011) About
BMI for children and teens.
[28] He, L., Ishii, K., Shibata, A., Adachi, M., Nonoue, K. and
Oka K. (2012) Patterns of physical activity outside of
school time among Japanese junior high school students.
Journal of School Health, in Press.
[29] Robert-Wilson, J., Levesque, L. and Holden, R.R. (2007)
Development of a questionnaire assessing school physical
activity environment. Measurement in Physical Educa-
tion and Exercise Science, 11, 93-107.
[30] Ryan, G. and Dzewaltowski, D. (2002) Comparing the
relationships between different types of self-efficacy and
physical activity in youth. Health Education & Behavior,
29, 491-504. doi:10.1177/109019810202900408
[31] North Carolina State University (2012) Structural equa-
tion modeling: Statnotes. North Carolina State University,
[32] Kline, R.B. (1998) Principle and practice of structural
equation modeling. Guilford, New York.
[33] Schumacker, R.E. and Lomax, R.G. (2004) A beginner’s
guide to structural equation modeling. 2nd Edition, Law-
rence Erlbaum Associates, Mahwah.
[34] Hamparsum, B. (1987) Model selection and Akaike’s
information criterion (AIC): The general theory and its
analytical extensions. Psychometrika, 52, 345-370.
[35] Duncan, S., Duncan, T. and Strycker, L. (2005) Sources
and types of social support in youth physical activity.
Health Psychology, 24, 3-10.
[36] Hohepa, M., Scragg, R., Schofield, G., Kolt, G.S. and
Schaaf, D. (2007) Social support for youth physical ac-
tivity: Importance of siblings, parents, friends and school
support across a segmented school day. International Jour-
nal of Behavioral Nutrition and Physical Activity, 4, 54.
[37] Lytle, L.A., Murray, D.M., Evenson, K.R., Lytle, L.A.,
Murray, D.M. and Evenson, K.R., et al. (2009) Mediators
affecting girls’ levels of physical activity outside of school:
Findings from the trial of activity in adolescent girls. An-
nals of Behavioral Medicine, 38, 124-136.
[38] Patnode, C.D., Lytle, L.A., Erickson, D.J., Sirard, J.R.,
Barr-Anderson, D. and Story, M. (2010) The relative in-
fluence of demographic, individual, social, and environ-
mental factors on physical activity among boys and girls.
International Journal of Behavioral Nutrition and Physi-
cal Activity, 7, 79. doi:10.1186/1479-5868-7-79
[39] Alderman, B.L., Benham-Deal, T.B. and Jenkins, J.M.
(2010) Change in parental influence on children’s physi-
cal activity over time. Journal of Physical Activity and
Health, 7, 60-67.
[40] Welk, G., Wood, K. and Morss, G. (2003) Parental influ-
ences on physical activity in children: An exploration of
potential mechanisms. Pediatric Exercise Science, 15,
[41] Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan (2010) 2010’s national survey result of
physical and athletic capacity and exercise habits.
[42] Robbins, L.B., Stommel, M. and Hamel, L.M. (2008)
Social support for physical activity of middle school stu-
dents. Public Health Nursing, 25, 451-460.
[43] Biddle, S.J.H., Whitehead, S.H.O., Donovan, T.M. and
Nevill, M.E. (2005) Correlates of participation in physical
activity for adolescent girls: A systematic review of recent
literature. Journal of Physical Activity and Health, 2,
[44] Salmon, J., Brown, H. and Hume, C. (2009) Effects of
strategies to promote children’s physical activity on po-
tential mediators. International Journal of Obesity, 33,
S66-S73. doi:10.1038/ijo.2009.21
[45] Haug E., Torsheim T. and Samdal O. (2008) Physical
environmental characteristics and individual interests as
correlates of physical activity in Norwegian secondary
schools: The health behaviour in school-aged children
study. International Journal of Behavioral Nutrition and
Physical Activity, 5, 47. doi:10.1186/1479-5868-5-47
[46] Haug, E., Torsheim, T., Sallis, J.F. and Samdal, O. (2010)
The characteristics of the outdoor school environment
associated with physical activity. Health Education Re-
search, 25, 248-256. doi:10.1093/her/cyn050
[47] Prins, R.G., van Empelen, P., Velde, S.J.T., Timperio, A.,
van Lenthe, F.J. and Tak N.I., et al. (2010) Availability of
sports facilities as moderator of the intention-sports par-
ticipation relationship among adolescents. Health Educa-
tion Research, 25, 489-497.
[48] Colabianchi, N., Maslow, A.L. and Swayampakala, K.
(2011) Features and amenities of school playgrounds: A
direct observation study of utilization and physical ac-
tivity levels outside of school time. International Journal
of Behavioral Nutrition and Physical Activity, 8, 32
[49] Maddison, R., Jiang, Y., Hoorn, S.V., Mhurchu, C.N.,
Exeter, D. and Utter J. (2010) Perceived versus actual
distance to local physical-activity facilities: Does it really
matter? Journal of Physical Activity and Health, 7, 323-
[50] McCormack, G., Giles-Corti, B., Lange, A., Smith, T.,
Copyright © 2013 SciRes. OPEN ACCE SS
L. He et al. / Open Journal of Preventive Medicine 3 (2013) 42-50
Copyright © 2013 SciRes. OPEN ACCE SS
Martin, K. and Pikora, T.J. (2004) An update of recent
evidence of the relationship between objective and self-
report measures of the physical environment and physical
activity behaviors. Journal of Science and Medicine in
Sport, 7, 81-92. doi:10.1016/S1440-2440(04)80282-2
[51] Sallis J., Prochaska J. and Taylor W. (2000) A review of
correlates of physical activity of children and adolescents.
Medicine & Science in Sports & Exercise, 32, 963-975.
[52] Van der Horst, K., Paw, M.J.C.A., Twisk, J.W.R. and Van
Mechelen, W. (2007) A brief review on correlates of phy-
sical activity and sedentariness in youth. Medicine &
Science in Sports & Exercise, 39, 1241-1250.
[53] Thompson, J.K., Heinberg, L.J., Altabe, M.N. and Tantleff-
Dunn, S. (1999) Exacting beauty: Theory, assessment,
and treatment of body image disturbance. American Psy-
chological Association, Washington DC.
[54] Abbott, B.D. and Barber, B.L. (2011) Differences in
functional and aesthetic body image between sedentary
girls and girls involved in sports and physical activity:
Does Sport Type Make a Difference? Psychology of Sport
and Exercise, 12, 333-342.
[55] Rauste-von Wright, M. (1989) Body image satisfaction in
adolescent girls and boys: A longitudinal study. Journal
of Youth and Adolescence, 18, 71-83.