International Journal of Obesity (2001) 25, 822±829

ß 2001 Nature Publishing Group All rights reserved 0307±0565/01 $15.00

www.nature.com/ijo

PAPER
Physical activity and determinants of physical activity
in obese and non-obese children
SG Trost1*, LM Kerr2, DS Ward3 and RR Pate2
1
School of Human Movement Studies, The University of Queensland, Brisbane, Queensland, Australia; 2Department of Exercise
Science, School of Public Health, University of South Carolina, Columbia, South Carolina, USA; and 3School of Public Health,
University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

OBJECTIVE: To compare the physical activity (PA) patterns and the hypothesized psychosocial and environmental determinants
of PA in an ethnically diverse sample of obese and non-obese middle school children.
DESIGN: Cross-sectional study.
SUBJECTS: One-hundred and thirty-three non-obese and 54 obese sixth grade children (mean age of 11.4  0.6). Obesity status
determined using the age-, race- and gender-speci®c 95th percentile for BMI from NHANES-1.
MEASUREMENTS: Objective measurements were collected of PA over a 7-day period using the CSA 7164 accelerometer: total
daily counts; daily moderate (3 ± 5.9 METs) physical activity (MPA); daily vigorous physical activity (  6 METs; VPA); and weekly
number of 5, 10 and 20 min bouts of moderate-to-vigorous physical activity (  3 METs, MVPA). Self-report measures were
collected of PA self-ef®cacy; social in¯uences regarding PA, beliefs about PA outcomes; perceived PA levels of parents and peers,
access to sporting and=or ®tness equipment at home, involvement in community-based PA organizations; participation in
community sports teams; and hours spent watching television or playing video games.
RESULTS: Compared to their non-obese counterparts, obese children exhibited signi®cantly lower daily accumulations of total
counts, MPA and VPA as well as signi®cantly fewer 5, 10 and 20 min bouts of MVPA. Obese children reported signi®cantly lower
levels of PA self-ef®cacy, were involved in signi®cantly fewer community organizations promoting PA and were signi®cantly less
likely to report their father or male guardian as physically active.
CONCLUSIONS: The results are consistent with the hypothesis that physical inactivity is an important contributing factor in the
maintenance of childhood obesity. Interventions to promote PA in obese children should endeavor to boost self-ef®cacy
perceptions regarding exercise, increase awareness of, and access to, community PA outlets, and increase parental modeling
of PA.
International Journal of Obesity (2001) 25, 822 ± 829

Keywords: exercise; inactivity; overweight; accelerometry; objective monitoring

Introduction off points) increased from 6 to 10.7% and 4.8 to 10.7%,

The prevalence of obesity among US children and adoles-
cents is increasing at an alarming rate. In the 10 y between
the second (1976 ± 1980) and third (1988 ± 1991) administra-
tion of the National Health and Nutrition Examination
Survey (NHANES), the prevalence of obesity among children
(6 ± 11 y) and adolescents (12 ± 17 y) (based on age- and
gender-speci®c 95th percentile body mass index (BMI) cut-
*Correspondence: SG Trost, School of Human Movement Studies,
The University of Queensland, Brisbane, QLD 4072, Australia.
E-mail: strost@hms.uq.edu.au
Received 7 June 2000; revised 19 December 2000;
accepted 15 January 2001
respectively.1 Recent reports con®rm similar increases in the
prevalence of obesity among preschoolers and infants.2,3 The
rising trend in pediatric obesity represents a critical public
health problem. Obese children and adolescents are at
increased risk for adult obesity4 and are more likely than
their lean counterparts to experience signi®cant short-term
health problems such as hyperlipidemia, hypertension, glu-
cose intolerance and orthopedic complications.5,6 Moreover,
the adverse social consequences of childhood and adolescent
obesity may have long-lasting negative effects on self-
esteem, body image and economic mobility.6,7
Lack of physical activity is hypothesized to be an impor-
tant contributing factor in the development and=or

maintenance of childhood obesity.8,9 However, previous
studies on the relationship between physical activity and
adiposity in youth have produced inconsistent results. While
some studies show overweight status or adiposity to be
inversely related to physical activity participation, others
report no association.10 ± 12 Such discrepant ®ndings may,
in part, be related to the dif®culties associated with obtain-
ing valid and reliable measures of physical activity in chil-
dren.13,14 To date, a wide range of methods have been used to
measure physical activity behavior in lean and obese chil-
dren. These include self-report questionnaires, direct obser-
vation, heart rate monitors, motion sensors such as
accelerometers and pedometers, and doubly-labeled water
(DLW). However, due to their low cost and ease of adminis-
tration, self-report methods are the most commonly used
method. This is problematic, given that self-reports are sub-
ject to considerable recall bias and have limited validity and
reliability among children.13 ± 16
To avoid the limitations of self-report, a relatively small
number of investigations have utilized heart rate monitoring
(HRM) and doubly-labeled water (DLW) to examine weight-
related differences in physical activity and=or energy expen-
diture. However, these measures have limitations as well.
With respect to heart rate monitoring, it is widely recognized
that the relationship between heart rate and energy expen-
diture is in¯uenced by a large number of factors including
age, body size, emotional stress, temperature and cardio-
respiratory ®tness.15,17 Additionally, because heart rate
tends to lag behind changes in movement and tends to
remain elevated after the cessation of movement, heart rate
monitoring may mask the intermittent activity patterns of
children.17 The DLW technique provides accurate assess-
ments of total energy expenditure over extended time peri-
ods (1 ± 3 weeks), but provides no information on the
frequency, intensity and duration of physical activity.18
Moreover, estimates of physical activity energy expenditure
are highly dependent on accurate assessments of resting
metabolic rate and dietary induced thermogenesis.19
Accelerometers that detect and record the magnitude of
movement on a real-time basis have the potential to overcome
some of the limitations of HRM and DLW. They provide reliable
information about the frequency, duration and intensity of
physical activity within a given day or over several days.18,20
Moreover, their small size and robust design features make
them suitable for use in ®eld-based studies involving moderate
to large numbers of children and adolescents.17,18 To date,
however, very few studies have utilized accelerometers to
assess obesity-related differences in physical activity, and
none have used contemporary models that assess physical
activity patterns (ie bouts of activity) over time.
Knowing the `determinants', or the factors that in¯uence
physical activity, in obese youth is an important prerequisite
to designing effective intervention strategies for this popula-
tion.21 Presently, however, our knowledge of the psycho-
social and environmental factors that in¯uence physical
activity behavior in obese children is limited. Previous stu-
Physical activity in obese and non-obese youth
SG Trost et al
823
dies have identi®ed physical activity self-ef®cacy, beliefs and
social norms related to physical activity, involvement in
community-based physical activity organizations, access to
equipment at home and parental physical activity as factors
associated with, or predictive of, physical activity behavior
in children and adolescents.14,22 However, to our knowledge,
the extent to which these variables differ in obese and non-
obese children has not been previously studied. If obesity-
related differences in physical activity can be linked to
differences in speci®c determinants of physical activity beha-
vior, these determinants can then be targeted for change in
intervention programs designed speci®cally for the needs of
overweight children.
The purpose of this study was to compare the physical
activity patterns and the hypothesized psychosocial and
environmental determinants of physical activity behavior
in an ethnically diverse sample of obese and non-obese
middle school children. In contrast to previous studies
comparing activity levels or energy expenditure in obese
and lean children, we utilized a state-of-the-art accelero-
meter to objectively measure both the quantity and intensity
of physical activity over a 7 day period.
Methods
Subjects
Subjects for this study were 213 sixth grade students from
four randomly selected public middle schools in Columbia,
South Carolina. The proportion of African-American stu-
dents attending these schools ranged from 41 to 55%,
while the percentage of females ranged from 45.3 to
51.4%. Between 42.9 and 82.7% of the students attending
these schools were eligible to receive free or reduced priced
lunches. Students participated in the study on a voluntary
basis and were recruited from required physical education
classes. The initial study group was 51.6% female, 55.9%
African-American, with a mean age of 11.4  0.6. After dele-
tions for incomplete physical activity data (n ˆ 15) and
missing height and weight (n ˆ 11), the ®nal sample con-
sisted of 187 students (98 females, 89 males). The descriptive
statistics for this group (52.4% female, 55.6% African-
American, mean age of 11.4  0.6) indicated that the demo-
graphic characteristics remained unchanged by the
exclusion of these participants. Prior to participation in
the study, written informed consent was obtained
from each student and his or her primary guardian. The
study was approved by the University of South Carolina
Institutional Review Board.
Anthropometric measures
Height and weight assessments were conducted in a private
setting with students dressed in light clothing. Height was
measured to the nearest 1.0 cm using a portable stadiometer.
Weight was measured to the nearest 0.2 kg with a standard
physician's beam scale (Detecto). Body mass index (BMI) was
International Journal of Obesity
 Physical activity in obese and non-obese youth
SG Trost et al
824
calculated as body weight in kilograms divided by height in
meters squared (kg=m2).
Weight status
Consistent with recent population surveys,23 participants
were classi®ed as obese if their BMI was equal to or greater
than the sex-, race- and age-speci®c 95th percentile from the
®rst National Health and Nutrition Examination Survey
(NHANES-1).24,25 Using this criterion, the number of parti-
cipants classi®ed as obese and non-obese was 54 and 133,
respectively.
Measurement of physical activity
Instrumentation. Objective assessments of physical activ-
ity behavior were obtained using the Computer Science and
Applications Inc. (CSA) 7164 activity monitor (Shalimar,
Florida). Brie¯y, the CSA 7164 is a uniaxial accelerometer
designed to detect vertical acceleration ranging in magni-
tude from 0.05 to 2.00 G with frequency response of 0.25 ±
2.50 Hz. These parameters allow for the detection of normal
human motion and will reject high frequency vibrations
encountered in activities such as operation of a lawn
mower. The ®ltered acceleration signal is digitized and the
magnitude is summed over a user-speci®ed epoch interval.
At the end of each epoch, the summed value or activity
`count' is stored in memory and the integrator is reset.26
For the current study, a 1 min time interval was used. Trost
and co-workers27 recently assessed the validity and inter-
instrument reliability of the CSA 7164 activity monitor in
children aged 10 ± 14. CSA activity counts were strongly
correlated with energy expenditure during treadmill walking
and running (Pearson r ˆ 0.86). The intraclass correlation
for two CSA 7164 monitors worn simultaneously was 0.87,
indicating a strong degree of inter-instrument reliability.
Protocol. Students were out®tted with a single CSA 7164
activity monitor during their regularly scheduled physical
education class. Consistent with previous studies, monitors
were attached to adjustable elastic belts and worn over the
right hip. After receiving detailed instructions regarding the
care and use of the monitors, students were instructed to
wear the CSA monitor during the waking hours for 7 con-
secutive days. At the time of distribution, students were
given a 7 day log sheet to record the times the monitor
was worn and to provide information about participation in
non-weight bearing activities such as swimming, cycling and
weight training. Upon removal of the activity monitor the
following week, stored activity counts were downloaded and
saved to a personal computer for subsequent data reduction
and analysis.
Data reduction. Minute-by-minute activity counts were
uploaded to a QBASIC data reduction program written by
the primary author for determination of time spent in
International Journal of Obesity
moderate (3 ± 5.9 METs) and vigorous (  6 METs) physical
activity during each 60 min segment of the 7 day monitoring
period. The age-speci®c count ranges corresponding to the
above intensity levels were derived from the energy expen-
diture prediction equation developed by Freedson et al:28
METs ˆ 2:757 ‡ …0:0015  counts= min†
ÿ…0:08957  age …y†
ÿ…0:000038  counts= min age …y†
In an independent sample of 80 children and adolescents
aged 6 ± 18 y, this equation accounted for 90% of the var-
iance in observed MET levels and predicted energy expendi-
ture during treadmill running and walking within  1.1
METs. The correlation between predicted MET level and
observed MET level was 0.86.28
Daily totals for participation in moderate physical activity
(MPA) and vigorous physical activity (VPA) were calculated
by summing the MPA and VPA totals from the 60 min time
blocks between 9 am and 9 pm The 12 h time interval was
selected to replicate previous monitoring studies29 ± 31 and to
control for individual differences in time spent wearing the
monitors. None of the participants were involved in sports
practices or physical education classes prior to 9:00 am. MPA
and VPA scores recorded for each day of the monitoring
period were averaged to produce an estimate of usual VPA
and MPA. Students with less than 7 days of complete mon-
itoring data (  100 000 counts on each monitoring day) were
excluded from the analyses. We have previously shown that
7 days of monitoring provides reliable estimates of daily
participation in MVPA among middle school students.32
To evaluate the contribution of `non-monitored' activities
to overall activity level, physical activity scores were calcu-
lated with and without the inclusion of self-reported parti-
cipation in swimming, cycling and weight training.
Inclusion of these data resulted in no changes to the mean
MPA and VPA scores and were not included in the analyses.
To examine patterns of physical activity behavior (ie
continuous bouts), an additional program was run to calcu-
late the weekly number of 5, 10 and 20 min bouts with
physical activity at an intensity greater than or equal to 3
METs (MVPA) Within the 20 min bouts, students were per-
mitted a `break in the action' or interruption interval of
2 min. That is, during a 20 min bout, counts were permitted
to drop below the set cut-off for up to two consecutive
minutes. For the 10 min bout, counts were permitted to
drop below the set cut-off for 1 min. No interruption interval
was permitted in the calculation of 5 min bouts.
Determinants of physical activity
Students completed a questionnaire designed to measure
hypothesized demographic, psychosocial and environ-
mental determinants of physical activity. The determinant
variables were selected from Social Cognitive Theory33 and
the Theory of Reasoned Action=Planned Behavior.34 The

Table 1 Scales used to measure hypothesized psychosocial determinants of physical
activity
Scale Cronbach's a
Self-ef®cacy 0.71
(16 items, range 1±3)
Social in¯uences 0.72
(10 items, range 1±3)
Beliefs Ð outcomes 0.67
(16 items, range 1±3)
questionnaire was administered in a classroom setting by the
primary author who read the items to students using a
standardized script. At each administration, an assistant
moved around the classroom to answer any questions and
check for students who had problems. Prior to data collec-
tion, the questionnaire was pilot tested to ensure that the
reading level and response format was appropriate for sixth
grade students.
Psychosocial variables. Hypothesized psychosocial deter-
minants of physical activity included physical activity self-
ef®cacy, social norms regarding physical activity, and beliefs
regarding physical activity outcomes. The physical activity
self-ef®cacy, social in¯uence and belief scales were modeled
on the measurement scales developed by Saunders et al.35
Responses to each item were recorded on three-point scales
(yes, no and not sure). A brief description of these scales
and their associated reliability coef®cients (Cronbach's a) is
provided in Table 1.
Environmental variables. Students completed a series of
single items designed to measure hypothesized environ-
mental determinants of physical activity behavior. Consis-
tent with Bandura's concept of the physical and social
environment,33 these included perceived physical activity
of parents and friends, access to sporting and=or ®tness
equipment at home, involvement in community physical
activity organizations, participation in community sports
teams, and self-reported hours spent watching television
or playing video games. These items were modi®ed from
Physical activity in obese and non-obese youth
SG Trost et al
825
Concept=sample items
Perceived con®dence in ability to be active:
 I think I can be physically active even if I
can watch television or play video games
instead.
 I think I can be physically active even if it is
very hot or cold outside.
 I think I can be physically active even if I
have a lot of homework.
In¯uence of family members, friends, and
teachers on physical activity behavior:
 My mother thinks I should be physically
active.
 The students in my class think I should be
physically active.
 My best friend thinks I should be physically
active.
Beliefs about consequences of being
physically active
If I were to be physically active during my
free time in the next two weeks it would:
 Get or keep me in shape.
 Make me more attractive.
 Be fun.
measures used in the National Children and Youth Fitness
Study36 and the 1990 Centers for Disease Control and
Prevention's Youth Risk Behavior Survey.37 The psycho-
metric properties of these measures have been reported
elsewhere.38,39
Statistical analysis
All statistical analyses were conducted with SAS (version
6.12). Group differences with respect to age, height, weight
and BMI were tested using independent t-tests. Differences in
the percentage of females and African American students in
the non-obese and obese groups, respectively, were tested
using a w2 test. Group differences on the physical activity and
continuous determinant variables were tested using a one-
way ANCOVA with sex and race=ethnicity serving as covari-
ates. Differences in the categorical determinants variables
were tested using a Mantel ± Haenszel w2 analyses. Statistical
signi®cance was set an a level of 0.05.
Results
Descriptive statistics for the obese and non-obese partici-
pants are shown in Table 2. Relative to their lean counter-
parts, obese children were signi®cantly heavier and more
likely to be African-American (P < 0.05). No signi®cant group
differences were observed for age, height, or percentage
female.
Figure 1 shows the means (  s.e.) for total daily counts,
daily minutes of MPA, and daily minutes of VPA. Compared
International Journal of Obesity
 Physical activity in obese and non-obese youth
SG Trost et al
826
Table 2 Descriptive statistics for obese (n ˆ 54) and non-obese Table 3 Means and percentages for the determinant variables in
(n ˆ 133) sixth grade children obese (n ˆ 54) and non-obese children (n ˆ 133)

Variable Non-obese Obese Variable Non-obese Obese P-value

Age 11.4  0.6 11.4  0.6 Self-ef®cacy 2.6  0.2 2.4  0.4 0.02
Height (m) 1.55  0.08 1.58  0.07 Social in¯uences 2.4  0.4 2.3  0.6 0.53
Weight (kg) 43.0  7.0 68.8  11.1* Beliefs Ð outcomes 2.6  0.2 2.5  0.4 0.52
BMI (kg=m2) 18.0  1.9 27.5  3.1* Community organizations 10.2  0.1 9.4  0.2 0.005
African-American (%) 49.6% 70.4% Home equipment 7.5  0.3 6.8  0.4 0.14
Male (%) 48.1% 51.8% Father active (yes) 67.7% 51.9% 0.04
Mother active (yes) 66.9% 59.3% 0.32
*P < 0.05. Best friend active (yes) 72.9% 61.1% 0.11
Sports participation (yes) 34.6% 33.3% 0.87
> 3 h TV daily 61.1% 53.4% 0.33
to non-obese children, obese children exhibited signi®cantly
lower total counts per day (28.3104  2.01104 vs
37.7104  1.41104; P ˆ 0.003); daily participation in MPA
(62.6  4.5 vs 78.2  3.2 min=day; P ˆ 0.002); and daily parti- were signi®cantly less likely report their father or male
cipation in VPA (7.1  1.3 vs 13.5  0.9 min=day; P ˆ 0.001). guardian as physically active.
Figure 2 shows the mean (  s.e.) number of 5, 10 and
20 min bouts of MVPA per week. Relative to their non-obese
counterparts, obese children exhibited signi®cantly fewer Discussion
5-min bouts (15.9  1.8 vs 23.4  1.3; P ˆ 0.001), 10 min Our ®ndings are consistent with the hypothesis that physical
bouts (8.7  1.1 vs 12.6  0.7, P ˆ 0.002), and 20 min bouts inactivity is an important contributing factor in the main-
(3.9  0.6 vs 5.8  0.4, P ˆ 0.009) of MVPA over the 7 day tenance of childhood obesity. Relative to their non-obese
monitoring period. counterparts, obese children exhibited signi®cantly lower
Group differences with respect to the hypothesized psy- daily accumulations of moderate and vigorous physical
chosocial and environmental determinants of physical activ- activity and participated in signi®cantly fewer continuous
ity are reported in Table 3. Compared to non-obese youth, 5, 10 and 20 min bouts of moderate-to-vigorous physical
obese children reported signi®cantly lower levels of physical activity. Consistent with these observations, we found sig-
activity self-ef®cacy, were involved in signi®cantly fewer ni®cant obesity-related differences in several key social-
community organizations promoting physical activity, and cognitive determinants of youth physical activity behavior.

Figure 1 Movement counts and estimated daily participation in moderate and vigorous physical activity in 133 non-obese and 54 obese children.

International Journal of Obesity


Figure 2 Weekly frequency of objectively measured 5, 10 and 20 min bouts of MVPA in 133 non-obese and 54 obese children.
A major strength of this study was the use of an objec-
tive monitoring device to quantify both the quantity and
intensity of physical activity. This is an important consid-
eration, as young children have dif®culty recalling their
past behavior accurately.15,16 Using an established algo-
rithm of converting CSA accelerometer counts to units of
absolute energy expenditure (METs),22 we were able to
assess daily participation in both moderate and vigorous
physical activity. Furthermore, because the CSA 7164 sam-
ples and stores data on a real-time basis, we were able to
evaluate weight-related differences in weekly number of
short, medium, and long bouts of moderate-to-vigorous
physical activity. The consistency of our ®ndings across all
six physical activity variables, coupled with our success in
obtaining 7 days of complete monitoring data in a rela-
tively large sample of free-living children, reinforces the
view that objective measurement devices such as accelero-
meters are useful assessment tools in the study of childhood
obesity.
Previous studies using motion sensors to assess physical
activity have failed to observe signi®cant differences
between obese and non-obese children. Romanella et al,40
observed no signi®cant differences between obese and non-
obese children after assessing physical activity for two
consecutive days with a Caltrac accelerometer. Similarly,
Wilkinson et al41 reported no signi®cant differences between
obese and non-obese 12-y-old boys after measuring physical
activity with a pedometer on a single day. The discrepancy
between our ®ndings and those of previous studies may, in
part, be attributable to the small sample sizes of earlier
Physical activity in obese and non-obese youth
SG Trost et al
827
studies, differences in the de®nition of obesity, inferior
activity monitoring technology, and the length of the mon-
itoring period. In support of the latter point, Rowlands et al42
reported signi®cant inverse correlations between physical
activity and percentage body fat after 6 days of monitoring
with the TriTrac-R3D accelerometer. In contrast, correlations
based on just one day of monitoring were substantially lower
and failed to reach statistical signi®cance.
An important ®nding of the present study was the sig-
ni®cant difference between obese and non-obese children
with respect to physical activity self-ef®cacy. This indicated
that, within our sample of sixth grade students, children
classi®ed as obese were signi®cantly less con®dent in their
ability to overcome barriers to physical activity, ask parents
to provide opportunities for physical activity, and choose
physically active pursuits over sedentary ones. According
to social-cognitive theory,33 self-ef®cacy perceptions are
derived from four principle sources of information: past
performances, vicarious experiences (modeling); verbal
persuasion; and physiological state. Therefore, to increase
perceptions of physical activity self-ef®cacy among obese
children, physical activity intervention programs should:
(1) provide enjoyable, developmentally appropriate activities
than enable overweight children to experience success (ie
emphasize moderate intensity activities such as walking); (2)
create opportunities for obese youth to observe in¯uential
others (eg parents and peers) perform physical activity; (3)
verbally encourage children to participate in physical activ-
ity (ie `you can do it'); and (4) reduce any anxiety associated
with participation in physical activity by signi®cantly
International Journal of Obesity
 Physical activity in obese and non-obese youth
SG Trost et al
828
reducing or eliminating competition or grading from
planned activities.
Previous studies have demonstrated parental physical
activity to be a salient predictor of physical activity in
normal-weight children.14,22 Few studies, however, have
examined parental physical activity in obese and non-
obese children. Fogelholm et al,43 studied the physical activ-
ity differences in obese and non-obese children as well as
parent ± child associations of obesity and physical activity.
Consistent with the results the present study, obese children
exhibited signi®cantly lower habitual physical activity scores
than non-obese children. Parental physical activity was
strongly and positively correlated to child physical activity;
however, in contrast to our own ®ndings, parental activity
level was not associated with child obesity status. This
difference is most likely attributable to our use of the
child's perception of parental physical activity level rather
than actual assessments of parent physical activity behavior.
While the signi®cance of perceptions should not be dis-
counted,44 future studies investigating parent ± child associa-
tions in physical activity and obesity status should employ
objective measures of physical activity in both parents and
children.
Some limitations of this study warrant consideration.
First, the cross-sectional nature of this study design pre-
cluded us from inferring a causal relationship between phy-
sical activity and obesity status. Thus, the lower activity
pro®le exhibited by the obese children relative to their
non-obese counterparts may have been the consequence of
obesity and not the cause of it. Second, although we were
alarmed to ®nd that approximately 30% of our sixth grade
sample could be classi®ed as obese, the modest sample size of
our obese group precluded our ability to conduct sex-speci®c
analyses. Third, because of logistic concerns, our study
included a relatively small number of determinant variables
and did not collect data directly from parents. Therefore,
future studies should expand the list of determinants to
include physical self-perception variables such as social
physique anxiety and body attractiveness; and more
comprehensive measures of parental and peer support=
encouragement for physical activity.
In summary, compared to non-obese children, obese
children exhibited signi®cantly lower daily accumulations
of moderate and vigorous physical activity and participated
in signi®cantly fewer 5, 10 and 20 min bouts of moderate-to-
vigorous physical activity. Obese children reported lower
levels of physical activity self-ef®cacy, were involved in
fewer community organizations promoting physical activity,
and were less likely to report their father or male guardian as
physically active. These ®ndings suggest that ef®cacy percep-
tions regarding exercise, awareness of, and access to, com-
munity physical activity outlets, and parental modeling of
physical activity are potential targets for physical activity
intervention programs involving overweight children.
International Journal of Obesity
References
1 Troiano RP, Flegal KM. Overweight children and adolescents:
description, epidemiology, and demographics. Pediatrics 1998;
101: 497 ± 504.
2 Ogden CL, Troiano RP, Briefel RR, Kuczmarski, Flegal KM, John-
son CL. Prevalence of overweight among preschool children in
the United States, 1971 through 1994. Pediatrics 1997; 99: E1.
3 Mei Z, Scanlon KS, Grummer-Strawn LM, Freedman DS, Yip R,
Trowbridge FL. Increasing prevalence of overweight among US
low-income preschool children. The Centers for Disease Control
and Prevention Pediatric Nutrition Surveillance, 1983 to 1995
Pediatrics 1998; 101: E12.
4 Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH. Predict-
ing obesity in young adulthood from childhood and parental
obesity. New Engl J Med 1997; 337: 869 ± 873.
5 Must A, Strauss RS. Risks and consequences of childhood and
adolescent obesity. Int J Obes Relat Metab Disord 1999; 23(Suppl 2):
S2 ± S11.
6 Dietz WH. Health consequences of obesity in youth: childhood
predictors of adult disease. Pediatrics 1998; 101: 518 ± 525.
7 Neumark-Sztainer D. The weight dilemma: a range of philoso-
phical perspectives. Int J Obes Relat Metab Disord 1999; 23(Suppl
2): S31 ± S37.
8 Goran MI. Energy expenditure, body composition and disease
risk in children and adolescents. Proc Nutr Soc 1997; 56: 195 ± 209.
9 Weinsier RL, Hunter GR, Heini AF, Goran MI, Sell SM. The
etiology of obesity: relative contributions of metabolic factors,
diet, and physical activity. Am J Med 1998; 105: 145 ± 150.
10 Bar-Or O, Baranowski T. Physical activity, adiposity, and obesity
among adolescents. Pediatr Exerc Sci 1994; 6: 348 ± 360.
11 Pate RR. Physical activity in children and youth: relationship to
obesity. Contemp Nutr 1993; 18(2).
12 Ward DS, Evans R. Physical activity, aerobic ®tness, and obesity in
children. Med Exerc Nutr Health 1995; 4: 3 ± 16.
13 Goran MI. Measurement issues related to studies of childhood
obesity: assessment of body composition, body fat distribution,
physical activity, and food intake. Pediatrics 1998; 101: 505 ± 518.
14 Kohl III HW, Hobbs KE. Development of physical activity
behaviors among children and adolescents. Pediatrics 1998; 101:
549 ± 554.
15 Pate RR. Physical activity assessment in children and adolescents.
Crit Rev Food Sci Nutr 1993; 33: 321 ± 326.
16 Kohl III HW, Fulton JE, Caspersen CJ. Assessment of physical
activity among children and adolescents. Prev Med 2000; 31:
S11 ± S33.
17 Rowlands AV, Eston RG, Ingledew DK. Measurement of physical
activity in children with particular reference to the use of heart
rate and pedometry. Sports Med 1997; 24: 258 ± 272.
18 Harro M, Riddoch C. Physical activity. In: Armstrong N, van
Mechelen W (eds). Paediatric exercise science and medicine.
Oxford University Press: Oxford; 2000. pp 77 ± 84.
19 Dietz WH. Does energy expenditure affect changes in body fat in
children? Am J Clin Nutr 1998; 67: 191 ± 191.
20 Welk GJ, Corbin CB, Dale D. Measurement issues in the assess-
ment of physical activity in children. Res Q Exerc Sport 2000;
71(Suppl): 59 ± 73.
21 Baranowski T, Anderson C, Carmack C. Mediating variable frame-
work in physical activity interventions. How are we doing? How
might we do better? Am J Prev Med 1998; 15: 266 ± 297.
22 Sallis JF, Prochaska JJ, Taylor WC. A review of correlates of
physical activity of children and adolescents. Med Sci Sports
Exerc 2000; 32: 963 ± 975.
23 Troiano RP, Flegal KM. Overweight prevalence among youth in
the United States: why so many different numbers? Int J Obes
Relat Metab Disord 1999; 23(Suppl 2): S22 ± S27.
24 Must A, Dallall GE, Dietz WH. Reference data for obesity: 85th
and 95th percentile for body mass index (wt=ht2) and tricep
skinfold thickness. Am J Clin Nutr 1991; 53: 839 ± 846.

25 Must A, Dallall GE, Dietz WH. Reference data for obesity: 85th
and 95th percentile for body mass index (wt=ht2) and tricep
skinfold thickness Ð a correction. Am J Clin Nutr 1991; 54: 773.
26 Computer Science and Applications Inc. Wrist activity monitor
users manual model 7164, 1995.
27 Trost SG, Ward DS, Moorehead SM, Watson PD, Riner W, Burke
JR. Validity of the computer science and applications (CSA)
activity monitor in children. Med Sci Sports Exerc 1998; 30:
629 ± 633.
28 Freedson PS, Sirard J, Trost SG, Dowda M, Pate RR, Sallis JF.
Calibration of a uniaxial accelerometer for estimating exercise
intensity in children and youth. Pediatr Exerc Sci (in press).
29 Gilliam TB, Freedson PS, Geenen DL, Shahraray B. Physical
activity patterns determined by heart rate monitoring in 6 ± 7-y-
old children. Med Sci Sports Exerc 1981; 13: 65 ± 67.
30 Janz KF, Golden JC, Hansen JR, Mahoney LT. Heart rate monitor-
ing of physical activity in children and adolescents: the Musca-
tine Study. Pediatrics 1992; 89: 256 ± 261.
31 Armstrong N, Balding J, Gentle P, Kirby B. Patterns of physical
activity among 11 to 16 year old British children. Br Med J 1990;
301: 203 ± 205.
32 Trost SG, Pate RR, Freedson PS, Sallis JF, Taylor WC. Using
objective physical activity measures with youth: How many
days of monitoring are needed? Med Sci Sport Exerc 2000; 32:
426 ± 431.
33 Bandura A. Social foundations of thought and action. A social
cognitive theory. Prentice-Hall: Englewood Cliffs, NJ; 1986.
34 Ajzen, I, Fishbein M. Understanding attitudes and predicting social
behavior. Prentice Hall: Englewood Cliffs, NJ; 1980.
Physical activity in obese and non-obese youth
SG Trost et al
829
35 Saunders R, Pate RR, Felton GM, Dowda M, Weinrich MC, Ward
DS, Parsons MA, Baranowski T. Development of questionnaires to
measure in¯uences on children's physical activity. Prev Med 1997;
26: 241 ± 247.
36 Ross JG, Pate RR, Caspersen CJ, Damberg CL, Svilar M. The national
children and youth ®tness study II: home and community in
children's exercise habits. J Phys Educ Rec Dance 1987; 58: 85 ± 92.
37 Heath GW, Pratt M, Warren CW, Kann L. Physical activity patterns
in American high school students. Results from the 1990 Youth Risk
Behavior Survey. Arch Pediatr Adolesc Med 1994; 148: 1131 ± 1136.
38 Pate RR, Dowda M, Ross JG. Associations between physical
activity and physical ®tness in American children. Am J Dis
Child 1990; 144: 1123 ± 1129.
39 Brener ND, Collins JL, Kann L, Warren CW, Warren BI. Reliability
of the Youth Risk Behavior Survey Questionnaire. Am J Epidemiol
1995; 141: 575 ± 580
40 Romanella N, Wakat D, Loyd B, Kelly L. Physical activity and
attitudes in lean and obese high school girls. Int J Obes 1991; 15:
407 ± 414.
41 Wilkinson P, Parkin J, Pearlson G, Strong H, Sykes P. Energy intake
and physical activity in obese children. Br Med J 1977; 1: 756.
42 Rowlands AV, Eston RG, Ingledew DK. Relationship between
activity levels, aerobic ®tness, and body fat in 8- to 10-yr-old
children. J Appl Physiol 1999; 86: 1428 ± 1435.
43 Fogelholm M, Nuutinen O, Myohanen E, Saatela T. Parent ± child
relationship of physical activity patterns and obesity. Int J Obes
Relat Metab Disord 1999; 23: 1262 ± 1268.
44 Worsley A, Coonan W, Leitch D, Crawford D. Slim and obese chil-
dren's perceptions of physical activities. Int J Obes 1983; 8: 201 ± 211.
International Journal of Obesity
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.