Journal of Physical Activity and Health, 2011, 8, 613  -625

© 2011 Human Kinetics, Inc.

Individual, Social, and Physical Environment Factors

Associated With Electronic Media Use Among Children:
Sedentary Behavior at Home
Joanna Granich, Michael Rosenberg, Matthew W. Knuiman, and Anna Timperio

Background: Individual, home social and physical environment correlates of electronic media (EM) use
among children were examined and pattern of differences on school and weekend days. Methods: Youth (n
= 298) aged 11 to 12 years self-reported time spent using EM (TV, video/DVD, computer use, and electronic
games) on a typical school and a weekend day, each dichotomized at the median to indicate heavy and light
EM users. Anthropometric measurements were taken. Logistic regression examined correlates of EM use.
Results: In total, 87% of participants exceeded electronic media use recommendations of ≤ 2 hrs/day. Watch-
ing TV during breakfast (OR = 3.17) and after school (OR = 2.07), watching TV with mother (OR = 1.96), no
rule(s) limiting time for computer game usage (OR = 2.30), having multiple (OR = 2.99) EM devices in the
bedroom and BMI (OR = 1.15) were associated with higher odds of being heavy EM user on a school day.
Boys (OR = 2.35) and participants who usually watched TV at midday (OR = 2.91) and late at night (OR =
2.04) had higher odds of being a heavy EM user on the weekend. Conclusions: Efforts to modify children’s
EM use should focus on a mix of intervention strategies that address patterns and reinforcement of TV view-
ing, household rules limiting screen time, and the presence of EM devices in the child’s bedroom.

Keywords: youth, family, television viewing, physical activity, body mass index

Electronic media (EM) use occupies the largest
amount of sedentary activity among children,1 and may
involve engagement with television (TV), videos (VHS),
DVDs, nonacademic computer use, and electronic games.
This type of EM use, especially TV viewing,2 has been
linked to the increasing problem of overweight and
obesity in youth. Although, the relationship between TV
viewing and adiposity is well established,3–6 its clinical
relevance is currently considered small.3 Nevertheless,
overall EM use may be a risk factor for the development
of chronic conditions that track into later years.7 Despite
social and educational benefits,8 TV viewing has also
been associated with low academic achievement,9 anti-
social behavior,10 sleeping problems11 and poor dietary
habits.12 Console gaming and computer and internet use
have also been shown to have both positive and adverse
health benefits among youth.13
In light of the evidence linking children’s screen time
with overweight and obesity, it has been recommended
that children should not spend more than 2 hrs/day using
Granich and Rosenberg are with the School of Sport Science,
Exercise, and Health, The University of Western Australia,
Perth, Australia. Knuiman is with the School of Population
Health, The University of Western Australia, Perth, Australia.
Timperio is with the School of Exercise and Nutrition Sciences,
Deakin University, Burwood, Australia.
EM for entertainment during daylight hours on any given
day.14,15 Of concern, however, in the US, 11- to 14-year-
old youth spend an average of 4.25 hrs/day watching TV,
videos and DVDs.16 Although most Australian studies
have reported lower average screen times [typically
between 3–4 hrs/day, of which 2–3 hrs/day constituted
TV (only) time], these studies still report estimates of
higher than 2 hrs/day on average,17–21 with boys usually
spending more time engaged with EM than girls across
childhood.17,18 In addition, screen time has been shown
to increase with age,16,18 but typically only until late
adolescence.22 Olds and colleagues23 have questioned
the assessment of screen-time recommendations, where
the adoption of different calculation methods may yield
different prevalence estimates for screen time. Neverthe-
less, by averaging daily engagement in screen time across
the week, it was found that children whose mean daily
screen time exceeded 2 hrs/day had 63% greater odds of
being overweight for their sex and age compared with
those who met current recommendations.24 Given these
odds, the possibility of those who failed to meet the rec-
ommendation having cardio-metabolic implications25 is
likely. Thus, preadolescence appears to be an opportune
time to intervene and modify EM use before significant
increases occur26 and adverse health effects develop.
As EM use mainly occurs at home, the family home
environment may be an important setting for the develop-
ment of electronic-based sedentary behavior. Most homes

613
614   Granich et al
are media-rich environments, with nearly all households
having access to a TV. In the US, the average home has
4 TV sets.27 Of the homes with multiple TV sets, 66%
have 1 TV set in a child’s bedroom.27 In Australia, 60% of
households have 2 or more TV sets28 and approximately
30% of 10 to 12 year-olds have a TV in their bedroom.17
Furthermore, 80% of Australian 10- to 14-year-olds have
at least 1 video/computer game and about 30% of children
have 3 or more electronic devices at their disposal.16,17
An EM-filled home is highly associated with screen time
and body mass index (BMI) among children.17
The family home is a setting where EM use, particu-
larly TV viewing, is prevalent and forms part of families’
day-to-day leisure time.29 Previous studies have identified
a broad range of predictors of TV viewing or overall
screen time among young children and adolescents,
including gender,2,4 sociodemographics,7,17 child’s weight
status2,7,24 and adiposity,4 availability and accessibility of
electronic devices,13,17,30 physical activity,4,7,12,17,29 poor
dietary intake,12,30 paternal BMI,7 lack of restriction
over TV use,13,17 coviewing with family and friends,29,31
eating in front of TV,29 as well as parental and sibling TV
viewing habits.29,30 However, few studies have examined
these influencing factors simultaneously.32–35 Even fewer
have studied these influencing factors within the context
of a school and a weekend day. It is possible that differ-
ent factors may operate to influence EM on school and
weekend days and such information is important for the
development of interventions. This study aimed to exam-
ine associations between individual, and home social
and physical electronic environment factors and EM use
among 11- to 12-year-old children. Identifying correlates
of heavy EM use is likely to facilitate the development of
specific strategies to limit excessive EM use in children.
Methods
Design
In this cross-sectional study, a self-administered question-
naire and objective anthropometric measurements were
used to collect data from 11- to 12-year-olds (Grade 7).
The data were collected between October and December
2004. The study was approved by school Principals and
The University of Western Australia Human Research
Ethics Committee.
Sample Selection and Recruitment
A postcode36 stratified random sample of 15 primary
(elementary) schools from metropolitan Perth, Western
Australia, were selected and all 11- to 12-year-olds (n =
553) at each of these school invited to participate in this
study. Participation was voluntary. The return of a signed
informed consent form by the student and their primary
carer was a requirement for participation and students
nominated their level of participation: questionnaire (54%
response rate) and/or anthropometric measurements (49%
response rate).
Measures and Test-Retest Reliability
The development of the questionnaire was based on
data generated from a qualitative study,37 items used in
a descriptive study investigating children’s leisure activi-
ties17,38 as well as literature on sedentary behavior and
socioecological constructs.32,33 Test-retest reliability was
established by administering the survey twice, 10 days
apart among a subsample of 143 adolescents. Intraclass
correlation coefficients (ICC) were calculated for ordinal
and continuous variables,39 while Kappa values and per-
cent agreement were calculated for nominal variables.40
Electronic Media Use
Children were asked to report the total time per day (in
hours or minutes) they spent in 5 EM activities on a
typical school day (outside school hours) and a typical
weekend day: TV viewing, watching videos/DVD, using
a computer for fun, using a computer for homework, and
playing electronic games (hand-held and those that hook
up to a TV set). This instrument was adapted from an
existing measure.41 Total duration of EM use per school
day and per weekend day was computed. The test-retest
reliability (ICC) for total EM use was 0.78 for a typi-
cal school day and 0.67 for a typical weekend day. The
ICC’s for individual sedentary activities ranged from
0.51 to 0.76.
Individual Measures
Participants were asked how frequently (ICC = 0.58)
and for how many hours (ICC = 0.84) they usually par-
ticipated in physical activity (PA) outside school hours
(PA defined as ‘very active and out of breath or sweat’).
Participants were also asked where they usually watched
TV (own bedroom, other bedroom, dining/meals room,
living/lounge/theater room: Kappa = 0.49; percent
agreement = 86%) and when they usually watched TV
on a typical school day (before and straight after school,
during dinner, after dinner, late at night, Kappa = 0.41
to 0.72; percent agreement = 80% to 87%) and weekend
day (morning, lunchtime, afternoon, during dinner,
after dinner, late at night: Kappa = 0.45 to 0.58; percent
agreement = 73% to 83%). Children were asked what
time they usually get home from school and responded
by providing the hour and minutes (ICC = 0.80). Outside
of main meals, snacking while watching TV was also
measured with a yes or no response category (Kappa =
0.76; percent agreement = 92%). Perceptions regarding
the amount of TV they watched (ICC = 0.60) and if a
sitter (other than a family member) were usually present
at home when they returned home from school (Kappa
= 0.56; percent agreement = 76%) were also assessed in
the questionnaire.
Anthropometric measures were obtained by Level
1 ISAK (International Society for the Advancement of
Kinanthropometry) accredited technicians.42 All mea-
surements (height, weight, waist circumference, hip
circumference, and skinfolds) were taken using standard

protocols according to ISAK. Anthropometric equipment
used included a portable stadiometer, a weighing scale,
girth tape and Slimline skinfold calipers. Participants’
BMI [weight(kg)/height(m)2] were calculated and clas-
sified as acceptable weight, overweight, or obese, based
on international sex-and age-specific cut-off points.43 The
girth measures were taken twice, and mean girth values
were calculated for each child. Four skinfold thicknesses
(triceps, subscapular, suprailiac, and midabdominal) were
also measured twice and the mean computed for each
individual skinfold. All skinfold measurements were then
summed to compute a total skinfold score.
Home Social Environment Measures
Participants were asked how frequently their significant
others (ie, mom, dad, sister/s, and brother/s) watched TV,
played electronic games and played computer games and
how frequently they themselves watched TV with each
of these family members. Response options were never,
sometimes, often, and doesn’t apply (ICC = 0.40–0.67).
Participants were asked separate questions regarding
whether they (Kappa = 0.39–0.74; percent agreement =
79%–88%) and their family (Kappa = 0.39–0.52; per-
cent agreement) watch TV during breakfast, lunch, and
dinner, respectively. Participants were also presented with
several EM-based rules and restrictions and asked which
of these rules applied in their home (ICC = 0.42–0.67).
Perceptions about the frequency household members
encourage them to watch less TV were also assessed
(response options: never, sometimes, often, doesn’t apply
[ICC = 0.46–0.60]).
Home Physical Electronic Environment
Measures
Participants were presented with a list of 8 electronic
devices, including pay TV and internet service, and
asked which of these they had in their home (Kappa
= 0.68–0.83; percent agreement = 91%–99%) and in
their bedroom (Kappa = 0.70–0.89; percent agreement
= 93%–98%).
Sociodemographics Measures
Demographic items included gender, age, language
mostly spoken at home, residential suburb, parents’
education, family composition, and family living arrange-
ment. Maternal and paternal highest level of education
was categorized as tertiary education (university),
postsecondary education (TAFE—technical and further
education vocations), high school (completed certificate),
and primary (elementary) school (completed certificate).
The residential postcode was used to generate a proxy
for socioeconomic status (SES) using the Index of Rela-
tive Socio-Economic Disadvantage;36 participants were
ranked into tertiles of SES (low, medium, and high) based
on scores for their postcode.
Electronic Media Use Among Children   615
Data Analyses
Descriptive statistics [mean and (±SD)] were calculated
for each of the 5 different types of EM behaviors (Table
1). A sum of the (5) individual EM components generated
a composite score for which a highly skewed distribution
containing some implausible values (>12hours per day)
was found. So, the composite score was then truncated at
the 95% percentile to reduce the effect of outliers and a
mean (±SD) for total EM use calculated (Table 1). Given
this, the median was used to dichotomize the sample into
heavy and light EM users (Table 2). The BMI variable
was also skewed and thus, truncated at the 95th percentile.
To allow for the inclusion of as many children as pos-
sible in the multivariate analyses, simple imputation of
missing data (for 75 participants) was employed where
≤ 20% of values for a variable of interest were missing.
The mean was used to replace missing values for continu-
ous variables and the mode was used to replace missing
values for categorical variables, enabling a total of 298
participants to be included in all analyses. Other statistics
are presented as percentage and the strength of the few
bivariate associations between continuous variables (ie,
BMI and TV mins/day) are shown as Pearson’s correla-
tion (r). Differences between heavy and light EM user
groups for typical school and typical weekend days were
assessed using the t test for quantitative variables and
the Pearson’s chi-square test for categorical variables.
Logistic regression modeling with a backward variable
elimination strategy was used to find the significant and
independent predictors of heavy EM use separately for the
individual, social and home physical environment blocks
of variables. The process commenced with a multivariate
model containing all variables showing significant differ-
ences (P < .05) between heavy and light EM use (from
Table 2), and then eliminating nonsignificant (P > .05)
variables in stepwise fashion. The retained variables from
each block were then included together in a single final
multivariate model and estimated odds ratios presented
to show the strength of their associations with heavy EM
use. Data analyses were performed using SPSS V15.0
for Windows.
Results
Demographic Characteristics
The average age of participants was 12.53 (SD ± .36)
years. Gender was almost equal (147 boys and 151 girls)
and 89% of participants came from families that spoke
English at home. The majority of mothers (51%) and
fathers (55%) had a tertiary (university) education. Thirty
percent of participants resided in low, 30% in middle,
and 40% in high socioeconomic areas. Most children
(67%) lived in households with 2 parents and 54% had
2 or more siblings.
 Table 1  Mean (SD) Duration (Mins/day) of Different Types of Electronic-Based Sedentary Pursuits
by Adolescents on a School and a Weekend Day
All n = 298 Boys n = 147 Girls n = 151
Type of electronic pursuit Mins/day ± SD Mins/day ± SD Mins/day ± SD P
Total electronic media use#‡
  School day 290 ± 178 319 ± 197 260 ± 151 0.00**
  Weekend day 351 ± 200 400 ± 210 301 ± 176 0.00**
TV (only) viewing
  School day 138 ± 72 142 ± 77 133 ± 67 0.27
  Weekend day 148 ± 98 160 ±104 136 ± 90 0.04*
Video/DVD (only) viewing
  school day 93 ± 67 80 ± 63 111 ± 68 0.01**
  weekend day 114 ± 59 113 ± 60 115 ± 58 0.85
Computer use for fun
  School day 71 ± 63 85 ± 71 57 ± 50 0.00**
  Weekend day 90 ± 77 104 ± 86 75 ± 62 0.01**
Computer use for homework
  School day 56 ± 36 51 ± 35 60 ± 36 0.07
  Weekend day 54 ± 38 50 ± 40 57 ± 36 0.30
Electronic game playing
  School day 71 ± 65 80 ± 69 49 ± 50 0.01**
  Weekend day 88 ± 69 103 ± 72 58 ± 53 0.00**
# Composite measure of TV, Video/DVD viewing, and computer use for fun (playing games and using the internet to surf and chat with others)
and homework (word processing and researching on the internet), and playing electronic game systems including portable/hand-held, game cubes/
consoles, and games on mobile phones.
‡ Truncation (@ 95th percentile) applied to the total electronic media use score only.

* t test analysis; significant differences between boys and girls, P < .05.
** t test analysis; significant differences between boys and girls, P < .01.

Table 2  Factors Associated With Level of Electronic Media Use on a School and Weekend Day; Table
Shows Mean for Quantitative Factors and Percent for Categorical Factors
Type of electronic media (EM) users
School day Weekend day
Light user Heavy user Light user Heavy user
Factors an = 155% an = 143% P an = 155% an = 143% P
Individual
  Gender
    Boy 44.5 54.5 0.08 39.4 60.1 0.00**
    Girl 55.5 45.5 60.6 39.9
  Socio-Economic Status (SES)+
    Low SES 16.8 44.1 0.00** 23.9 36.4 0.05
    Middle SES 29.0 28.7 30.3 27.3
    High SES 54.2 27.3 45.8 36.4
  Frequency of PA outside school hours
    ≤1 time per week 11.0 22.4 0.00** 14.2 18.9 0.15
    2–3 times per week 31.0 37.1 31.0 37.1
    4–6 times per week 40.0 23.8 38.1 25.9
    Everyday 18.1 16.8 16.8 18.2
  Duration of PA outside school hours
    ≤1 hour per week 12.3 23.1 0.13 16.1 18.9 0.51
    2 hours per week 16.8 14.7 13.5 18.2
    3 hours per week 14.8 16.8 18.7 12.6
    4 hours per week 15.5 12.6 13.5 14.7
    5 or more hours per week 40.6 32.9 38.1 35.7
616 (continued)
Table 2 (continued)
Type of electronic media (EM) users
School day Weekend day
Light user Heavy user Light user Heavy user
Factors an = 155% an = 143% P an = 155% an = 143% P
  Where child usually watches TV
    Own bedroom 7.1 16.1 0.01* 8.4 14.7 0.08
    Elsewhere in the house 92.9 83.9 91.6 85.3
  Time (pm) when child gets home from school 3.52 3.35 0.01* - - -
  Sitter/carer (other than a family member) present
at home when child gets home from school
    Yes 0.6 7.0 0.00** - - -
    No 99.4 93.0 - - -
  Period of the day when child usually watches TV
on a school day
    Before school 37.1 62.9 0.00** - - -
    After school 40.2 59.8 0.00** - - -
    During dinner 42.2 57.8 0.00** - - -
    After dinner 51.1 48.9 0.55 - - -
    Late at night 34.2 65.8 0.00** - - -
  Period of the day when child usually watches TV
on a weekend day
    Morning - - - 50.3 49.7 0.48
    Midday - - - 30.4 69.6 0.00**
    During dinner - - - 46.0 54.0 0.07
    After dinner - - - 39.8 60.2 0.00**
    Late at night - - - 48.2 51.8 0.00**
  Snacking while watching TV (outside of main meals)
    Yes 67.1 79.7 0.01* 67.1 79.7 0.01*
    No 32.9 20.3 32.9 20.3
  Child’s perception about their quantity of TV viewing
    Right amount 57.0 43.0 0.00** 55.1 44.9 0.04*
  BMI 19.11 20.09 0.00** 19.41 19.77 0.33
  Waist circumference (cm) 63.90 65.84 0.01* 64.69 64.99 0.70
  Hip circumference (cm) 80.70 82.21 0.18 81.94 80.85 0.33
  Skinfold score# (mm) 49.01 54.17 0.04* 52.02 50.90 0.66
Social
  TV coviewing with mom
    Sometimes 65.5 53.6 0.04* 56.3 45.2 0.06
    Often 34.5 46.4 43.7 54.8
  TV coviewing with dad
    Sometimes 65.2 62.8 0.67 67.6 60.3 0.21
    Often 34.8 37.2 32.4 39.7
  TV coviewing with brother/s^
    Sometimes 41.7 45.0 0.63 46.0 40.2 0.39
    Often 58.3 55.0 54.0 59.8
  TV coviewing with sister/sXXX
    Sometimes 49.4 53.3 0.60 50.6 52.3 0.82
    Often 50.6 46.7 49.4 47.7
  Child (only) usually watches TV
during a main meal
    Breakfast 32.7 67.3 0.00** 40.2 59.8 0.00**
    Lunch - - - 33.3 66.7 0.00**
    Dinner 45.2 54.8 0.02* 50.0 50.0 0.37
(continued)
617
Table 2 (continued)
Type of electronic media (EM) users
School day Weekend day
Light user Heavy user Light user Heavy user
Factors an = 155% an = 143% P an = 155% an = 143% P
  Family usually watches TV during a main meal
    Breakfast 27.9 72.1 0.00** 36.1 63.9 0.00**
    Lunch 38.2 61.8 0.01** 42.6 57.4 0.08
    Dinner 45.6 54.4 0.00** 48.5 51.5 0.15
  Household rules about EM use
    No TV before school only 69.4 30.6 0.00** - - -
    No TV after school only 66.7 33.3 0.00** - - -
    No TV before homework (hw) 61.5 38.5 0.00** 54.6 45.4 0.42
    No late night TV 68.5 31.5 0.00** 53.7 46.3 0.66
    Time limit on portable e-games 77.4 22.6 0.00** 64.2 35.8 0.00**
    Time limit on TV-based e-games 61.8 38.2 0.00** 48.3 51.7 0.39
    Time limit on PC use (minus hw) 66.0 34.0 0.00** 55.7 44.3 0.34
    Time limit on playing PC games 69.8 30.2 0.00** 59.4 40.6 0.08
    No PC or e-games before hw 60.0 40.0 0.00** 51.1 48.9 0.49
  Frequency dad plays e-games
    Never 84.7 69.0 0.00** 85.3 68.7 0.00**
    Often‡ 15.3 31.0 14.7 31.3
  Frequency mom plays e-games
    Never 88.2 86.8 0.73 90.0 85.0 0.24
    Often‡ 11.8 13.2 10.0 15.0
  Frequency sister/sXXX plays e-games
    Never 55.9 35.9 0.01** 52.9 38.2 0.07
    Often‡ 44.1 64.1 47.1 61.8
  Frequency brother/s^ plays e-games
    Never 63.6 50.6 0.07 59.1 55.8 0.64
    Often‡ 36.4 49.4 40.9 44.2
Physical
  Number of electronic screen-based devices present
in child’s bedroom
    None 38.5 0.00** 58.1 44.1 0.01**
    1 21.7 22.6 20.3
    2 14.7 6.5 14.7
    3 or more 25.2 12.9 21.0
  Internet access at home
    Yes 78.3 0.00** 81.9 86.7 0.25
    No 21.7 18.1 13.3
  Pay TV access at home
    Yes 27.3 0.42 21.3 29.4 0.10
    No 72.7 78.7 70.6
  Pay TV access in child’s bedroom
    Yes 1.4 0.71 0.0 3.5 0.01**
    No 98.6 100.0 96.5
a Type of electronic media user cut-points are based on this sample’s median for total time (mins) in EM. The cut point for a light EM user on a typical school day

is < 240.00 mins/day and a heavy EM user is > 240.01 mins. The cut-point for a light EM user on a typical weekend day is < 320.00 mins and for heavy EM user
it is >320.01 mins. * Significant differences between light and heavy EM users, P < .05. ** Significant differences between light and heavy EM users, P < .01.
+ Based on residential postcodes and classified according to SEIFA.36 # The skinfold score is a sum of 4 skinfold thickness’ (mm) taken on triceps, subscapular,

suprailiac and midabdominal body sites. ^ Only children who had a brother/s were used in this cross-tabulation. XXX Only children who had a sister/s were used
in this cross-tabulation. ‡ Collapsed category of ‘sometimes’ and ‘often’ as few response counts had less than 5 cases violating chi-square test assumption.

618

Electronic Media Use
Participants engaged in EM use for an average of 290
mins/day on a typical school day and 351 mins/day on a
typical weekend day, with boys spending more time than
girls in total EM use on both days (Table 1). Boys spent
more time than girls using a personal computer for fun
and playing electronic games on a school and a weekend
day. Boys also watched more TV on a typical weekend
day than girls. Overall, 84% of children exceeded the
guidelines for EM use (2 hrs/day)14,15 on a school day
and 90% on a typical weekend day.
Associations of Individual, Home
Social, and Physical Factors With Heavy
Electronic Media Use
Table 2 shows univariate associations between individual,
home social, and physical environment variables and EM
use. Heavy and light EM users on a typical school day
and a typical weekend day differed on many individual,
home social, and physical environment variables included
in this study. In addition, data examined within gender
groups (data not shown in Table 2) showed significant
differences in mean BMI between light and heavy EM
users among girls (18.81 vs. 21.06, P = .001) but not
among boys (19.30 vs. 19.39, P = .865). A greater pro-
portion of female heavy EM users (37%) were found to
be overweight or obese compared with light EM users
(13%). Boys average waist girth measured 64.7cm
and girls 65.0cm. Based on the 90th percentile waist
circumference for sex and age (12 yrs),44 24% of girls
and 10% of boys within the heavy EM user groups may
be at an elevated health risk of developing risk factors
that may be associated with metabolic syndrome45 and
cardiovascular disease.46
A weak but significant correlation between BMI and
total EM use per day (r = .13, P = .029) was also found by
this study. However, no significant correlations between
total time (mins/day) spent in front of TV (including
video and DVD) and various adiposity indices such as
BMI (r = .03, P = .599), WC (r = .02, P = .659) and
skinfold score (r = .04, P = .945) were found.
Table 3 shows the significant variables from Table
2 that retained significance in the multivariate analysis.
The significant and independent determinants of heavy
school day EM use were usually watching TV after
school (OR = 2.07), having a sitter present at home when
they returned home from school (OR = 18.85), having a
higher BMI (for every 1 unit increase in BMI, the odds
of adolescent being a heavy EM user increased by 15%),
often coviewing TV with their mother on a school day
(OR = 1.96), usually watching TV during breakfast (OR
= 3.17), having a household rule prohibiting late night
TV (OR = 0.32), not having a household rule limiting
the time spent on personal computer games (OR = 2.30),
and having 2 or more EM screen-based devices in the
bedroom (OR = 3.0). The significant and independent
Electronic Media Use Among Children   619
determinants of heavy weekend-day EM use were being
a boy (OR = 2.35), and usually watching TV around
midday (OR = 2.91), late at night (OR = 2.04) and during
lunch (OR = 1.87).
Discussion
In this study, individual, family social, and physical home
factors were compared between 11- to 12-year-olds who
were heavy and light EM users. Total EM use estimates
for this study were significantly higher than reported
elsewhere,17,18,20,34,47–49 however the overall findings
are consistent with many studies showing that a high
proportion of children exceed current recommendations
for EM use16–18,26,29,31,47,48 and that high EM use may
contribute to risk of adiposity.12,18,24,47,49–51 Together,
these findings highlight a need for action to reduce EM
use among youth. This study also identified a pattern of
difference in EM use by days of the week (school and
weekend days respectively) and by individual, social, and
physical home-based factors that independently predicted
heavy EM use.
On a school day, the before and after-school period
was the most popular time to watch TV in addition to TV
viewing with a female carer (typically mother) and having
someone (a nonfamily member) home after school for
heavy EM users. Furthermore, the presence of 2 or more
EM screen-based devices in the child’s bedroom, home
rules prohibiting late night TV viewing and time limits
on personal computer games together with an increased
BMI were also independent predictors of heavy EM use
on a school day.
This study also found that EM use pattern differed
by gender where boys spent more time than girls in front
of screens, particularly on weekends, a finding consistent
with other studies.17,18,20,34,47,48 The likelihood of being a
heavy EM user on the weekend was also greater among
boys. This suggests that girls may have other interests
or preferences for other types of leisure activities apart
from EM use, although it is possible that these may be
still sedentary (ie, sitting and chatting to friends),17,37
given that girls typically accumulate less physical activ-
ity than boys.52
All of these influencing factors (summarized in
Figure 1) are important targets for youth-based interven-
tions and are broadly consistent with socioecological32
and social cognitive33 frameworks where individual fac-
tors, family social norms, as well as the home physical
environment appear to influence discretionary EM use
at home. However, overall, there were few correlates of
weekend EM use compared with school day EM use and
these tended to be related to TV viewing times (at midday,
during lunch and late at night) and not to individual or
family-related factors.
The results regarding relationships between physical
activity participation and patterns of EM use by type of
day suggested that both sedentary and physical activity
behaviors can coexist, an assumption previously raised
Table 3  Multivariate Model With Significant and Independent Individual, Home Social, and
Physical Environment Predictors of Heavy Electronic Media Use on a School and a Weekend Day
Factors n Odds ratio 95% CI P
School day
  Period of day when child usually watches TV—after school
    No 107 1.00†
    Yes 191 2.07 1.15–3.72 0.01
  Sitter/carer (other than a family member) present when child gets home
from school
    No 287 1.00†
    Yes 11 18.85 1.94– 0.01
182.73
  BMI 298 1.15 1.03–1.25 0.00
  TV coviewing with mom
    Sometimes 183 1.00†
    Often 115 1.96 1.10–3.49 0.02
  Child (only) usually watches TV during main meal—breakfast
    No 191 1.00†
    Yes 107 3.17 1.77–5.67 0.00
  Household rule about EM use—no late night TV
    No 190 1.00†
    Yes 108 0.32 0.17–0.59 0.00
  Household rule about EM use—time limit on PC games
    Yes 96 1.00†
    No 202 2.30 1.24–4.28 0.00
  Frequency sister/s plays e-games
    Never 66 1.00†
    Often 80 2.03 0.93–4.43 0.07
  Number of electronic screen-based devices present in child’s bedroom
    None 153 1.00†
    1 64 1.22 0.61–2.47 0.56
    2 31 3.06 1.17–8.00 0.02
    3 or more 50 2.99 1.34–6.70 0.00
Weekend day
  Gender
    Girl 151 1.00†
    Boy 147 2.35 1.43–3.87 0.00
  Period of the day when child usually watches TV—midday
    No 206 1.00†
    Yes 92 2.91 1.65–5.14 0.00
  Period of the day when child usually watches TV—late at night
    No 53 1.00†
    Yes 245 2.04 1.04–4.00 0.03
  Child (only) usually watches TV during main meal—lunch
    No 232 1.00†
    Yes 66 1.87 0.99–3.53 0.05
†Referent group = light EM users. The cut-point for a light EM user on a typical school day is < 240.00 mins/day and a heavy EM user is > 240.01
mins. The cut-point for a light EM user on a typical weekend day is < 320.00 mins and for heavy EM user it is > 320.01 mins.

620
 Electronic Media Use Among Children   621

Figure 1 — Factors influencing children’s electronic-media use at home. * Variables showed a significant association with heavy
electronic media use at bivariate level (P < .05). ‡ Variables showed a significant association with heavy electronic media use at
multivariate level and adjusted for all variables significant (P < .05) in Table 2.

elsewhere.37,53,54 Frequency of physical activity (but not
duration) was associated with EM use in the unadjusted
analysis. However, in the adjusted model physical activity
was no longer significant. This adds weight to the notion
that sedentariness and physical activity are independent
behaviors.55 Given that children who watched TV after
school were more than twice as likely to be heavy EM
users and that physical activity guidelines specifically
recommend that EM use for entertainment be limited
during daylight hours,15 and with the greatest range of
622   Granich et al
alternative activities (active or otherwise) is likely to be
available during this time period, the after-school period
is likely to be important. Focusing intervention efforts to
reduce children’s EM use during this period of the school
day may therefore be opportune.
Despite school days posing a constraint on children’s
leisure time, this study found a distinct pattern of EM use
where on a school day, heavy EM users were character-
ized by watching TV after school and also late into the
night. The presence of electronic devices in the bedroom
appeared to be a consistent correlate of heavy screen-time
use during a school day. Together these factors demon-
strate that encouraging parents to either remove or avoid
placing TVs and other electronic devices in children’s
bedrooms, and providing alternative opportunities to
screen-based activities, particularly after school, may
by important. Removing EM equipment from children’s
bedrooms may also reduce the likelihood of watching TV
late into the night and may allow parents to have more
control over their children’s EM use. In support of this,
rules restricting children’s TV viewing late into the night
and placing time limits on personal computer games were
independently associated with lower likelihood of school
day heavy EM use. This highlights that parental guidance
and enforcement of rules is likely to be an important part
of efforts to reduce children’s EM use. Previous research
has also found associations between home rules, EM use,
and children’s screen-based behavior.13,17,34,38,61 House-
hold rules, however, were not associated with EM use
on the weekend in the multivariate model in this study.
This may be because rules are less likely to be enforced
by parents on weekends, which may be viewed as more
leisurely and as ‘time-out.’ This assumption is supported
by the greater amount of EM use reported by participants
on a weekend day compared with a school day. While
introducing household rules regarding EM use is impor-
tant, upholding such rules in these circumstances may
be challenging. Therefore, a consistent enforcement of
rules may be necessary for the sustainability of home
EM use rules.
A similar pattern of TV viewing during meal times
was observed by school (breakfast) and weekend (lunch)
days among heavy EM users. This presents as another
opportunity for parents to reduce their children’s screen-
based behavior. Watching TV during meals has been
linked with overweight and obesity.29,50 Although TV
viewing during meals may be a secondary task, it may
contribute to high overall EM use by leading to extended
viewing before and after meals. Meal time viewing may
also encourage viewing at other times since TV and
what type of program is on TV can often be a point of
distractibility.62 Encouraging parents to remove TV as a
focal point during meal times may be an important tool
for further reducing children’s screen time.
Coviewing of TV with mother but not other family
members (ie, father or siblings) was found to be inde-
pendently associated with heavy EM use on a school day,
possibly due to the reinforcing nature of the behavior.
Role modeling and reinforcement of EM use by other
family members were not significant in the final multivari-
ate model. This is inconsistent with previous studies29,30,34
that have found role modeling and/or behavioral rein-
forcement by family members (ie, parents and siblings)
in general to be correlates of screen-related sedentary
behavior, so this is worthy of further exploration.
In this study, BMI was the only index of adiposity
significantly associated with heavy EM use in the mul-
tivariate model (for school day). This finding provides
additional support for the need to intervene to reduce
the amount of time children spend using EM for enter-
tainment given that overweight children are likely to
remain overweight as they age,56,57 and the significant
health implications associated with overweight during
youth5,25,45,58,59 and later in life.7,60 Compared with other
studies investigating correlates of EM use in youth3,29,35
this study examined the widest range of anthropomet-
ric measures (BMI, waist and hip circumference, and
skinfold score) with EM use. Although, no relationship
between TV viewing and most of the body fatness indices
apart from BMI was found (data not shown), the signifi-
cant relationship between BMI and EM is consistent with
other studies3,6 and supports the notion that TV viewing
alone may not be an ideal single marker of sedentariness.3
The high screen time found by this study is also
alarming but notably, this study is one of few17,20,21 that
has comprehensively included a range of different screen-
based activities. This study measured 5 distinct types of
sedentary screen-based activities (ie, TV only, Video/
DVD, computer use for fun, electronic games) including
computer use for homework as part of the overall EM use
prevalence estimate. Therefore, it is plausible that this
may accounted for the difference in time spent with EM
compared with other studies18,47–49 as it captured greater
range of electronic mediums that children engage with
and thus higher amount of screen time.
Limitations of this study included the cross-sectional
study design (which limits inferences about causality) and
the reliance on self-report measures that had generally
not been validated (which may result in bias). It is also
plausible that the general high level of EM use reported
in this study may have been overestimated by some
children and it is possible that the truncation of extreme
values at the 95th percentile to reduce their influence
in the bivariate and multivariate analysis may not have
fully accommodated this over-reporting bias. In addition,
quarter of the sample had missing data for variables of
interest, necessitating the imputation of missing values so
as to minimize the impact of missing data in the overall
analyses. This has been accommodated using a simple
imputation method. Further, numerous statistical tests
were conducted in this study which may have led to bias
and false-positive errors in the bivariate analysis. How-
ever, this is unlikely at the final multivariate level as most
of the variables that remained were significant at P < .01.

The sample included only 11- to 12-year-olds and was
skewed toward higher parental education and area-level
SES, and the active consent process may have resulted
in a sample with a greater interest in the topic than the
broader population, all of which may limit the generaliz-
ability of the results. Finally, the current study focused
on the influence of individual factors and the immediate
social and physical family home environment factors on
EM use. It did not consider whether the social environ-
ment variables (apart from time of day and mealtime
viewing patterns) differed on weekdays and weekends,
nor did it consider broader external factors which may
also influence EM use, such as the characteristics of the
local neighborhood.63
Electronic media has a strong presence in family
homes and EM use comprises a large proportion of
children’s leisure time. This study confirmed the per-
vasiveness of EM use among youth. The results of this
study indicated that factors within the family home are
important influences on children’s EM use and specifi-
cally suggests that intervention strategies should focus on
reducing EM use in the before- and after-school periods,
as well as late in the evenings. Moreover, introducing
and maintaining home rules and limits around TV and
other EM use throughout the week, particularly rules
prohibiting TV during meal times as well as minimizing
access to electronic devices in the bedroom, are likely to
be important. Modifying paternal behaviors that might
reinforce TV viewing such as maternal coviewing of TV
may also be important. Further research should include
longitudinal designs and consider including other seden-
tary pursuits, particularly social sedentary behaviors such
as sitting and chatting to friends and using telephones.
The mixed use of electronic devices such as mobile
phones with internet capabilities and connectivity, and
other emerging technologies, may further contribute to
sedentary screen time and are therefore worthy of study.
Given that the results of this study lend themselves to the
development of intervention strategies, future research
may also explore children’s self-efficacy to modify
electronic-based sedentary behavior, self-regulation of
EM use, goal setting strategies and decision making
about alternative nonscreen based leisure activities, as
well as parental behavioral capability to modify chil-
dren’s EM use.
Acknowledgments
This research was funded by the Western Australian Health Pro-
motion Foundation (Healthway). JG was supported by a Health
Promotion Research Scholarship from Healthway and AT was
supported by a Public Health Research Fellowship from the Vic-
torian Health Promotion Foundation. The authors acknowledge
the contribution of all Health Science and Human Movement
undergraduate and postgraduate students who assisted in the
collection of the anthropometric data. The authors would also
like to thank all primary (elementary) school students and
teachers who were involved in this study.
Electronic Media Use Among Children   623
References
1. Australian Bureau of Statistics. Children’s participation in
cultural and leisure activities. Canberra, Australia: ABS;
April 2006. (cat.no. 4901.0).
2. Kautiainen S, Koivusilta L, Lintonen T, Virtanen SM,
Rimpela A. Use of information and communication tech-
nology and prevalence of overweight and obesity among
adolescents. International Journal of Obesity & Related
Metabolic Disorders. 2005;29(8):925–933.
3. Marshall S, Biddle S, Gorely T, Cameron N, Murdey
I. Relationships between media use, body fatness and
physical activity in children and youth: a meta-analysis.
International Journal of Obesity Related Metabolic Dis-
orders. 2004;28(10):1238–1246.
4. Dollman J, Ridley K. Differences in body fatness, fat pat-
terning and cardio-respiratory fitness between groups of
Australian children formed on the basis of physical activ-
ity and television viewing guidelines. J Phys Act Health.
2006;3:191–199.
5. Zoeller RF, Jr. Physical activity, sedentary behavior, and
overweight/obesity in youth: evidence from cross-sec-
tional, longitudinal, and interventional studies. American
Journal of Lifestyle Medicine. 2009;3(2):110–114.
6. Hesketh K, Wake M, Graham M, Waters E. Stability of
television viewing and electronic game/computer use in a
prospective cohort study of Australian children: relation-
ship with body mass index. Int J Behav Nutr Phys Act.
2007;4(60): .
7. Hancox RJ, Milne BJ, Poulton R. Association between
child and adolescent television viewing and adult
health: a longitudinal birth cohort study. The Lancet.
2004;364(9430):257–262.
8. Close R. Television and language development in the early
years: a review of literature. London: National Literacy
Trust; 2004.
9. Hancox RJ, Milne BJ, Poulton R. Association of television
viewing during childhood with poor educational achieve-
ment . Archives of Pediatrics and Adolescent Medicine.
2005;159(7):614–618.
10. Bickham DS, Rich M. Is television viewing associated
with social isolation?: roles of exposure time, viewing
context, and violent content . Archives of Pediatrics and
Adolescent Medicine. 2006;160(4):387–392.
11. Johnson JG, Cohen P, Kasen S, First MB, Brook JS. Asso-
ciation between television viewing and sleep problems
during adolescence and early adulthood . Archives of Pedi-
atrics and Adolescent Medicine. 2004;158(6):562–568.
12. Salmon J, Campbell KJ, Crawford DA. Television viewing
habits associated with obesity risk factors: a survey of Mel-
bourne school children. Med J Aust. 2006;184(2):64–67.
13. Sharif I, Sargent J. Association between television, movie,
and video game exposure and school performance. Pedi-
atrics. 2006;118(4):e1061–e1070.
14. American Academy of Pediatrics Committee on Public
Education. AAP: children, adolescents and television.
Pediatrics. 2001;107:423–426.
15. Department of Health and Ageing. Active kids are healthy
kids: Australia’s physical activity recommendations for
5-12 year olds. Canberra, Australia: ACT; 2004.
16. Roberts DF, Foehr UG, Rideout V. Generation M: media
in the lives of 8-18 year-olds: The Henry J Kaiser Family
Foundation: 2005.
624   Granich et al
17. Salmon J, Telford A, Crawford D. The Children’s Leisure
Activities Study (CLASS): summary report. Melbourne,
Australia: Centre for Physical Activity and Nutrition
Research: Deakin University; 2004.
18. Booth M, Okely A, Denney-Wilson E, Hardy L, Yang B,
Dobbins T. NSW Schools Physical Activity and Nutrition
Survey (SPANS) 2004: summary report. Sydney, Australia:
Department of Health; 2006.
19. Population Health Research Centre (ACT). The report on
the ACT Year 6 Physical Activity and Nutrition Survey.
Canberra, ACT. Australian Capital Territory: Government;
2007.
20. Olds T, Ridley K, Dollman J. Screenieboppers and extreme
screenies: the place of screen time in the time budgets
of 10-13 year-old Australian children. Aust N Z J Public
Health. 2006;30(2):137–142.
21. Department of Health and Ageing. 2007 Australian
National Children’s Nutrition and Physical Activity
Survey: main findings. Canberra, Australia: Common-
wealth of Australia; 2008.
22. Olds T, Wake M, Patton G, et al. How do school-day activ-
ity patterns differ with age and gender across adolescence?
J Adolesc Health. 2009;44(1):64–72.
23. Olds T, Ridley K, Wake M, et al. How should activity
guidelines for young people be operationalised? Int J
Behav Nutr Phys Act. 2007;4(1):43.
24. Spinks AB, Macpherson AK, Bain C, McClure RJ. Com-
pliance with the Australian national physical activity
guidelines for children: relationship to overweight status.
J Sci Med Sport. 2007;10:156–163.
25. Burke V, Beilin LJ, Simmer K, et al. Predictors of body
mass index and associations with cardiovascular risk fac-
tors in Australian children: a prospective cohort study.
International Journal of Obesity & Related Metabolic
Disorders: Journal of the International Association for
the Study of Obesity. 2005;29(1):15–23.
26. Brodersen NH, Steptoe A, Boniface DR, Wardle J, Hillsdon
M. Trends in physical activity and sedentary behaviour in
adolescence: ethnic and socioeconomic differences. British
Journal of Sports Medicine. 2007;41(3):140–144.
27. Vandewater EA, Rideout VJ, Wartella EA, Huang X, Lee
JH, Shim M-s. Digital childhood: electronic media and
technology use among infants, toddlers, and preschoolers.
Pediatrics. 2007;119(5):e1006–1015.
28. Nielson A. Australian TV trends 2001. AC Nielson Media
International; 2001.
29. Hardy LL, Baur LA, Garnett SP, et al. Family and home
correlates of television viewing in 12-13 year old ado-
lescents: the Nepean Study. Int J Behav Nutr Phys Act.
2006;3(24):1–9.
30. Gorely T, Marshall SJ, Biddle SJ. Couch kids: correlates
of television viewing among youth. Int J Behav Med.
2004;11(3):152–163.
31. Salmon J, Hume C, Ball K, Booth M, Crawford D. Indi-
vidual, social and home environment determinants of
change in children’s television viewing: The Switch-Play
Intervention. J Sci Med Sport. 2006;9(5):378–387.
32. Stokols D. Translating social ecological theory into
guidelines for community health promotion. Am J Health
Promot. 1996;10(4):282–298.
33. Baranowski T, Perry CL, Parcel GS. How individuals,
environments, and health behavior interact: Social cog-
nitive theory. In: Glanz K, Rimer BK, Lewis FM, eds.
Health behavior and health education: theory, research,
and practice. San Francisco: Jossey-Bass; 2002.
34. He M, Harris S, Piche L, Beynon C. Understanding
screen-related sedentary behavior and its contributing
factors among school-aged children: a socio-ecological
exploration. Am J Health Promot. 2009;23(5):299–308.
35. Van der Horst K, Paw MJ, Chin A, Twisk JSR, Van
Mechelen W. A brief review on correlates of physical
activity and sedentariness in youth. Med Sci Sports Exerc.
2007;39(8):1241–1250.
36. Trewain D. Socio-economic indexes for areas, Australia
2001: information paper, census of population and hous-
ing. Canberra, Australia: Australian Bureau of Statistics;
2003. ABS Catalogue No. 2039.0.
37. Granich J, Rosenberg M, Knuiman M, Timperio A. Under-
standing children’s sedentary behaviour: a qualitative
study of the family home environment. Health Educ Res.
2010;25(2):199–210.
38. Salmon J, Timperio A, Telford A, Carver A, Crawford D.
Association of family environment with children’s televi-
sion viewing and with low level of physical activity. Obes
Res. 2005;13(11):1939–1951.
39. de Mast J, van Wieringen W. Measurement system
analysis for bounded ordinal data. Qual Reliab Eng Int.
2004;20(5):383–395.
40. Haas M. Statistical methodology for reliability studies. J
Manipulative Physiol Ther. 1991;14(2):119–132.
41. Telford A, Salmon JD, Crawford D. Reliability and validity
of a self-report and proxy-report physical activity question-
naire for children: The Children’s Leisure Activities Study
Survey. Pediatr Exerc Sci. 2004;16:64–78.
42. Gore C, Olds T, Carter L, Norton K. Accreditation in
anthropometry: a curriculum guide. Belconnen, Australia:
Australian Sports Commission; 1998.
43. Cole TJ, Bellizzi MC, Flegal K, Dietz WH. Establishing a
standard definition for child overweight and obesity world-
wide: international survey. BMJ. 2000;320:1240–1243.
44. Eisenmann J. Waist circumference percentiles for 7-
to 15-year-old Australian children. Acta Paediatrica.
2005;94(9):1182–1185.
45. McCarthy HD. Body fat measurements in children as
predictors for the metabolic syndrome: focus on waist
circumference. Proc Nutr Soc. 2006;65(04):385–392.
46. Janssen I, Katzmarzyk PT, Srinivasan SR, et al. Combined
influence of body mass index and waist circumference on
coronary artery disease risk factors among children and
adolescents. Pediatrics. 2005;115(6):1623–1630.
47. Hardy LL, Dobbins T, Denney-Wilson E, Booth ML, Okely
AD. Sedentary behaviours among Australian adolescents.
Aust N Z J Public Health. 2006;30(6):534–540.
48. Hands B, Parker H, Glasson C, Brinkman S, Read H.
Physical activity and nutrition levels in Western Australian
children and adolescents: CAPANS Survey Report. Perth,
Western Australia: Western Australian Government; 2004.
49. Hamar P, Biddle S, Soos I, Takacs B, Huszar A. The
prevalence of sedentary behaviours and physical activity
in Hungarian youth. Eur J Public Health. 2009;2009: .
50. Saelens BE, Sallis JF, Nader PR, Broyles SL, Berry CC,
Taras HL. Home environmental influences on children’s
television watching from early to middle childhood. J Dev
Behav Pediatr. 2002;23(3):127–132.
51. Ortega FB, Tresaco B, Ruiz JR, et al. Cardiorespiratory
fitness and sedentary activities are associated with adipos-
ity in adolescents. Obesity. 2007;15(6):1589–1599.
52. Nader PR, Bradley RH, Houts RM, McRitchie SL, O’Brien
M. Moderate-to-vigorous physical activity from ages 9 to
15 years. JAMA. 2008;300(3):295–305.

53. Biddle SJ, Gorely T, Marshall SJ, Murdey I, Cameron
N. Physical activity and sedentary behaviours in youth:
issues and controversies. J R Soc Promot Health.
2004;124(1):29–33.
54. Marshall SJ, Biddle SJH, Sallis JF, McKenzie TL, Conway
TL. Clustering of sedentary behaviors and physical activity
among youth: a cross-national study. Pediatr Exerc Sci.
2002;14:401–417.
55. Biddle SJH. Sedentary Behavior (editorial and commen-
try). Am J Prev Med. 2007;33(6):502–504.
56. Booth ML, Dobbins T, Okely AD, Denney-Wilson E,
Hardy LL. Trends in the prevalence of overweight and
obesity among young Australians, 1985, 1997, and 2004.
Obesity. 2007;15(5):1089–1095.
57. Venn AJ, Thomson RJ, Schmidt MD, et al. Overweight
and obesity from childhood to adulthood: a follow-up of
participants in the 1985 Australian Schools Health and
Fitness Survey. Medical Journal of Australia. 2007;186(9).
58. Katzmarzyk PT, Srinivasan SR, Chen W, Malina RM,
Bouchard C, Berenson GS. Body mass index, waist cir-
cumference, and clustering of cardiovascular disease risk
factors in a biracial sample of children and adolescents.
Pediatrics. 2004;114(2):e198–205.
Electronic Media Use Among Children   625
59. Weiss R, Dziura J, Burgert TS, et al. Obesity and the
metabolic syndrome in children and adolescents. N Engl
J Med. 2004;350(23):2362–2374.
60. Morrison JA, Friedman LA, Gray-McGuire C. Metabolic
syndrome in childhood predicts adult cardiovascular dis-
ease 25 years later: The Princeton Lipid Research Clinics
follow-up study. Pediatrics. 2007;120(2):340–345.
61. Zabinski MF, Norman GJ, Sallis JF, Calfas KJ, Patrick K.
Patterns of sedentary behavior among adolescents. Health
Psychol. 2007;26(1):113–120.
62. Martin CK, Coulon SM, Markward N, Greenway FL,
Anton SD. Association between energy intake and viewing
television, distractibility, and memory for advertisements.
Am J Clin Nutr. 2009;89(1):37–44.
63. Roemmich J, Epstein L, Raja S, Yin L. The neighborhood
and home environments: disparate relationships with
physical activity and sedentary behaviors in youth. Ann
Behav Med. 2007;33(1):29–38.
Copyright of Journal of Physical Activity & Health is the property of Human Kinetics Publishers, Inc. and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.