Social and Environmental Factors Related to Boys’ and Girls’ Park-Based Physical Activity

Abstract

Parks provide opportunities for physical activity for children. This study examined sex differences in correlates of park-based physical activity because differences may indicate that a standard environmental intervention to increase activity among children may not equally benefit boys and girls.
The System for Observation Play and Recreation in Communities was used to measure physical activity among 2,712 children and adolescents in 20 neighborhood parks in Durham, North Carolina, in 2007. Sedentary activity, walking, vigorous park activity, and energy expenditure were the primary outcome variables. Hierarchical logit regression models of physical activity were estimated separately for boys and girls.
Type of activity area and presence of other active children were positively associated with boys' and girls' physical activity, and presence of a parent was negatively associated. A significant interaction involving number of recreation facilities in combination with formal activities was positively associated with girls' activity. A significant interaction involving formal park activity and young boys (aged 0-5 y) was negatively associated with park-based physical activity.
Activity area and social correlates of park-based physical activity were similar for boys and girls; findings for formal park programming, age, and number of facilities were mixed. Results show that girls' physical activity was more strongly affected by social effects (eg, presence of other active children) whereas boys' physical activity was more strongly influenced by the availability of park facilities. These results can inform park planning and design. Additional studies are necessary to clarify sex differences in correlates of park-based physical activity.

Full text

PREVENTING CHRONIC DISEASE
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY
Volume 12, E97 JUNE 2015
ORIGINAL RESEARCH
Social and Environmental Factors Related
to Boys’ and Girls’ Park-Based Physical
Activity
JasonN.Bocarro,PhD; MyronF.Floyd,PhD; WilliamR.Smith,PhD;
MichaelB.Edwards,PhD; CourtneyL.Schultz,MS; PerverBaran,PhD;
RobinA.Moore,MCP; NildaCosco,PhD; LuisJ.Suau,PhD 
Suggested citation for this article:
Bocarro JN, Floyd MF,
Smith WR, Edwards MB, Schultz CL, Baran P, et al. Social and
Environmental Factors Related to Boys’ and Girls’ Park-Based
Physical Activity. Prev Chronic Dis 2015;12:140532. DOI: http://
dx.doi.org/10.5888/pcd12.140532.
PEER REVIEWED
Abstract
Introduction
Parks provide opportunities for physical activity for children. This
study examined sex differences in correlates of park-based physic-
al activity because differences may indicate that a standard envir-
onmental intervention to increase activity among children may not
equally benefit boys and girls.
Methods
The System for Observation Play and Recreation in Communities
was used to measure physical activity among 2,712 children and
adolescents in 20 neighborhood parks in Durham, North Carolina,
in 2007. Sedentary activity, walking, vigorous park activity, and
energy expenditure were the primary outcome variables. Hierarch-
ical logit regression models of physical activity were estimated
separately for boys and girls.
Results
Type of activity area and presence of other active children were
positively associated with boys’ and girls’ physical activity, and
presence of a parent was negatively associated. A significant inter-
action involving number of recreation facilities in combination
with formal activities was positively associated with girls’ activity.
A significant interaction involving formal park activity and young
boys (aged 0–5 y) was negatively associated with park-based
physical activity.
Conclusion
Activity area and social correlates of park-based physical activity
were similar for boys and girls; findings for formal park program-
ming, age, and number of facilities were mixed. Results show that
girls’ physical activity was more strongly affected by social ef-
fects (eg, presence of other active children) whereas boys’ physic-
al activity was more strongly influenced by the availability of park
facilities. These results can inform park planning and design. Ad-
ditional studies are necessary to clarify sex differences in correl-
ates of park-based physical activity.
Introduction
Most US children do not accumulate the recommended 60 minutes
of daily physical activity, and adherence to recommendations is
significantly lower among girls (1). Public parks and playgrounds
are key components of environmental interventions to increase
physical activity among children during nonschool hours and can
be modified through public policy to further encourage daily phys-
ical activity (2).
Availability of a variety of recreation facilities and proximity to
them have been associated with increased physical activity among
adults (3) and children (4), with some exceptions (5). Improve-
ments and renovations in parks (6) and school playgrounds (7)
have been associated with increased park-based physical activity
among children. Although studies have shown there are sex differ-
ences among youth park-based physical activity (8), research ex-
amining such differences in the context of social and environment-
al factors such as parental supervision and other children is lim-
ited (9).
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health
and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2015/14_0532.htm • Centers for Disease Control and Prevention 1

Previous studies provide evidence that associations between envir-
onmental variables and physical activity vary between boys and
girls (10,11). Sex differences present challenges for understand-
ing how the built environment affords opportunities for physical
activity among children. Such differences may indicate that a
standard environmental intervention to increase activity among
children may not equally benefit boys and girls. Understanding
whether social and environmental determinants influence physical
activity differently among boys and girls can be used to guide de-
cisions related to programming options and the design and retrofit-
ting of parks. Therefore, this study sought to determine if associ-
ations among social and environmental characteristics of parks and
park-based physical activity among children varied by sex. We hy-
pothesized that boys’ park use would result in greater intensity
levels of park-based physical activity than girls’ park use across
different park activity settings. Second, we hypothesized that so-
cial and environmental correlates of park-based physical activity
vary between boys and girls.
Methods
Study settings
Direct observations were conducted in 20 neighborhood parks in
Durham, North Carolina, in 2007. To ensure socioeconomic di-
versity of population and inclusion of neighborhoods with a mix-
ture of races/ethnicities, this study focused on the mostly residen-
tial central area of the city. Simple random sampling was used to
select 20 of 38 available parks. Mean park size was 10.3 acres
(range, 0.5–45.9 acres). For each park, a 400-meter network buf-
fer was constructed using ArcGIS 9.1 (Esri) to define park service
areas (12,13). Park service areas are geographic catchment areas
encompassing a population of potential users (12). Their mean
size was 121.8 acres (range, 35.4–305.5 acres). Park service areas
were used to identify contextual characteristics of study parks.
Total population of census blocks within park service areas ranged
from 32 to 1,796, with a mean of 780. African American residents
were the largest racial/ethnic group (mean,55.1%; range,
1.4%–100%), followed by non-Hispanic white (mean, 36.1%,
range 0%–96.9%) and Hispanic residents (mean = 9.7%, range
0%–32.4%). The mean percentage of children aged 18 years or
younger was 24.3% and ranged from 16.0% to 35.7%
(www.census.gov//census2000/states/nc.html).
Measures
Physical activity
Park-based physical activity was measured using the System for
Observing Play and Recreation in Communities (SOPARC) (14).
Developed for open environments such as parks and playgrounds,
SOPARC consists of systematic momentary time sampling of pre-
determined activity areas in parks. Park activity areas were
scanned by trained observers visually sweeping from left to right,
and the codes representing children’s activity levels were recor-
ded on a standardized form along with other contextual informa-
tion. Observations occurred for 8 weeks from 10AM to 7PM EDT
on all weekend days and randomly selected weekdays in May,
June, and July 2007. Each park was observed 16 times during the
study period (8 weekend days and 8 weekdays). Observed activity
was coded as sedentary (eg, standing, sitting, lying down), walk-
ing (eg, walking, other moderate intensity activities), or vigorous
(eg, running, climbing, jumping) as validated in previous studies
(15,16). The codes also provided estimates of energy expenditure
rate (EER). EER was estimated by equating categories of activit-
ies with constants (sedentary, 0.051 kcal/kg/min; walking, 0.096
kcal/kg/min; very active, 0.144 kcal/kg/min) (17,18).
Demographic and social correlates
The SOPARC age category was modified to account for 3 age cat-
egories: young child (0–5 y), middle-child (6–12 y), and older
children or adolescents (13–18 y). The observation codes for dif-
ferent age groups were introduced after the research staff was able
to recognize and code SOPARC physical activity levels. Age was
determined by observable physical and biological features (eg,
height) and social context (eg, with a parent). The description of
the training protocols are reported elsewhere (19). Paired observa-
tions from 4 study weeks produced data to assess interrater reliab-
ility for all physical activity codes for each age group. Adequate
reliability was observed for physical activity codes using the 3 age
groups (Table 1).
Presence of an adult was coded by observers as 1) not present, 2)
supervising adult (eg, teacher, coach), 3) parent or caregiver, or 4)
don’t know (mean κ, 0.6). Presence of other active children was a
dichotomous variable indicating the presence of other moderately
or very active children in an activity area (1 = yes). Formality of
play was measured using 4 attributes: no play observed, free play,
informal organized play (eg, group playground play, pick-up soc-
cer), and formal organized play (eg, individual or group athletic
event) (mean κ, 0.8).
Park environment correlates
SOPARC observations were conducted in predetermined zones
(park activity areas) in each park. The areas for SOPARC observa-
tions were mapped by 3 members of the research team. Mean
acres for SOPARC zones was 0.35 (standard deviation, 0.45). En-
vironmental features in each activity area were measured using the
Environmental Assessment of Public Recreation Spaces (EAPRS)
instrument (20). Audits for the presence of facilities and amenities
were conducted for each SOPARC zone (N = 134) during day-
PREVENTING CHRONIC DISEASE VOLUME 12, E97
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JUNE 2015
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
2 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2015/14_0532.htm

light hours by 2 pairs of 2 raters working independently. Features
serving as primary supports for physical activity were categorized
as recreation facilities (eg, trails, playground equipment). Second-
ary features (eg, tables, benches) were treated as park amenities
(21). The mean κ across audited features was 0.9. Counts of facil-
ities and amenities in park activity areas were derived for sub-
sequent analyses. Activity area types were categorized on the basis
of their designated use. These included playgrounds, courts, fields,
open areas, swimming pools, and picnic areas and shelters. Size of
activity areas was measured by calculating the area of the poly-
gons comprising the activity areas. All study procedures were ap-
proved by the North Carolina State University’s institutional re-
view board.
Analysis
First, differences in energy expenditure among boys and girls were
compared for different activity settings (eg, playgrounds, sport
fields, courts). Second, sex-stratified logit models of park-based
physical activity were estimated to examine whether different pat-
terns of associations existed for boys and girls.
Analysis of variance was used to test for sex differences in intens-
ity of physical activity associated with different park activity
areas. Hierarchical generalized regression models stratified by sex
were used to examine whether social and environmental correl-
ates of park-based physical activity varied between boys and girls.
Hierarchical generalized linear models (22) were estimated be-
cause of the hierarchical structure of the data (individual children
within park zones) and the use of a 3-level ordinal dependent vari-
able (sedentary, walking, and very active). To address the hypo-
thesis related to the environmental characteristics of park activity
areas, unconditional models (intercept only) were estimated to es-
tablish whether variation in individual park-based physical activ-
ity levels were significantly associated with differences in park
activity areas. Independent variables of interest were examined by
using fixed effects and odds ratios controlling for other model pre-
dictors. Sedentary served as the reference category in regression
models. Within-level and cross-level interactions were also ex-
amined. Analyses were performed in May 2011 using SAS ver-
sion 9.2 (SAS Corp).
Results
Characteristics of park users and activity areas are shown in Table
2. During the study period, 2,712 children were observed, and
43.5% were girls (Table 2). Among girls, 50.4% were categorized
in the 0 to 5 years age group, followed by 39.2% in the 6 to 12
years age group and 10.3% in the 13 to 18 years age group.
Among boys, the 6 to 12 years age group was most frequently ob-
served (42.3%) followed by 36.6% for the 0 to 5 years age group
and 21.1% for the 13 to 18 years age group. Boys were more
likely to be observed in informally and formally organized park
activities (16.1 and 9.4, respectively, for boys vs 5.8 and 4.1, re-
spectively, for girls). Girls were also more likely than boys to be
observed in free play (68.1 vs. 53.5) (
χ
2
= 108.8,
P
< .001).
Regarding supervision, a parent or guardian was more likely to be
present among girls (55.9%, girls vs 44.6%, boys). Presence of
other adult supervisors was more likely to be observed among
boys (25.1%, boys vs 19.9%, girls). Activity levels were similar
across boys and girls, with boys slightly more active than girls
overall. Of girls, 56.1% were observed in sedentary activity com-
pared with 49.9% of boys. Of girls, 31.4% were observed in walk-
ing activity behaviors compared with 36.6% of boys.
Results of hierarchical generalized regression
model
Girls’ park-based physical activity
Examination of the fixed effects indicated that significant vari-
ation existed in thresholds (intercepts) across all park activity
areas between sedentary activity and vigorous activity (intercept 1)
but not between sedentary and walking (intercept 2) (Table 3).
Controlling for predictor variables, girls across all park activity
areas were equally likely to be observed sedentary as they were to
be observed walking (moderate intensity activities).
Individual level effects (level 1) and cross-level interactions.
Within park activity areas and controlling for other variables in the
model, the presence of a parent was associated with lower odds
(odds ratio [OR], 0.47) of an increased level of physical activity
for girls. The presence of other active children in a park activity
area had the strongest positive association (OR, 4.85) with girls’
park-based physical activity. A significant positive interaction in-
volving the number of recreation facilities and formally organized
activity was associated with 2.21 increased odds of higher park-
based physical activity. That is, girls participating in more formal
and organized play had higher odds of increased physical activity
in activity areas with a greater number of recreation facilities.
Boys’ park-based activity
Examination of the fixed effects indicated that significant vari-
ation existed in thresholds (intercepts) across park activity areas
between sedentary and vigorous activity (intercept 1) but not
between sedentary and walking (intercept 2). Controlling for pre-
dictor variables, boys across park activity areas were no less likely
to be observed sedentary as they were to be observed walking.
Individual level effects (level 1) and cross-level interactions. As
with girls, but to a lesser extent, the presence of a parent was asso-
ciated with a lower likelihood (OR, 0.66) of high levels of physic-
PREVENTING CHRONIC DISEASE VOLUME 12, E97
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JUNE 2015
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2015/14_0532.htm • Centers for Disease Control and Prevention 3

al activity. A stronger interaction of age and formality of play as-
sociated with lower odds of increased physical activity was found
among boys in this sample in comparison with girls. Among boys
categorized as age 0 to 5, each increased level of formalization and
organization in park activity was associated with a 37% decrease
in odds of higher levels of physical activity. Boys observed in park
activity areas identified as picnic areas and shelters were 70% less
likely to be engaged in high levels of physical activity than boys in
other types of park activity areas. High levels of park-based phys-
ical activity were associated with courts (eg, basketball, tennis,
etc.) and presence of other active children.
In sum, high levels of girl’s park-based activity were associated
with presence of other active children and a combination of form-
al activity and an increasing number of recreation facilities in park
activity areas. Common correlates between girls and boys were
presence of a parent and presence of other active children and a
low likelihood of increased activity associated with picnic areas.
Among boys, increased activity was also associated with athletic
courts.
Discussion
Neighborhood parks have potential to help children accumulate re-
commended levels of daily physical activity. This study sought to
determine whether social and environmental correlates of park-
based physical activity differed between boys and girls. Multivari-
ate sex-specific models provided evidence of similarities and note-
worthy differences in social and environmental correlates for boys
and girls. Models for boys and girls showed that the type of park
activity area was associated with increased likelihood of vigorous
activity but not with walking intensity activity. This finding is im-
portant in 3 respects. First, it indicates that characteristics of park
activity areas may be uniquely associated with vigorous activity, a
“relative rare” intensity level (1). Second, previous studies indic-
ate that vigorous intensity is more strongly associated with youth
fitness levels and weight status (23,24). Finally, such findings
provide direction for identifying appropriate spatial scales for
park-based physical activity interventions. Interventions can tar-
get multiple scales (ie, neighborhoods, parks, and areas within
parks). In light of data from earlier studies (6–9), these findings
suggest that activity areas within parks can be the focus of envir-
onmental or policy change.
Relationships between social correlates and physical activity were
similar for boys and girls; however, noteworthy differences
emerged. Presence of other active children had a stronger positive
association with park-based physical activity than any other pre-
dictor variable. The association was stronger among girls where
presence of other active children increased the odds of higher
activity levels by 4.85 times compared with 3.12 times for boys.
Methods used in the current study precluded assessment of rela-
tionships among children. The findings are nevertheless consist-
ent with studies that show friendships among adolescents are
strong predictors of their physical activity levels (9). Presence of
parents was associated with a lower likelihood of increased activ-
ity among children and adolescents with a stronger negative trend
observed among girls. These findings support the hypothesis that
gender norms affect how some parents respond to children’s beha-
vior. Possibly parents’ efforts to supervise and monitor their chil-
dren’s safety inadvertently curtails higher levels of activity. The
pervasiveness of gender norms may encourage some parents to be
more protective of girls and encourage “safer” sedentary play
activities. Other research suggests parental anxieties about safety
are more constraining of young children’s physical activity than
the level of public recreation facilities (25). Communicating with
parents about the social and physiologic importance of active play
is a major challenge in children’s health promotion (26,27). The
results suggest that active play among girls should be emphasized.
Differences between boys and girls were observed for interactions
involving formality of park activities. Among boys in the 0 to 5
age group, formal park activity was associated with a low level of
physical activity. Thus, park programming designed to promote
physical activity for young boys could minimize formal games and
emphasize free play (26). By contrast, among girls, formal park
activity and a greater number of recreation facilities appeared to
increase high levels of physical activity. Timperio et al (28) repor-
ted that an increased number of neighborhood recreation facilities
was inversely associated with walking activity among young girls.
Our findings suggest that programming and availability of facilit-
ies within parks appear to be more important for girls than boys
among all age groups. This may be because parents are more re-
luctant to let girls play outside if unsupervised (29) and may be
less reluctant if programming and facilities are available because
of their association with supervision.
Strengths of this study include use of sex-stratified models to ex-
amine social and environmental correlates of park-based physical
activity; measurement of 3 different age groups of children; use of
validated assessment tools (SOPARC and EAPRS) to measure ac-
tual park use, park-based physical activity, and environmental fea-
tures; statistical analyses adjusted for clustering within park activ-
ity areas; and parks sampled from racially diverse neighborhoods.
Primary limitations were that observations of children occurred at
1 point in time and not continuously during the course of
children’s visit to the park; energy expenditure was not measured
directly; the study design was cross-sectional; we did not know
PREVENTING CHRONIC DISEASE VOLUME 12, E97
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JUNE 2015
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
4 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2015/14_0532.htm

whether parks users were demographically similar to the park ser-
vice area population; and results were not generalizable beyond
the summer season.
To our knowledge this is the first study using sex-stratified mod-
els to examine social and environmental correlates of park-based
physical activity among children and adolescents. Overall, similar
patterns of associations were observed for boys and girls. Charac-
teristics of activity areas, presence of parents, and presence of oth-
er active children exhibited similar effects. Key differences were
type of activity areas among boys, formality of park activity
among young boys, and presence of recreation facilities interact-
ing with formality of park activity among girls. Public parks and
recreation facilities are important community resources for pro-
moting physical activity (2). To maximize parks’ contribution to
children’s total physical activity, future studies should build on
these results to identify and evaluate social and environmental
factors with strong potential to support walking and vigorous
activity levels within parks. Parks are built environmental features
with high relevance for children (13,30), and more studies are
needed to learn how the children’s sex influences the effect of so-
cial and built environment correlates on park-based activity.
Park-program specialists and researchers should consider the role
parents and other children play in encouraging activity. Our res-
ults indicate that parents may discourage high intensity activity
among children with a more noticeable influence among girls.
Presence of other active children appears conducive to higher in-
tensity activity, especially for girls. Presence of specific park facil-
ities should also be considered. Courts were more conducive for
boys’ activity. However, picnic areas were less conducive to phys-
ical activity among both boys and girls. A combination of organ-
ized park activities and recreation facilities produced greater activ-
ity among girls.
These findings suggest that environmental correlates that support
physical activity in parks differ across age groups and between
girls and boys. Enhanced knowledge about social and environ-
mental factors that increase the frequency and intensity of physic-
al activity among children should help park administrators, pro-
grammers, and planners to design parks that meet the needs of
children of all ages and of both sexes. To better understand how
parks benefit children’s physical activity, potential differences in
the characteristics of park environments that encourage or discour-
age boys’ and girls’ physical activity should be further examined.
Acknowledgments
This study was funded by a grant from Active Living Research
(no. 59449), a research program of the Robert Wood Johnson
Foundation. The authors thank the City of Durham, North Caro-
lina, Parks and Recreation Department, for its assistance with this
study.
Author Information
Corresponding Author: Jason N. Bocarro, PhD, Department of
Parks, Recreation, and Tourism Management, North Carolina
State University, Box 8004, Biltmore Hall, Raleigh, NC 27695.
Telephone: 919-513-8025. Email: jnbocarro@ncsu.edu.
Author Affiliations: Myron F. Floyd, William R. Smith, Michael
B. Edwards, Courtney L. Schultz, Perver Baran, Robin A. Moore,
Nilda Cosco, North Carolina State University, Raleigh, North
Carolina; Luis J. Suau, Shaw University, Raleigh, North Carolina.
References
Troiano RP, Berrigan D, Dodd KW, Mâsse LC, Tilert T,
McDowell M. Physical activity in the United States measured
by accelerometer. Med Sci Sports Exerc 2008;40(1):181–8.
1.
Institute of Medicine. Local government actions to prevent
childhood obesity. Washington (DC): National Academy of
Sciences; 2009.
2.
Diez Roux AV, Evenson KR, McGinn AP, Brown DG, Moore
L, Brines S, et al. Availability of recreational resources and
physical activity in adults. Am J Public Health 2007;
97(3):493–9.
3.
Cohen DA, Ashwood JS, Scott MM, Overton A, Evenson KR,
Staten LK, et al. Public parks and physical activity among
adolescent girls. Pediatrics 2006;118(5):e1381–9.
4.
Adkins S, Sherwood NE, Story M, Davis M. Physical activity
among African-American girls: the role of parents and the
home environment. Obes Res 2004;12(S9,Suppl):38S–45S.
5.
Gardsjord HS, Tveit MS, Nordh H. Promoting youth’s
physical activity through park design: linking theory and
practice in a public health perspective. Landscape Research.
2014;39(1):70–81.
6.
Colabianchi N, Maslow AL, Swayampakala K. Features and
amenities of school playgrounds: a direct observation study of
utilization and physical activity levels outside of school time.
Int J Behav Nutr Phys Act 2011;8(1):32.
7.
Weimann H, Björk J, Rylander L, Bergman P, Eiben G.
Neighbourhood environment and physical activity among
young children: a cross-sectional study from Sweden. Scand J
Public Health 2015;43(3):283–93.
8.
PREVENTING CHRONIC DISEASE VOLUME 12, E97
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JUNE 2015
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2015/14_0532.htm • Centers for Disease Control and Prevention 5

Coughenour C, Coker L, Bungum TJ. Environmental and
social determinants of youth physical activity intensity levels
at neighborhood parks in Las Vegas, NV. J Community Health
2014;39(6):1092–6.
9.
Carver A, Timperio A, Hesketh K, Crawford D. Are safety-
related features of the road environment associated with
smaller declines in physical activity among youth? J Urban
Health 2010;87(1):29–43.
10.
Yan AF, Voorhees CC, Clifton K, Burnier C. “Do you see
what I see?” — correlates of multidimensional measures of
neighborhood types and perceived physical activity-related
neighborhood barriers and facilitators for urban youth. Prev
Med 2010;50(Suppl 1):S18–23.
11.
Nicholls S. Measuring the accessibility and equity of public
parks: a case study using GIS. Managing Leis. 2001;
6(4):201–19.
12.
Evenson KR, Wen F, Hillier A, Cohen DA. Assessing the
contribution of parks to physical activity using global
positioning system and accelerometry. Med Sci Sports Exerc
2013;45(10):1981–7.
13.
McKenzie TL, Cohen DA, Sehgal A, Williamson S, Golinelli
D. System for Observing Play and Recreation in Communities
(SOPARC): reliability and feasibility measures. J Phys Act
Health 2006;3(Suppl 1):S208–22.
14.
Rowe PJ, Schuldheisz JM, van der Mars H. Measuring
physical activity in physical education for use with first to
eighth grade students. Pediatr Exerc Sci 1997;9:136–49. http://
www.humankinetics.com/acucustom/sitename/Documents/
DocumentItem/12401.pdf. Accessed April 17, 2015.
15.
Rowe P, van der Mars H, Schuldheisz J, Fox S. Measuring
students’ physical activity levels: Validating SOFIT for use
with high school students. J Teach Phys Educ 2004;23:235–51.
http://www.humankinetics.com/acucustom/sitename/
Documents/DocumentItem/3983.pdf. Accessed April 17, 2015.
16.
McKenzie TL, Sallis JF, Nader PR, Patterson TL, Elder JP,
Berry CC, et al. BEACHES: an observational system for
assessing children’s eating and physical activity behaviors and
associated events. J Appl Behav Anal 1991;24(1):141–51.
17.
McKenzie TL, Marshall SJ, Sallis JF, Conway TL. Leisure-
time physical activity in school environments: an observational
study using SOPLAY. Prev Med 2000;30(1):70–7.
18.
Bocarro JN, Floyd MF, Moore R, Baran P, Danninger T, Smith
W, et al. Developing a reliable measure of physical activity
among children and adolescents in different age groups using
the System for Observing Physical Activity and Recreation in
Communities (SOPARC). J Phys Act Health 2009;
6(6):699–707.
19.
Saelens BE, Frank LD, Auffrey C, Whitaker RC, Burdette HL,
Colabianchi N. Measuring physical environments of parks and
playgrounds: EAPRS instrument development and inter-rater
reliability. J Phys Act Health 2006;3(Supp 1):S190–207. http://
activelivingresearch.net/files/JPAH_13_Saelens.pdf. Accessed
April 17, 2015.
20.
Kaczynski AT, Stanis SAW, Besenyi GM, Child S.
Differences in youth and adult physical activity in park settings
by sex and race/ethnicity. Prev Chronic Dis 2013;10:E42.
21.
Raudenbush SW, Bryk AS. Hierarchical linear models:
applications and data analysis methods. Vol 1. Second edition.
Thousand Oaks (CA): Sage Publications; 2002.
22.
Gutin B, Yin Z, Humphries MC, Barbeau P. Relations of
moderate and vigorous physical activity to fitness and fatness
in adolescents. Am J Clin Nutr 2005;81(4):746–50.
23.
Gutin B. Child obesity can be reduced with vigorous activity
rather than restriction of energy intake. Obesity (Silver Spring)
2008;16(10):2193–6.
24.
Valentine G, McKendrick J. Children’s outdoor play:
exploring parental concerns about children’s safety and the
changing nature of childhood. Geoforum 1997;28(2):219–35.
25.
Burdette HL, Whitaker RC. Resurrecting free play in young
children: looking beyond fitness and fatness to attention,
affiliation, and affect. Arch Pediatr Adolesc Med 2005;
159(1):46–50.
26.
Alexander SA, Fusco C, Frohlich KL. ‘You have to do 60
minutes of physical activity per day . . . I saw it on TV’:
children’s constructions of play in the context of Canadian
public health discourse of playing for health. Sociol Health Illn
2015;37(2):227–40.
27.
Timperio A, Giles-Corti B, Crawford D, Andrianopoulos N,
Ball K, Salmon J, et al. Features of public open spaces and
physical activity among children: findings from the CLAN
study. Prev Med 2008;47(5):514–8.
28.
Cleland V, Timperio A, Salmon J, Hume C, Baur LA,
Crawford D. Predictors of time spent outdoors among children:
5-year longitudinal findings. J Epidemiol Community Health
2010;64(5):400–6.
29.
Han B, Cohen DA, Derose KP, Marsh T, Williamson S, Raaen
L. How much neighborhood parks contribute to local residents’
physical activity in the City of Los Angeles: a meta-analysis.
Prev Med 2014;69(Suppl 1):S106–10.
30.
PREVENTING CHRONIC DISEASE VOLUME 12, E97
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JUNE 2015
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
6 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2015/14_0532.htm

Tables
Table 1. Cohen’s κ Coefficients and Percentage Observer Agreement by Age Group, Sex, and Physical Activity Code,
Durham, North Carolina, 2007
Child Sex, Age
a
,
and Activity Level
κ Coefficient (% Observer Agreement)
Training (127
Paired
Observations)
Week 1 (130
Paired
Observations)
Week 4 (152
Paired
Observations)
Week 5 (117
Paired
Observations)
Week 7 (120
Paired
Observations)
Week 8 (75
Paired
Observations)
Female, YC, S 0.67 (93.7) 0.57 (96.2) 0.66 (98.7) 0.56 (94.0) 0.80 (97.5) 0.81 (97.3)
Female, YC, W 0.26 (92.1) 0.38 (95.4) 0.74 (98.7) 0.15 (92.3) 0.59 (98.3) 0 (98.7)
Female, YC, V 0.56 (93.7) 0 (99.2) 1.00 (100) −0.01 (98.3) 0.66 (99.2) 0 (98.7)
Male, YC, S 0.63 (96.1) 0.39 (97.7) 0.80 (98.7) 0.84 (98.3) 0.79 (98.3) 0.49 (97.3)
Male, YC, W 0.77 (96.1) 0.71 (98.5) 0.80 (98.7) 0.59 (98.3) 0.56 (97.5) 0.66 (98.7)
Male, YC, V 0.65 (96.9) 1.00 (100) 0.80 (99.3) 0.24 (97.4) 0.80 (99.2) 1.00 (100)
Female, MC, S 0.68 (92.9) 0 (99.2) 0.66 (97.4) 0.81 (98.3) 0.95 (99.2) 0.87 (98.7)
Female, MC, W 0.44 (89.0) 0.49 (97.7) 0.63 (97.4) 0.66 (98.3) 1.00 (100) 0 (98.7)
Female, MC, V 0.49 (91.3) 0.66 (99.2) 0.63 (97.4) 0 (98.3) 1.00 (100) (100)
b
Male, MC, S 0.55 (89.0) 1.00 (100) 0.65 (97.4) 0.59 (96.6) 0.75 (96.7) 0.90 (98.7)
Male, MC, W 0.60 (91.3) (100)
b
0.66 (96.7) 0.66 (99.2) 0.87 (98.3) 0.66 (98.7)
Male, MC, V 0.51 (94.5) 0 (99.2) 0.93 (99.3) 0.49 (98.3) 0.56 (97.5) (100)
b
Female, OC, S 0.55 (96.9) (100)
b
0.66 (99.3) 0.66 (99.2) 0.66 (99.2) 1.00 (100)
b
Female, OC, W 0.43 (96.1) (100)
b
0.89 (99.3) (100)
b
1.00 (100)
b
(100)
b
Female, OC, V (100)
b
(100)
b
1.00 (100) (100)
b
0 (99.2) (100)
b
Male, OC, S 0.68 (94.5) (100)
b
0.50 (99.3) 0.83 (99.2) 0.66 (99.2) 0 (98.7)
Male, OC, W 0.66 (93.7) (100)
b
(100)
b
0.62 (97.4) 0.66 (99.2) −0.03 (93.3)
Male, OC, V 0.83 (97.7) (100)
b
0.66 (99.3) 0.80 (99.2) (100)
b
0 (98.7)
Abbreviations: MC, middle child; OC, older child; S, sedentary physical activity; V, vigorous physical activity; W, walking or moderately active physical
activity; YC, young child.
a
Young child = age 0 to 5 years, middle child = age 6 to 12 years, old child = age 13 to 18 years.
b
No variability observed; therefore, a weighted κ cannot provide a meaningful test statistic.
PREVENTING CHRONIC DISEASE VOLUME 12, E97
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JUNE 2015
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2015/14_0532.htm • Centers for Disease Control and Prevention 7

Table 2. Characteristics of Children and Adolescents Using Parks and Park Activity Areas, Durham, North Carolina, 2007
Characteristic Girls Boys
Park Users, n (%)
Sex
Girls 1,180 (43.5) —
Boys — 1,532 (56.5)
Physical activity level
Sedentary 662 (56.1) 765 (49.9)
Walking 370 (31.4) 558 (36.4)
Vigorous 148 (12.5) 209 (13.6)
Age group
0–5 y 595 (50.4) 560 (36.6)
6–12 y 463(39.2) 648 (42.3)
13–18 y 122 (10.3) 324 (21.1)
Style of play
No play 255 (21.9) 318 (21.0)
Free play 793 (68.1) 811 (53.5)
Informal organized 68 (5.8) 244 (16.1)
Formal organized 48 (4.1) 142 (9.4)
Parent/guardian present 660 (55.9) 684 (44.6)
Supervising adult present 235 (19.9) 384 (25.1)
Park areas
Sex of park area users
Girls only or boys and girls, no. of park areas 89 —
Boys only or boys and girls, no of park areas — 95
Zone type
Playground 24 (26.9) 24 (25.3)
Courts 14 (15.7) 15 (15.8)
Fields 7 (7.8) 10 (10.5)
Trail/walking path 11 (12.4) 12 (12.6)
Shelter/picnic area 14 (15.7) 14 (14.7)
Open space 11 (12.4) 15 (15.8)
Other 8 (8.9) 5 (5.3)
Areas with other active children present, % 23.8 26.3
Zone size, sq ft, mean (SD) 15,966 (20,499) 15,938 (19,279)
No. of facilities, mean (SD) 1.1 (0.8) 1.1 (0.8)
No. of amenities, mean (SD) 1.3 (1.5) 1.2 (1.4)
Abbreviation: SD, standard deviation.
PREVENTING CHRONIC DISEASE VOLUME 12, E97
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JUNE 2015
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
8 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2015/14_0532.htm

Table 3. Hierarchical Cumulative Logit Model for Park-Based Physical Activity of Children and Adolescents, Durham, North
Carolina, 2007
Characteristic
Girls (Level 1 N = 1,180, Level 2 N = 87) Boys (Level 1 N = 1,532, Level 2 N = 95)
Estimate (SE)
P
Value
Odds Ratio
(95% CI) Estimate (SE)
P
Value
Odds Ratio
(95% CI)
Fixed Effects
Main effects
Intercept 1
a
(vigorous
activity)
−1.91 (0.461) <.001 — −2.32 (0.374) <.001 —
Intercept 2
a
(walking
activity)
0.172 (0.456) .70 — −0.124 (0.368) .73 —
Age category 1 (0–5 y) 0.438 (0.394) .26 1.54
(0.77–2.31)
0.338 (0.307) .27 1.4 (0.79 to
2.00)
Age category 2 (6–12 y) 0.336 (0.395) .39 1.39 (0.62–
2.16)
0.151 (0.282) .59 1.16 (0.60 to
1.71)
Age category 3
b
(13–18 y) 1[Reference]
Parent present −0.756 (0.193) <.001 0.469
(0.09–0.85)
−0.423 (0.172) .01 0.655 (0.32 to
0.99)
Supervising nonparental
adult present
−0.322 (0.22) .14 0.725
(0.29–1.16)
−0.339 (0.178) .05 0.712 (0.36 to
1.06)
Formality of play −0.251 (0.314) .42 0.778
(0.16–1.39)
0.203 (0.202) .31 1.23 (0.83 to
1.63)
Level 1 interaction, by age
b
Age 1 × formality of play −0.164 (0.334) .62 0.849
(0.19–1.50)
−0.46 (0.215) .03 0.631 (0.20 to
1.06)
Age 2 × formality of play −0.229 (0.336) .49 0.795
(0.14–1.45)
−0.17 (0.194) .38 0.806 (0.46 to
1.22)
Age 3 × formality of play
b
1[Reference]
Park activity area characteristics
Zone size, area in ft
2
3.19E–06
(7.29E–06)
.66 1 (0.99–1.00) −6.76E–06
(−6.12E–06)
.27 1 (−0.99 to
1.00)
Recreational facilities −0.394 (0.211) .06 0.674
(0.26–1.09)
0.099 (0.2) .62 1.1 (0.71 to
1.49)
Park amenities −0.025 (0.107) .81 0.975
(0.76–1.18)
0.022 (0.098) .81 1.02 (0.83 to
1.21)
Other active children in
area (1= yes)
1.58 (0.167) <.001 4.85
(4.52–5.18)
1.14 (0.143) <.001 3.12 (2.83 to
3.40)
Zone type
Picnic/shelter −1.03 (0.545) .05 0.357 (−0.71 to
1.43)
−1.22 (0.334) .01 0.295 (−0.36 to
0.94)
Courts 0.253 (0.436) .56 1.29
(0.44–2.14)
0.746 (0.337) .02 2.11 (1.45 to
2.77)
Cross-level interaction
Abbreviations: E, exponent value; SE, standard error; —, not applicable
a
Sedentary was used as reference category for the ordinal outcome variable.
b
Age 1 = 0 to 5 years; age 2 = 6 to 12 years; age 3 = age 13 to 18 years. Age category 3 was used as reference category.
(continued on next page)
PREVENTING CHRONIC DISEASE VOLUME 12, E97
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JUNE 2015
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
www.cdc.gov/pcd/issues/2015/14_0532.htm • Centers for Disease Control and Prevention 9

(continued)
Table 3. Hierarchical Cumulative Logit Model for Park-Based Physical Activity of Children and Adolescents, Durham, North
Carolina, 2007
Characteristic
Girls (Level 1 N = 1,180, Level 2 N = 87) Boys (Level 1 N = 1,532, Level 2 N = 95)
Estimate (SE)
P
Value
Odds Ratio
(95% CI) Estimate (SE)
P
Value
Odds Ratio
(95% CI)
Recreational facilities ×
formality of play
0.343 (0.109) .002 2.21
(1.99–2.42)
0.032 (0.107) .77 1.03 (0.82 to
1.24)
Variance components
Level 2 variance (τ
00
) 0.793 (0.255) — — 0.648 (0.192) — —
Goodness of fit (independence)
−2 log L 9,205.41 — — 11,757 — —
Independence −2 log L 9,251.73 — — 11,805.15 — —
χ
2
46.32 <.001 — 48.15 <.001 —
Abbreviations: E, exponent value; SE, standard error; —, not applicable
a
Sedentary was used as reference category for the ordinal outcome variable.
b
Age 1 = 0 to 5 years; age 2 = 6 to 12 years; age 3 = age 13 to 18 years. Age category 3 was used as reference category.
PREVENTING CHRONIC DISEASE VOLUME 12, E97
PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY JUNE 2015
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,
the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
10 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2015/14_0532.htm