Longitudinal associations between psychosocial, home, and neighborhood factors and children’s physical activity

Abstract

Background:

Physical activity (PA) provides important health benefits to children, and a large percentage of children’s PA occurs at home. The purpose of this study was to examine associations between psychosocial, home, and neighborhood environmental factors and children’s reported PA at home and in the neighborhood, during the transition from elementary to high school.

Methods:

A total of 555 participants (44% boys) were recruited in grade 5 and followed through grades 6,7, and 9. Children self-reported PA in three locations – at home, in the neighborhood and on the street. Children reported parent support and neighborhood environment, parents reported PA equipment, and a windshield survey assessed incivilities and outside PA equipment. Longitudinal Poisson models evaluated the relationships between environmental variables and three self-reported PA variables, adjusting for gender, race/ethnicity, and parent education.

Results:

Parent support and PA equipment were significant positive predictors of home PA. Child’s perceived environment (positive), and incivilities (negative) were significant predictors of neighborhood PA. Parental support, perceived environment, and outside PA equipment were positive significant predictors of street PA.

Conclusions:

This study supports the need for both family and community/neighborhood PA interventions that encourage parents to support child PA and for communities to reduce incivilities.

Introduction

Physical activity provides many health benefits for children, including better cardiorespiratory fitness, stronger muscles and bones, and healthier weight status.^1,2 Experts recommend that children and youth accumulate 60 min of moderate-to-vigorous physical activity (MVPA) per day, but studies show that few children and youth meet this guideline³ and that the prevalence declines with age.³ On school days, children and youth age 6 and older accumulate about half of their daily MVPA after school and during the evenings.⁴

As reported in several studies, a significant percentage of physical activity occurs at home and in the neighborhood.^5,6 Kneeshaw-Price and colleagues reported that 6–11-year-old children spent 47.5% of their time at home and accumulated 44.6% of their daily MVPA while there.⁵ They spent an additional 0.8% of their time in the neighborhood, where they accumulated another 1.5% of their daily MVPA. Similarly, Carlson et al.⁷ reported that 12–16-year-olds participated in 33.7% of their MVPA at or near their homes. Accordingly, it is important to consider both home and neighborhood locations for children’s PA.⁸

Social ecological models emphasize multiple influences on physical activity behavior⁹ and provide a useful framework for conceptualizing influences on physical activity.¹⁰ Studies^8,11,12 and reviews^10,13 identified multiple factors associated with children’s PA, including home factors (opportunities at home and PA equipment), psychosocial factors (parent support), and neighborhood environmental factors. Furthermore it is important to consider social and environmental influences on PA as well as to focus on both the home and neighborhood.

Research into the relationships between home and neighborhood factors and physical activity has been largely cross-sectional^13,14, with few exceptions.^15,16 Little research has been done to determine the relationship between home and neighborhood environmental factors and participation in physical activity performed specifically at home and in the neighborhood. Parental involvement with a physical activity intervention, as well as the characteristics of the built environment (e.g. ease of access to neighborhood facilities) have been weakly associated with physical activity change in the short term.^13,17,18 However, few studies have examined longitudinally, from the perspective of both the parents and children, the simultaneous influence of parent support and access to opportunities for physical activity in the home and neighborhood^19,20 Therefore the purpose of this study was to examine associations between psychosocial, home, and neighborhood environmental factors and child-reported physical activity at home and in the neighborhood among a sample of children who transitioned from elementary school to high school, for the total group and by sex.

Methods

Participants and Setting

Children were enrolled in the longitudinal Transitions and Activity Changes in Kids (TRACK) study, designed to examine multiple influences and predictors of change in physical activity behavior, which began when participants were in 5^th grade (n=1084) and followed them into high school. The children were recruited in 2010 from 21 elementary schools in two school districts in South Carolina. The first district was in a county in the north-central section of the state, with a 2010 population of 226,046 persons (75% white, 19.4% black and 11.2% living in poverty). The second district was in a county in the central part of the state, with a population in 2010 of 107,490 persons (48% white, 48% black and 19.1% living in poverty).²¹ Recruitment assemblies were held in all schools, during which all 5^th graders were invited to participate. They received information regarding the data collection and informed consent forms for their parents to read, complete and sign.

Data collection procedures occurred over two sessions per child at the school and were administered by a trained measurement team at each measurement period (5^th, 6^th, 7^th and 9^th grades). At the first visit, participants completed the student questionnaire (e.g., neighborhood perception items, parent support items), and the measurement team took anthropometric measures. During the second visit participants completed the Physical Activity Choice survey (PAC), which assessed types of physical activities performed. Students took home the parent survey for their parents to complete each year. Parents (87% were mothers at the 5^th grade measurement period) completed questionnaires about their child and family, and about access to types of physical activity equipment in the home. Parents signed consent forms and children gave their assent before beginning any study procedures. The Institutional Review Board at the University of South Carolina approved all protocols.

The self-reported race/ethnic breakdown of the 1084 children was 36.2% white, 35.0% black, 11.2% Hispanic, and 17.7% other race/ethnicity (including multiracial). In the present analyses, children were excluded if they did not complete the Physical Activity Choice (PAC) instrument in the 5^th grade (n=297), if they did not complete at least two out of three PAC instruments in 6^th, 7^th, and 9^th grade (n= 184), were missing parent-reported education (n=17), or were missing a windshield survey (e.g. neighborhood observations, n=31), leaving 555 children who were included in the analyses. There were no differences at baseline in the percent by gender, parent education or BMI between those in the analyses and those not included, using Chi-square or t-tests. However, those included were on average younger (10.5±0.5 versus 10.7±0.6) and a higher percent were black than those not included in the analyses.

Demographic Characteristics

Each child responded to two questions about race/ethnicity. The first asked whether the child thought of himself/herself as Hispanic or Latino. The second asked the child to check as many categories as applied (white, African American/black, Asian, American Indian/Alaska Native, and other). Trained research staff members measured participants’ height to the nearest 0.1 cm using a portable stadiometer and weight to the nearest 0.1 kg using an electronic scale, after the child removed heavy clothing and shoes. BMI (kg/m²) and BMI z -scores (based on CDC growth charts)²² were calculated. Parents reported their highest level of education (1=attended high school, 2=completed high school, 3=attended college or technical school, 4=attended graduate school, 5=completed graduate school), and their relationship to the child.

Physical activity

Physical activity was self-reported using two instruments. Each year (5^th, 6^th, 7^th and 9^th grades) children completed the Physical Activity Choice (PAC) survey, which included a list of 48 physical activities and asked where the activities were performed over the last five days (e.g. school, park, home, neighborhood).²³ More than one place could be checked, and the list included both home and neighborhood locations. Sums of the activities were calculated for each child who responded yes, separately for home and then neighborhood, and ranged from 0 to 17. Beginning in the 6^th grade, four street physical activity items²⁴ were included in the student questionnaire that asked whether physical activity was performed during the past five days: 1) in the street in front of your home 2) in your front yard or driveway 3) in the front yard or driveway of a neighbor who lives on your street 4) in a vacant (empty) lot on your street. The number of days over the past 5 for each item was also collected (0–20 possible score). Cronbach’s alpha for the four items was 0.75 at the 6^th grade measurement.

Parental support

Parental support for physical activity was reported by the child using five items²⁵ that asked them to report how frequently during a normal week their parents did the following: encourage you to do physical activities or play sports; do a physical activity or play sports with you; provide transportation to a place where you can do physical activity or play; watch you participate in physical activities or sports; or tell you that you are doing well in physical activities or sports? The five responses ranged from none (coded as 0) to daily (coded as 5). In the present sample, the Cronbach’s alpha was 0.90 for parent support.

Home and Neighborhood Environment

Parents reported which of 14 types of physical activity-related equipment their child has access to in the home or yard (e.g. basketball hoop, active equipment, bicycle).²⁶ Children completed nine items that addressed perceived characteristics of the neighborhood²⁷ that have been associated with after-school moderate-to-vigorous physical activity.²⁸ The items inquired about places to go: sidewalks; bike or walking trails; safe to walk/jog; people can see walkers/bikers from their homes; see girls or boys playing outdoors; interesting things to look at while walking; streets are well lit; and there are playgrounds, parks and gyms close to home. There were four possible responses for the items, ranging from disagree a lot (coded as 1) to agree a lot (coded as 4). The Cronbach’s alpha for the 5^th grade was 0.74.

A Windshield Survey^29,30 was completed for the street segment corresponding to each participant’s home address, between 5^th and 6^th grade and again when the child was in high school. The street segment was from one intersection to another and total length did not exceed 0.5 miles.³¹ Trained research assistants drove the length of the participants’ street segments. One data collector conducted data assessment while the other drove the vehicle. The data collector who was observing rode in the passenger seat and observed the space on the right side of the car. The data collectors drove each segment up to 8 times to collect necessary information (4 times going in one direction, 4 times coming back, to observe both sides of the street). Only one person recorded information except when reliability was being assessed. Windshield survey data were collected in the summer months (May-September).

Three scales were created from the windshield data: physical incivilities (7 items, e.g., litter, graffiti), territoriality (6 items, e.g., security signs, decorations), and social spaces (9 items, e.g., presence of yards, visible people).^30,32 Inter-observer reliabilities for the three scales were >0.80. Observers also recorded the presence of physical activity equipment in participants’ yard/front of the house, including balls/toys, sports goals, and wheeled toys (e.g., bike). For the present analysis, outside equipment was coded as none versus one or more.

Statistical Analysis

Descriptive characteristics of the children were calculated for the total group and by sex, and t-tests and Chi-square were used to determine if there were sex differences. Means (SD) were presented for home PA, neighborhood PA, and street PA for total group and by sex. To determine if there were differences by sex and time, models were tested using longitudinal Poisson regression (Genmod in SAS 9.4)³³ with school as a random variable. Means and (SD) for the total group were presented for each of the potential physical activity determinates (physical activity equipment, physical activity support, perceived environment, incivilities, territoriality, social spaces and outside equipment) by year of the study, and mixed model repeated measures were used to test if there were differences over time.

Separate longitudinal Poisson regression models (Genmod in SAS 9.4)³³ were conducted for each of the self-reported physical activity variables (PA home, PA neighborhood and street PA) for the total group and by sex. Models included time varying variables: child’s reported parent support, parent’s reported physical activity equipment, child’s perceived neighborhood environment, and the four windshield survey variables (territoriality, physical incivilities, social spaces, and physical activity equipment in the yard). Also included were interactions by time for each of these variables. All models were adjusted for child’s race, gender and parent education (i.e., socioeconomic status), and school was treated as a random variable. A backward elimination was then conducted to eliminate non-significant terms (p≥.05) other than gender, race and parent education for each model. For ease of interpretation, all continuous variables were centered by subtracting the grand mean of the variable. Time was coded as 0 (5^th grade), 1 (6^th grade), 2 (7^th grade), and 4 (9^th grade) for home PA and neighborhood PA, and as 0 (6^th grade), 1 (7^th grade), and 3 (for 9^th grade) for street PA. The intercept and slope (time) were modeled as random effects and children were nested in their elementary school. Also, unconditional growth models were run for comparison of fit indices.

Results

Of the 555 5^th grade boys (43.6%) and girls (56.4%), approximately 41% were black, 36% white, 8% Hispanic and 15% other races (Table 1). Self-reported home PA, neighborhood PA, and street PA (Table 2) declined over time (p<.001). Home PA over the past 5 days declined from 4 activities in 5^th grade, to about 2 activities in 9^th grade. Similarly, neighborhood PA declined from about 2 activities in 5^th grade to <1 in 9^th grade, and street PA declined from 6.6 activities in 6^th grade to 3.8 in 9^th grade (p<.001). Reports of home PA were higher for girls than for boys (p<.001). Boys reported more days of street activity than girls (p=.001).

Table 1.

Characteristics of 555 5^th graders

Characteristic	Mean (SD) or Percent			p-value*
Characteristic	Total group	Boys n=242	Girls n=313
Gender, boys	43.6%
Age, years	10.5 (0.5)	10.5 (0.6)	10.5 (0.5)	.78
BMI†, kg/m2	21.4 (5.0)	20.8 (4.8)	21.8 (5.2)	.02
BMI z-score†	0.94 (1.1)	0.88 (1.1)	0.99 (1.0)	.19
Race				.50
Black	40.7%	43.8%	38.3%
Hispanic	8.1%	8.7%	7.7%
Other	15.1%	14.5%	15.7%
White	36.0%	33.1%	38.3%
Parent education, >HS	57.5%	59.1%	56.2%	.50

Abbreviations: HS, high school

^†n for BMI and BMI z score=549 in the total group

^*p-value for difference between boys and girls

Table 2.

Self-reported physical activity

	5th		6th		7th		9th
Variable and source	Total Group
	n	Mean (SD)	n	Mean (SD)	n	Mean (SD)	n	Mean (SD)
Home PA1	555	4.0 (3.0)	541	2.8 (2.3)	541	2.0 (2.0)	217	1.9 (2.0)
Neighborhood PA2	555	1.8 (2.3)	541	1.2 (1.8)	541	0.8 (1.4)	217	0.8 (1.4)
Street PA3			550	6.6 (5.2)	552	6.0 (4.9)	255	3.8 (4.4)
	Boys
Home PA	242	3.6 (2.6)	238	2.5 (2.3)	238	1.8 (1.9)	87	1.6 (1.9)
Neighborhood PA	242	1.5 (2.0)	238	1.2 (1.8)	238	0.7 (1.2)	87	0.8 (1.2)
Street total PA			241	6.8 (5.5)	239	6.6 (4.8)	103	4.8 (5.0)
	Girls
Home PA4	313	4.4 (3.2)	303	3.0 (2.3)	303	2.1 (2.0)	130	2.0 (2.1)
Neighborhood PA	313	2.0 (2.5)	303	1.2 (1.8)	303	0.8 (1.5)	130	0.8 (1.5)
Street PA5			309	6.4 (5.1)	313	5.5 (4.9)	152	3.1 (3.7)

Abbreviations: PA, physical activity

¹Home PA declined over time, p<.001.

²Neighborhood PA declined over time, p<.001.

³Street PA declined over time (p<.001).

⁴Home PA differed between boys and girls, p<.001.

⁵Street PA differed between boys and girls, p=.002.

Note: Scores ranged from 0 to 48 possible activities over 5 days for Home and or Neighborhood PA. Street PA ranged from 0 to 20 (count of 0–4 items over by number of days (0–5).

Table 3 provides means (SD) for psychosocial and environmental variables. The number of physical activity equipment items reported by the parent was ≥6 at each time point. Parental physical activity support and perceived environment reported by the child declined from 7^th to 9^th grade (p<.001). The windshield survey physical incivilities and territoriality variables improved from elementary school to high school (p<001). Social spaces and observed outside equipment remained stable over the two measurement periods.

Table 3.

Sample size and mean (SD) of psychosocial and environment factors over time for total group

	Possible range	5th		6th		7th		9th		p-value
Physical activity equipment1	0–14	507	6.3 (2.6)	481	6.2 (2.6)	440	6.1 (2.4)	195	6.0 (2.8)	.46
Physical activity support2	0–5	528	3.4 (0.9)	535	3.4 (0.9)	516	3.3 (0.9)	255	3.0 (1.1)	<.001
Perceived environment3	1–4	555	2.8 (0.6)	552	2.8 (0.7)	552	2.8 (0.7)	255	2.6 (0.7)	<.001
Incivilities4	0–1	535	0.3 (0.5)					263	0.5 (0.5)	<.001
Territoriality5	0–4	535	1.8 (0.9)					263	2.4 (1.0)	<.001
Social spaces6	0–9	535	3.0 (1.0)					263	3.0 (0.9)	.96
Outside equipment7	Yes, No	535	0.4 (0.5)					263	0.4 (0.5)	.22

¹Parent reported up to fourteen physical activity related equipment in the home or yard

²Child reported parent physical activity support using five items

³Child reported perceived characteristics of neighborhood using nine items

⁴Incivilities (e.g., litter, graffiti) reported during the windshield survey of child’s street segment

⁵Territoriality (e.g., security signs, decorations) reported during the windshield survey of child’s street segment

⁶Social spaces (e.g., presence of yard, visible people) reported during the windshield survey of child’s street segment

⁷Presence (yes, no) of outside physical activity related equipment in front house or child’s yard reported during the windshield survey of child’s street segment; 60% of children had no outside equipment at the 5^th/6^th grade measure.

The longitudinal Poisson regression models across the four time points are presented in Table 4. All three self-reported physical activity variables declined significantly (<.001) over time for the total group and by gender (<.01). Most of the conditional models showed improvement in fit compared to the models with only time. Gender differences were observed, with boys having lower self-reported physical activity at home and higher physical activity in street settings, compared to girls. Children whose parents had greater than a high school education reported lower physical activity at home.

Table 4.

Results from longitudinal Poisson regression for physical activity outcomes and psychosocial and environmental factors

	Home PA Grades 5–9			Neighborhood PA Grades 5–9			Street PA Grades 6–9
	Total	Boys	Girls	Total	Boys	Girls	Total	Boys	Girls
Intercept	1.23 (0.06)***	0.93 (0.09)***	1.30 (0.07)***	0.43 (0.10)***	−0.06 (0.13)	0.55 (0.12)***	1.80 (0.06)***	1.89 (0.08)***	1.88 (0.07)***
Time	−0.23 (0.02)***	−0.26 (0.04)***	−0.23 (0.03)***	−0.29 (0.04)***	−0.26 (0.05)***	−0.31 (0.05)***	−0.14 (0.02)***	−0.09 (0.03)**	−0.20 (0.03)***
Child reported Parent support	0.09 (0.03)***	0.08 (0.04)*	0.08 (0.03)*		0.20 (0.07)**		0.25 (0.03)***	0.20 (0.04)***	0.17 (0.05)**
Parent support *time									0.10 (0.03)**
Parent reported Home Equipment	0.06 (0.01)***	0.04 (0.01)**	0.06 (0.02)***
Home equip*time	−0.02 (0.01)*		−0.02 (0.01)*
Child’s Perceived environment				0.43 (0.06)***		0.38 (0.08)***	0.20 (0.04)***	0.21 (0.06)**	0.19 (0.06)**
Windshield
Incivilities		−0.21 (0.09)*		−0.29 (0.13)*	−0.32 (0.18)	−0.34 (0.20)
Incivilities*time				0.22 (0.07)**	0.18 (0.09)*	0.28 (0.10)*
Outside Equipment							0.11 (0.05)*
Race
Black	0.11 (0.06)	0.30 (0.10)**	0.01 (0.09)	0.01 (0.11)	0.47 (0.16)**	−0.15 (0.14)	−0.10 (0.07)	−0.06 (0.09)	−0.12 (0.09)
Hispanic	0.07 (0.09)	0.30 (0.13)*	−0.09 (0.14)	−0.09 (0.16)	0.28 (0.19)	−0.18 (0.26)	−0.03 (0.11)	0.03 (0.14)	−0.07 0.19)
Other	0.09 (0.07)	0.19 (0.10)	0.04 (0.09)	0.10 (0.13)	0.20 (0.21)	0.09 (0.16)	0.05 (0.07)	.11 (0.10)	0.002 (0.11)
White	Reference	Reference	Reference	Reference	Reference	Reference	Reference	Reference	Reference
Gender, boys	−0.20 (0.05)***			−0.14 (0.09)			0.14 (0.06)*
Parent education, > high school	−0.19 (0.05)***	−0.19 (0.08)*	−0.26 (0.07)***	−0.12 (0.09)	−0.30 (0.13)*	−0.06 (0.12)	−0.04 (0.06)	0.004 (0.07)	−0.05 (0.08)
Fit Criteria Unconditional model
QIC	−336.71	82.30	−432.49	1230.74	627.96	624.63	−2835.01	−1502.36	−1385.12
Fit Criteria Full models
QIC	−417.60	12.95	−402.13	1181.41	628.01	601.09	−3042.97	−1662.31	−1611.01

Abbreviations: PA, physical activity; QIC, quasi-likelihood information criteria, smaller is better.

Notes: Time-varying variables parent support, home equipment, perceived environment and window survey variables and continuous variables were centered.

^*p<.05,

^**p<.10,

^***p<.001

For the total group, child-reported parent support (p<.001) and parent-reported physical activity equipment (p<.001) in the home were positively associated with home PA, after adjusting for child’s gender, race and parent education. Home equipment was associated with change in home PA for the total group and in girls (interaction between home equipment and time) (p<.05) and indicates that children with more home physical activity equipment reported higher home PA in 5^th grade, but differences in home PA disappeared over time. Also, the presence of incivilities was negatively (<.05) related to home PA in boys.

Child’s perception of the environment (p<.001) was positively associated with neighborhood PA in the total group and for girls. For the total group, and for boys and girls, incivilities was associated with change in neighborhood PA. Children whose street segment included one or more incivilities reported lower neighborhood PA in the 5^th grade, but there was little difference in neighborhood PA by 9^th grade regardless of the number of incivilities (p<.01). Child-reported parent support was also positively related to neighborhood PA in boys (p<.01).

In the total group, child-reported parent support (p<.001), perceived environment (p<.001), and windshield-observed physical activity equipment (p<.05) were positively related to street PA. In girls, parent support was related to change in street PA (interaction between parent support and time, p<.01). Over time, girls with lower perceived parent support had lower reported street PA as compared to girls with higher perceived parent support.

Discussion

Collectively the findings of this study underscore the importance of considering multiple domains of influence (e.g., psychosocial and environmental), the specific context (e.g., home, street or neighborhood), and group (e.g., boys or girls) when developing programs to promote physical activity in youth. Consistent with the social ecological model, factors measured in multiple domains (i.e., child reported parent support, parent-reported PA equipment, child’s perception of the environment, objectively-measured incivilities, and outside equipment) were associated with home, neighborhood, and street PA in children transitioning from elementary school to high school. Furthermore, the specific relationships between these factors and self-reported physical activity varied among home, neighborhood, and street locations and by gender, as discussed in the following sections.

Determinants of Home PA

Parent support and availability of physical activity equipment were positively related in the total group and in both boys and girls over the four time points. These results are consistent with those of Tandon and colleagues¹² and Millstein and colleagues.⁸ However, in a review¹⁰ of studies, physical activity equipment was not consistently related to overall physical activity in youth. In the present study, the association between PA equipment and home PA declined over time. This perhaps suggest that older children who became more independent and who had equipment were active in places other than at home or were no longer active. In the 5^th grade, 91% of the children in this study reported performing one or more physical activities at home, emphasizing the importance of parent support and availability of play equipment for children in this age group.

Determinants of Neighborhood PA

The child’s perception of the environment was positively related to neighborhood PA (total group and in girls) and incivilities was negatively related to neighborhood PA (total group and in boys and girls) over the four time points. In boys, perceived parent support was also related to neighborhood PA. The finding that perceived environment is associated with neighborhood PA is consistent with previous reported literature.^8,27,28 Similarly, neighborhoods with high recreation-supportive environments have been associated longitudinally with better child weight outcomes and lower sedentary activity.¹⁶ Other studies have reported inconsistent findings on the association of incivilities and reported physical activity.³⁴ In the present study, over time, the difference in the amount of neighborhood PA between those with incivilities and without incivilities declined, suggesting that age seems to matter. Only about half of the children reported performing one or more physical activities in their neighborhood in the 5^th grade in the present study, perhaps because some children did not have a place to be active in the neighborhood (e.g., a park).

Determinants of Street PA

In the present study, over the three time points (6^th, 7^th, and 9^th grade), parent support, perceived environment and presence of outside physical activity equipment were positively related and presence of incivilities was negatively related to street PA in the total group. These results are consistent with those of other studies.^35,36 However, girls with lower perceived parent support reported lower street PA than girls with higher perceived parent support over time. In the present study, 90% of 6^th grade children reported being active in and around their street. The street near the home can be a great place for children to play and has prompted interventions such as Play Streets, which temporarily closes streets for children and their families to engage in active play.³⁷

Strengths and Weaknesses

Strengths of the present study include the longitudinal design that followed children from elementary school to high school, and the collection of physical activity data for three specific locations. Additionally, both boys and girls and children from several race/ethnic groups were included. A limitation of this study is that children reported location of physical activity. Also, this study included only children from two school districts in one state, limiting generalizability, and the sample size decreased over time. However, the statistical analyses utilized all available data and maximum likelihood estimation. Continued research is needed to consider other determinates (e.g., self-efficacy) that may influence the relationships that were found in this study, and to evaluate if there are race/ethnic differences in the relationships.

Conclusions

The present study highlights the importance of the neighborhood and home environments in promoting children’s physical activity. The physical environment included both perceptions of the neighborhood (structures such as sidewalks, parks and playgrounds, safe to walk/jog, seeing adults and children being active), and objectively measured incivilities (litter, graffiti). The study found positive relationships between self-reported physical activity and parent support, physical activity equipment and perceived environment. Several gender differences were observed. Girls’ physical activity was higher at home, while boys reported higher levels of street physical activity. The study also found differences in the associations between the psychosocial, home and environmental factors with PA at the different locations. The results of this study highlight the importance of considering context in physical activity interventions and support the need for both family and community/neighborhood physical activity intervention strategies that encourage parents to support their children and for communities to keep their streets clean and free of incivilities.