Physical inactivity is a major contributing factor to the obesity epidemic, and adolescence is associated with a decline in physical activity [1
]. Despite known health benefits, a large portion of the United States (U.S.) adolescent population does not engage in the recommended 60 min of accumulated moderate to vigorous physical activity (MVPA) daily [2
]. Of particular concern is that physical activity levels decline during the transition from childhood to adolescence, a period crucial to the development and creation of habitual participation in physical activity throughout the life course [3
]. One prominent explanation of inactivity is that adolescents are spending too little time playing outdoors due to social and physical neighborhood factors [5
]. It is possible that if parents restrict outdoor play due to concerns about their neighborhood, their adolescent offspring will be less active overall, as time spent outdoors is positively associated with overall physical activity [1
]. Generally, individuals in more disadvantaged neighborhoods have lower levels of physical activity and higher rates of obesity, even when controlling for individual-level socioeconomic status (SES). These relationships may be due to safety concerns (crime, victimization, poorly lighted streets); the physical environment (incivilities, access to parks/playgrounds, sidewalks, walkable destinations); or to differences in neighborhood social environment (perceived safety, social cohesion, social capital, social support, perceived collective efficacy) [5
Emerging social environmental research identifies perceived collective efficacy as a potentially influential determinant of physical activity opportunities for both adults [11
] and youth [6
]. Collective efficacy is a form of social capital and is defined as a measure of perceived social cohesion (mutual trust among neighborhoods) and social control (capacity and willingness of the group to intervene for a common goal) [14
]. Numerous social capital indicators, including collective efficacy, have been identified and tested in relation to health behaviors and outcomes, yet have resulted in mixed conclusions [7
]. For example, physical activity studies operationalizing the social environment as sense of belonging, social cohesion or norms of reciprocity illustrate no significance, whereas studies assessing collective efficacy report significant results [9
]. Furthermore, the majority of studies relating the social environment and physical activity assessed the impact on active transport, rather than outdoor play in general and were performed mainly in older populations [10
While a number of studies have assessed the impact of the social environment on children’s physical activity, few studies have considered physical activity-related parenting behaviors (e.g., constrained behaviors), which may ultimately restrict or facilitate adolescents’ physical activity [1
]. Adolescents who are free to play outdoors and travel actively without adult supervision accumulate more physical activity than those who are not; therefore understanding whether parental perceptions of their neighborhood impact physical activity-related parenting behaviors may be crucial to improving overall activity among adolescents [18
]. Constrained behavior is defined as the act of restricting offspring’s physical activity and is categorized as “avoidance” or “defensive” behaviors depending on whether physical activity was avoided or modified, respectively. Examples of avoidance behavior include parents driving their children to school instead of allowing them to walk or forbidding unsupervised outdoor play. Examples of defensive behavior include parental accompaniment while walking to school or restriction of outdoor play to the backyard [1
]. Carver and colleagues (2010) is the only study to examine environmental influence on constrained behaviors and found that perceived risk of harm to a child in the neighborhood was positively associated with constrained behavior resulting in lower levels of active transport and MVPA outside of school hours for children and adolescents [1
]. However, no studies to our knowledge have explored the influence of parental perceived collective efficacy on constrained behaviors, nor have any further explored the impact of the physical environment on parental perceptions and behaviors impacting adolescents’ activity.
The physical environment (e.g., neighborhood incivilities) may be the foundation for, or etiology of, perceived collective efficacy and may ultimately explain differences in perceptions of collective efficacy among parents living in low- versus high-incivility neighborhoods. The social disorganization theory [19
] and the norms and collective efficacy model [20
] posit that characteristics of the built environment influence the ability of community members to establish cohesive relationships and create a shared set of socially-accepted norms that promote the willingness to intervene on behalf of the common good [21
]. Therefore, achieving the recommended 60 min of daily physical activity is difficult due to a complex interaction among the physical environment, parental perceptions of the social environment and their constraints on adolescent outdoor play. Additional studies are necessary to understand the influence of the social environment on parenting behaviors and offspring’s physical activity, especially among populations who live in high-incivility neighborhoods. The current study examined relationships between parental perceived collective efficacy, constrained behaviors (e.g., avoidance or defensive behaviors) and adolescent offspring neighborhood physical activity in low- versus high-incivility neighborhoods. Conceptual path diagrams are presented in Figure 1
and Figure 2
This novel study explored associations among the social and physical environment and parenting behaviors to explain anticipated low levels of physical activity in a small sample of adolescents. This preliminary study was the first to report that parents who perceived low levels of collective efficacy were more likely to constrain their adolescents’ outdoor play practices. Yet, constrained behaviors did not impact neighborhood physical activity in these offspring. Results from multilevel regression analyses by incivilities suggested that parental perceptions of collective efficacy might be more influential in high- versus low-incivility neighborhoods. Our findings are consistent with the ecological perspective and stress the importance of understanding how the social and physical environments interact to impact parenting behaviors and adolescents’ neighborhood physical activity to reduce adverse health outcomes at early ages.
Parents who perceived higher levels of collective efficacy exhibited less avoidance and defensive behavior; however, gender, race and annual household income diminished these findings. Other studies have reported similar inverse relationships between other characteristics of the social environment (i.e., perceived risk and perceived safety) and parental constrained outdoor play practices [1
]. Parents of children 15–17 years of age who perceived greater risk of their child being harmed in their neighborhood had higher levels of avoidance and defensive behavior [1
]. In the same cohort, perceived personal safety was related to lower levels of avoidance and defensive behaviors [33
]. In our sample, significant relationships between perceptions of collective efficacy and constrained outdoor play practices did not remain after controlling for demographic variables (gender, race and annual household income). This may be due to low-income families being disproportionately exposed to a host of environmental factors that may further restrict their physical activity, such as limited access to safe areas to play, that were not considered in the current study [34
]. In our sample, a higher number of low-income and non-white participants lived in neighborhoods characterized by high incivilities compared to high-income and white participants, and collective efficacy may have been more influential in these high-incivility neighborhoods (Table 2
and Table 4
). Yet, our sample was predominantly white and high income (Table 2
), which was likely not an ideal sample in which to explore these relationships; moreover, the low variability in parental perceptions of collective efficacy may have limited our ability to detect significant relationships. Furthermore, parenting behaviors may be especially relevant in low-income populations, where children have less exposure to physically-active parents, fewer parental role models, less ability for parental supervision of outdoor play and less joint activity with parents [35
]. Thus, a more diverse sample that better represented the racial and socio-economic composition of the urban New Orleans area may have revealed stronger correlations between parents’ perceptions of collective efficacy and constrained outdoor play practices, especially in high-incivility neighborhoods.
Perceptions of collective efficacy were related to parenting behaviors, but direct relationships were not found between parents’ perceptions of collective efficacy and their children’s neighborhood physical activity. In contrast to our findings, numerous research studies in moderate to large child/adolescent populations reported positive associations between perceived collective efficacy and physical activity [5
]. In particular, children whose mothers perceived higher levels of collective efficacy in their neighborhood played outside for longer periods of time, watched less television and visited the park or playground more frequently [10
]. However, these studies were primarily performed in children under 11 years of age. Parental perceptions of collective efficacy may be less influential in older, more autonomous youth in the current study because parents/guardians (e.g., constrained outdoor play practices) may modify their actions to align with their belief that older offspring possess better self-regulatory skills.
The present study, to our knowledge, is the first to explore how physical environmental factors (i.e., incivilities) measured objectively may alter the relationship between parents’ perceptions of collective efficacy and: (1) parental constrained outdoor play practices; and (2) adolescents’ physical activity. Individuals in our sample who lived in high-incivility neighborhoods reported lower perceived collective efficacy and more avoidance and defensive behaviors than those living in low-incivility neighborhoods (Table 2
). This finding is supported by the “broken windows theory”, which states that the presence of disorder, such as graffiti, litter or abandoned homes, communicates a dangerous, unmonitored environment, which influences individual’s attitudes and perceptions [10
]. Although not statistically significant, regression and mediation analyses within low- and high-incivility neighborhoods suggest that relationships between perceptions of collective efficacy and behaviors are more meaningful in neighborhoods with high incivilities. Perceived collective efficacy may have a lower impact in neighborhoods with low incivilities, since neighborhoods with appealing physical environments have additional modes of social control, such as gated entrances, security that decreases graffiti, landscape services, well-kept sidewalks and roads for play [37
]. Therefore, collective efficacy may not be the key process by which parents make decisions regarding children’s outdoor play in low-incivility neighborhoods. Whereas, in neighborhoods with high incivilities, perceived collective efficacy may be a key process by which parents overcome the negative effects of incivilities, thereby increasing the importance of social cohesion and control among neighbors. This is supported by previous research indicating that the relationship between incivilities and fear of crime was mitigated by higher levels of perceived collective efficacy [37
Several studies report relationships between the social environment and physical activity, but a paucity of research has investigated whether parenting behaviors are the mechanism by which the environment impacts children’s physical activity and health outcomes [1
]. The current study is novel because it investigated parental constraint of outdoor play practices and how this relates to neighborhood physical activity. Results indicated that constrained outdoor play practices was not a significant mediator. This can be explained by the non-significant direct relationship between perceived collective efficacy and adolescent’s neighborhood physical activity, which ultimately precluded our ability to detect mediation. However, mediation results within high-incivility neighborhoods suggest that constrained outdoor play practices may be the mechanism by which perceived collective efficacy impacts neighborhood physical activity (Figure 5
). Carver and colleagues (2010) reported that higher levels of parental constraint on outdoor play due to perceived risk of harm resulted in less active transport and lower total objectively measured MVPA outside of school hours for both children aged 10–11 years and adolescents aged 15–17 [1
]. However, there were inconsistencies throughout their results. Constrained behavior influenced female, but not male adolescents’ MVPA, and hypothesized associations were true only for younger children and adolescent girls. Our results found no gender differences in constrained outdoor play practices, which may indicate that the adolescent population in the current study may be more autonomous. As stated earlier, adolescents from the current study presumably were less restricted by parenting behaviors than those reported for younger children [1
]. In Carver’s study, parents of children ages 10–11 years reported more avoidance and defensive behaviors compared to those of children aged 15–17 years [1
]. Adolescence is a time of increased autonomy when rules and boundaries regarding outdoor play may be re-negotiated [1
]. Additionally, greater autonomy may increase the pertinence of the child’s own perception of neighborhood collective efficacy on their physical activity behaviors, especially since adolescents may experience the neighborhood context in a manner distinct from their parents [1
]. Furthermore, adolescents may lose interest in specific outdoor locations in their neighborhood, such as their yard, cul de sac or local park or playground, which may result in decreased physical activity [43
]. Furthermore, the lack of significant relationships between constrained behavior and physical activity may also be explained by today’s children belonging to the ‘backseat generation’ that are chauffeured to structured leisure-time activities outside of their neighborhood more than previous generations [1
The low sample size limited several findings. Furthermore, due to the small sample and cross-sectional nature of this study, our mediation analyses are exploratory in nature. Nevertheless, our findings were able to detect a significant association between perceived collective efficacy and parental constrained behaviors. Subsequent research should examine hypothesized relationships in a larger, more diverse cohort. Regrettably, the target population, and resulting small sample size, was limited by the availability of potential volunteers (n = 146) who participated in the parent study, the MET study (2006–2012). The selection of this convenience group of returning study participants inadvertently generated another limitation. The MET cohort included children recruited throughout southeastern Louisiana and included primarily white families from high income households, which limits the generalizability of the findings to more diverse populations.
The inclusion of objectively measured physical activity data is a strength of this study; however, 47.8% of participants did not achieve valid wear time (a minimum of three days with 8 h of wear time per day) and, thus, were not included for analyses. Defining valid wear time is complex, and varying definitions have resulted in a range from 38%–84% of study samples that achieved usable accelerometry data [44
]. Furthermore, accelerometers were not equipped with global positioning technologies (GPS) to determine the location of physical activity. Therefore, the study relied on potentially biased parent-reported perceptions of their child’s neighborhood physical activity. Additional studies with a larger sample of valid accelerometer data that incorporates Global Positioning Systems (GPS) technology are warranted to increase the power to detect relationships between neighborhood characteristics and physical activity occurring in the neighborhood.
Although numerous factors were considered, additional factors may explain the overall low levels of physical activity in our sample. For example, participation in organized sports teams and the types of services available to youths in a neighborhood have been shown to influence adolescent’s physical activity [1
]. Furthermore, individual factors (e.g., self-efficacy) or neighborhood-level factors (e.g., access to greenspace, incivilities, peers in the neighborhood and living within a walkable distance to school or businesses) that were not considered may have influenced activity levels. Additionally, future research should consider psychosocial factors (i.e., self-esteem, psychological adjustment, substance abuse) that may be influenced by parenting practices and ultimately influence physical activity levels in our sample [45
]. However, a major strength of the current study was the inclusion of a novel conceptualization of the physical environment as incivilities, which were objectively measured by systematic social observation (SSO) at the parcel-level using Google Street View. Measures were aggregated to create summary measures for the street segment on which the child lived. Therefore, this unique application may not have truly captured the entire neighborhood-level incivilities. Further research is needed to determine if SSO performed on a larger area (e.g., for the entire block face or on 10% of street segments within the block group) would provide a more complete assessment of the associations among perceived collective efficacy and parenting behaviors, physical activity and obesity-related outcomes. Regardless of this limitation, the SSO method provided a more accurate overall assessment of incivilities because factors were measured at the smallest geographic unit possible (parcel-level), which potentially provides a more detailed description of neighborhoods and captures incivilities that are within an individual’s control (i.e., their residential and yard condition, litter in their yard, etc.) [30
]. This may be advantageous for micro-level studies of individual health outcomes, such as those included in the current study.
The social environment is broad; no definition or single measure encompasses its effect on an individual’s behaviors and health outcomes [7
]. The way in which the social environment is measured and conceptualized produces inconsistent results when examining physical activity and health [7
]. The current study was limited to the assessment of perceived collective efficacy. Thus, other social environmental factors (e.g., sense of belonging, trust and norms of reciprocity), which may be associated with parental restriction of physical activity, were not considered. Furthermore, overall high perceived collective efficacy scores and, therefore, low variation in levels of perceived collective efficacy among parents of study participants may have reduced the ability to detect relationships between perceived collective efficacy and physical activity. Lastly, the present study did not consider participants’ length of residence in their current neighborhood. Greater length of residence may have implied increased stability within a neighborhood and greater permanence of network affiliations, which may have increased collective efficacy among neighbors.