Neighborhood social cohesion and youth participation in physical activity in Chicago
Introduction
The 1996 Report of the US Surgeon General put physical activity and its relationship with health and disease prevention at the center of the public health agenda. This landmark report, released on the eve of the Centennial Olympic Games, promoted the idea that “we do not have to scale Olympian heights to achieve significant health benefits” and that “an active lifestyle is available to all” (US Department of Health and Human Services, 1996, PAGE). Despite these motivating words, the Surgeon General's Report and subsequent research has documented disparities in activity levels among young people associated with socioeconomic status (Pamuk, Makuc, Heck, Reuben, & Lochner, 1998), race/ethnicity (Gordon-Larsen, McMurray, & Popkin, 2000), gender, age (Pamuk et al., 1998), and geographic location (Simons-Morton et al., 1997). Physical inactivity is a major cause of preventable death in the United States (McGinnis & Foege, 1993) and the average level of physical activity falls far short of recommended levels of participation among both adults and adolescents (US Department of Health and Human Services, 2000). Understanding the factors that influence young people's participation in physical activity is a public health priority as activity levels decline with age (Harris, Gordon-Larsen, Chantala, & Udry, 2006) and physical activity patterns developed in youth may track into adulthood (Raitakari et al., 1994, Tammelin et al., 2003).
Much of the past research on physical activity among children and adolescents has focused on the individual-level correlates of participation in physical activity (Sallis, Prochaska, & Taylor, 2000). Researchers have examined factors such as social support from family and friends (Garcia et al., 1995, Strauss et al., 2001, Zakarian et al., 1994) and self-efficacy (DiLorenzo et al., 1998, Sallis et al., 2000, Strauss et al., 2001). However, many determinants of health can operate at the level of the local community.
The study of the role of social and physical contexts of neighborhoods in promoting positive outcomes among young people has been approached in several ways, often according to discipline of study and outcome of interest. Several sociological studies have examined the impact of local social context on adolescent and developmental outcomes (Coleman, 1988, Fauth et al., 2007, Mayer and Jencks, 1989, Sampson et al., 1999). The growing number of studies that address the relationship between social and physical environments and physical activity patterns in young people in health research (Ferreira et al., 2007) likely reflects calls for environmental and policy approaches to improve activity levels (Brownson et al., 2001, Brownson et al., 2006, King et al., 1995). From this work, it appears that several environmental characteristics such as access to facilities (Garcia et al., 1995, Gordon-Larsen et al., 2006, Sallis et al., 1997, Sallis et al., 1996, Sallis et al., 2001, Sallis et al., 1993), availability of programs for participating in physical activity (Gordon-Larsen et al., 2000, Trost et al., 1997), and neighborhood safety (Gomez et al., 2004, Gordon-Larsen et al., 2000, Sallis et al., 1996) may be important in the promotion of physical activity.
Fewer studies have explicitly examined the role of geographically defined social contexts on physical activity levels in adult (see for e.g., Fisher et al., 2004, Ross, 2000, Yen and Kaplan, 1999) or adolescent populations (Cohen et al., 2006, Karvonen and Rimpela, 1997, Lee and Cubbin, 2002, Molnar et al., 2004). On the whole, these studies indicated that several neighborhood characteristics (e.g., rates of unemployment, poverty, home ownership, and neighborhood disorder, lack of safety, social cohesion, and education) are promising correlates of physical activity behaviors and obesity. Studies that tested the relationship between measures of neighborhood-level social context (as opposed to individual perceptions of neighborhood context) and physical activity behaviors among young people (Karvonen and Rimpela, 1997, Lee and Cubbin, 2002, Molnar et al., 2004) came to divergent conclusions regarding the significance of the neighborhood context in influencing physical activity. These differences may derive from inconsistent strategies of measuring neighborhood attributes (e.g., census variables, survey of residents), or from differing age ranges of study participants and study settings (e.g., country of origin and urban setting versus national study).
In a recent study (Molnar et al., 2004), researchers used individual-level data from 1378 young people aged 11–15 and their caregivers to assess whether neighborhood safety and disorder were related to physical activity levels. Using data from the Project on Human Development in Chicago Neighborhoods, parental reports of hours young people spent in recreational programming were used to estimate physical activity levels. For the 80 neighborhood clusters selected to represent the range of socioeconomic and racial and ethnic diversity found in Chicago used for this study, neighborhood safety was derived from a separate sample of community residents and measures of neighborhood social disorder were developed using videotapes of 15,141 block faces (i.e., facing sides of the street on a given city block between intersecting cross-streets). In hierarchical linear regression models, lower neighborhood safety and higher levels of social disorder were significantly associated with less activity, after controlling for individual demographics.
Using data from a nationally representative survey that included information on physical activity and other cardiovascular health behaviors, Lee and Cubbin (2002) linked 8165 respondents between the ages of 12 and 21 to their respective census tracts. They then used 1990 Census data to develop several measures of neighborhood context (i.e., neighborhood socioeconomic status, social disorganization, racial/ethnic minority concentration and urbanization measured at the level of the census tract). Physical activity participation was assessed through self-reports of the number of days the young people participated in vigorous physical activity during the previous 7 days. These authors found limited evidence for neighborhood effects related to physical activity measured as a dichotomous outcome (i.e., no days versus any days of participation).
In the third study, Karvonen and Rimpela (1997) linked data on 33 sub-areas, or uniformly urban neighborhood districts in the city of Helsinki, Finland to information on several health behaviors and individual-level variables including social status of the head of the household and the adolescents' level of education for 1048 individuals aged 16–18 years. The authors contrast the relative heterogeneity of population socioeconomic position in a given sub-area of Finland with the socioeconomic segregation into small areas found in Great Britain (and also, we suggest, in the US). For the young women in Karvonen and Rimpela's analysis, living in an area characterized by a relatively high percentage of owner occupied housing was associated with higher participation in organized sports or other physical activities when compared with residence in areas with lower levels of owner occupied housing. The authors proposed that factors correlated with better housing such as “increased availability and access to high quality facilities combined with an upper class culture that highly favours fitness” (Karvonen & Rimpela, 1997) – rather than the quality of the housing per se – were the likely mechanisms explaining the relationship. However, another interpretation of this finding is that areas characterized by relatively higher levels of owner occupied housing are also advantaged in terms of levels of connectedness and solidarity.
Sampson et al. (1999) found that residential stability (associated with homeownership) was significantly associated with the relative frequency of social exchange within neighborhoods on issues of consequence for children such as advice about childrearing or job openings. According to Coleman (1990), this type of reciprocated exchange is enabled by levels of trustworthiness of the social environment, and according to Kawachi and Berkman (2000) it is a distinguishing feature of social cohesion.
Social cohesion is one aspect of the social environment of a neighborhood that has the potential to influence individual health and health-related behaviors such as physical activity. Social cohesion refers to two inter-related features of society: (1) the absence of latent social conflict; and (2) the presence of strong social bonds – often measured by levels of trust and norms of reciprocity (Kawachi & Berkman, 2000). The latter feature, also called social capital, has been recognized in various forms in the work of authors such as Coleman, 1988, Putnam, 1995, Bourdieu, 1986.
Kawachi and Berkman (2000) have proposed three pathways through which social cohesion may influence health at the neighborhood-level. These pathways include (1) social influence over health-related behaviors, (2) promoting access to services and amenities, and (3) directly affecting psychosocial processes. In light of prior research findings discussed above, neighborhood social cohesion could influence a young person's participation in physical activity via several potential mechanisms. The extent to which parents are involved with their children, as well as with other parents within a neighborhood, may improve the ability of a community to enforce healthy norms, including physical activity. Enhanced communication between members of a community may lead to quicker or more widespread adoption of healthy behaviors because neighbors know and trust one-another, and are in frequent contact (Kawachi, Kennedy, & Glass, 1999), a hypothesis in line with the influence of a culture favoring fitness suggested by the work of Karvonen and Rimpela (1997). Effective communication channels hypothesized to be found in neighborhoods with greater social cohesion may also increase parent or caregiver awareness about (and hence access to) programs and activities for young people (Coleman, 1988, Putnam, 1995). Community cohesion also facilitates collective action, such as lobbying for resources to improve the quality of local recreational spaces (e.g., converting disused railroads into bike trails). Social cohesion may thus theoretically enable better, more efficient use of available community resources for enhancing physical activity, thereby mitigating the negative effects of community disadvantage. Finally, cohesive neighborhoods may prove more conducive environments for physical activity because they limit negative conditions such a crime and disorder (Sampson, Raudenbush, & Earls, 1997). Based on the work of Molnar et al. (2004) neighborhood safety and visible disorder are also associated with participation in recreational programming.
In the present study, we examine evidence for associations between neighborhood social cohesion and parental reports of their child's participation in school or community-based recreational programs (i.e., extracurricular recreational activities such as sports, games or crafts) as well as the young person's self-reported physical activity participation, independent of individual characteristics. We hypothesized that young people would be more likely to participate in recreational programs and physical activity in neighborhood areas characterized by higher levels of social cohesion. We seek to expand the limited existing literature by investigating these hypotheses among a longitudinal sample of young people using measures of physical activity and recreational program participation.
Section snippets
Design
The Project on Human Development in Chicago Neighborhoods (PHDCN) is a longitudinal study of the development of child and adolescent criminal behavior, substance use, and prosocial behavior within the context of neighborhoods (Sampson et al., 1997). The PHDCN study design included both a city-wide community survey and a longitudinal cohort study conducted separately within a subset of neighborhoods (Earls & Buka, 1997). In the community survey, researchers first sampled city blocks within each
Results
Overall, the majority of respondents did not participate in school- or community-based recreational activity (66% and 86%, respectively). Fifty-eight percent of respondents were not recreationally active in either type of programming. The proportion of young people that did not participate in recreational programs differed by neighborhood education status (low- versus high-education areas) for school-based (68% versus 58%), community-based (90% versus 62%) and combined measures of total
Discussion
In this study, residing in a neighborhood with lower levels of neighborhood social cohesion at baseline was associated with increased likelihood that a young person did not participate in recreational programming (measured at baseline) and did not participate as frequently in more general types of physical activity measured at follow-up 2 years later. These associations between neighborhood social cohesion and participation in physical activity remained after adjusting for several
Conclusions
Neighborhood social cohesion was associated with participation in recreational programming and general physical activity, after adjusting for neighborhood educational status, the neighborhood availability of youth services, and several individual predictors of physical activity participation. This association may occur through neighborhood-level social and psychosocial processes that promote participation in physical activity. These results suggest that strategies that seek to enhance
Acknowledgements
This work was supported in part by the Centers for Disease Control and Prevention (Prevention Research Centers Grant U48/CCU115807) and the Harvey Fineberg Cancer Prevention Fellowship through the Harvard Center for Cancer Prevention and by grants from the John D. and Catherine T. MacArthur Foundation. This work is solely the responsibility of the authors and does not necessarily represent the official views of the Centers for Disease Control and Prevention or other granting institutions.
References (60)
- et al.
Collective efficacy and obesity: The potential influence of social factors on health
Social Science & Medicine
(2006) - et al.
Determinants of exercise among children. II. A longitudinal analysis
Preventive Medicine
(1998) - et al.
Violent crime and outdoor physical activity among inner-city youth
Preventive Medicine
(2004) - et al.
Urban small area variation in adolescents' health behaviour
Social Science & Medicine
(1997) - et al.
Social environment and physical activity: A review of concepts and evidence
Social Science & Medicine
(2006) Walking, exercising, and smoking: Does neighborhood matter?
Social Science & Medicine
(2000)- et al.
Ethnic, socioeconomic, and sex differences in physical activity among adolescents
Journal of Clinical Epidemiology
(1996) - et al.
Physical activity and social status in adolescence as predictors of physical inactivity in adulthood
Preventive Medicine
(2003) - et al.
A prospective study of the determinants of physical activity in rural fifth-grade children
Preventive Medicine
(1997) - et al.
Correlates of vigorous exercise in a predominantly low SES and minority high school population
Preventive Medicine
(1994)