The Paradox of Parks in Low-Income Areas

Abstract

Concerns about safety and perceived threats have been considered responsible for lower use of parks in high-poverty neighborhoods. To quantify the role of perceived threats on park use, we systematically observed 48 parks and surveyed park users and household residents in low-income neighborhoods in the City of Los Angeles. Across all parks, the majority of both park users and local residents perceived parks as safe or very safe. We noted apparently homeless individuals during nearly half of all observations, but very few instances of fighting, intimidating groups, smoking, and intoxication. The presence of homeless individuals was associated with higher numbers of park users while the presence of intoxicated persons was associated with lower numbers. Overall, the strongest predictors of increased park use were the presence of organized and supervised activities. Therefore, to increase park use, focusing resources on programming may be more fruitful than targeting perceived threats.

Keywords

parks physical activity safety collective efficacy mental health

Background

Parks are community resources that provide both space and facilities to support physical activity, but they are often underutilized (Cohen et al., 2013; Cohen et al., 2010; Kaczynski & Henderson, 2007). As a consequence of not using parks, many people do not get enough physical activity, which places them at greater risk for multiple chronic diseases, like heart disease, diabetes, and cancer (USDHHS, 2008). Parks in low-income neighborhoods are used less than those in high-income neighborhoods, but the reasons for this have not been fully delineated (Cohen et al., 2012). Perceived threats can be barriers to park use, and fears about crime, traffic safety, becoming injured, or being caught up in gang violence have all been cited as reasons some people avoid parks (Committee on Environmental Health, & Tester, 2009; Parks, Housemann, & Brownson, 2003; Shinew, Stodolska, Roman, & Yahner, 2013). A study of parks in New York City noted that although there may be more parks in lower income neighborhoods, there was lower social access because of higher crime, fewer park acres, and more noxious land uses (Weiss et al., 2011). Other contextual factors also influence park use including street connectivity (Kaczynski, Koohsari, Stanis, Bergstrom, & Sugiyama, 2014) and land use mix (Frank et al., 2012). Whereas one study indicated no difference in perceived accessibility to parks among different racial/ethnic groups (Carlson, Brooks, Brown, & Buchner, 2010), others have shown distinct differences in perceptions of park safety, with minority groups, including African Americans and Latinos, perceiving local parks as less safe (Boslaugh, Luke, Brownson, Naleid, & Kreuter, 2004; Tappe, Glanz, Sallis, Zhou, & Saelens, 2013).

Nevertheless, regardless of race/ethnicity or income level, positive community-level social factors may reduce fear if park users have confidence that community members are looking out for each other. Measures of collective efficacy suggest that when community members trust each other, have similar values, and would intervene on behalf of one another, people may feel safer and enjoy better health (Sundquist et al., 2014; Sampson, Raudenbush, & Earls, 1997). Parks constitute an important component of the social fabric in communities and at least two studies have found associations between parks or park use and collective efficacy (Broyles, Mowen, Theall, Gustat, & Rung, 2011; Cohen, Inagami, & Finch, 2008). Furthermore, parks have been associated with mental health benefits, possibly due to exposure to nature, positive social interactions that occur in parks, or more directly as a consequence of physical activity (Sturm & Cohen, 2014).

Another layer of influence on park use and people’s perception of parks involves management practices that affect the social environment. The degree to which parks are staffed, schedule programming and events, and maintain conditions, landscaping, and renovate facilities may play a large role in drawing users to parks and potentially overcoming perceived threats (Dolash, He, Yin, & Sosa, 2015; Shinew, Stodolska, Roman, & Yahner, 2013). Because parks are generally used less in low-income neighborhoods, we conducted a study of parks in high-poverty area to understand the relative importance of perceived threats as well as the role of collective efficacy on objectively measured park use.

Method

Conceptual Framework

To consider park use, we used relevant portions of a conceptual model of the role of parks in public health (Bedimo-Rung, Mowen, & Cohen, 2005). Specifically, we considered the factors that influence frequency of use and nonuse as influenced by two broad categories: the characteristics of potential park users and the environmental characteristics of parks themselves. Environmental characteristics include park features (size, facilities, and programming), condition (maintenance and incivilities), accessibility, aesthetics, safety (perceived and objective), and policies (management and budget). User characteristics such as age, gender, race-ethnicity, socioeconomic status, and residential location can influence park use at both the intra- and interpersonal levels (Bedimo-Rung et al., 2005).

Data on park use and park social environment

As part of an ongoing randomized controlled intervention trial, we collected baseline data in 48 parks in neighborhoods with a poverty level above the median for the city of Los Angeles (>19% households in poverty) between June 2013 and August 2014. These 48 parks comprised almost a complete census of eligible low-income area parks with recreation centers in the City of Los Angeles. A few parks were excluded because of location (in a housing project with limited public access) or safety concerns. We mapped each park and divided it into distinct target areas for observation. In each park, we counted all park users in every target area following a modified System for Observing Play and Recreation in Communities (SOPARC) protocol in which we recorded the activity level, gender, and perceived age and race/ethnicity grouping for each person separately (Cohen et al., 2011). We conducted observations on six randomly scheduled days (3 weekdays and 3 weekend days) during three different times of the day over a 6-month period (18 observation hours per park).

We also conducted intercept interviews with adult park users and with local residents living <1 mile from park. We randomly selected 30 households around each park, 10 within each of three strata—less than ¼ mile from the park, ¼ to ½ mile, and ½ mile to 1 mile. We conducted door-to-door surveys and asked respondents about their use of the park, their health, and included items to measure perceptions of the social environment in the park by adapting two subscales from Sampson’s index of collective efficacy–one subscale measuring perceived informal social control and the other measuring perceived social cohesion and trust (Sampson et al., 1997). The social cohesion and trust items were “people around here are willing to help each other,” “people in this park can be trusted,” “people in this park generally don’t get along with each other,” and “people in this park do not share the same values.” The items were answered on a scale of 1 to 5 from strongly agree to strongly disagree with a series of statements. Informal social control included the statements, “people in the park would intervene if children were spray-painting graffiti” and “people in the park would intervene if a fight broke out in the park.” The responses were on a 5-point scale from very likely to very unlikely.

We also included the Kessler-6 (Cairney, Veldhuizen, Wade, Kurdyak, & Streiner, 2007), which asks about symptoms of depression and anxiety as a measure of mental health. Perception of safety was measured by asking, “In general, how safe do you feel the park is?” We also asked about the safety of the neighborhood in which the park was located. Respondents answered whether they agreed/disagreed with the statements, “there is a high crime rate in this neighborhood,” “the crime rate in this neighborhood makes it unsafe to go on walks during the day,” and “the crime rate in this neighborhood makes it unsafe to go on walks at night.” The safety items are adapted from Saelens et al. (2012).

Other survey items were adopted from already validated scales including the measures of safety, based upon Saelens, Sallis, Black, and Chen (2003) and the self-report of park visits, which were validated by Evenson, Wen, Golinelli, Rodriguez, and Cohen (2013). The self-report of physical activity is based upon the International Physical Activity Questionnaire (IPAQ; Craig et al., 2003).

Contextual measures of perceived threats, incivilities, and park conditions

After each round of observing park users and their characteristics (3 times/day), trained field staff (all females, who observed in pairs) documented possible threatening or unpleasant situations and other conditions that might discourage park use. First, they noted whether they saw individuals who were smoking cigarettes or appeared to be intoxicated, instances of fighting, and groups of people who were intimidating to them. They also noted whether there were apparently homeless individuals (defined as persons who had many belonging with them, including suitcases, backpacks, trash bags, carts, or sleeping materials). Finally, they recorded other park conditions, including the presence of food vendors in and around the park and ongoing construction that interfered with park use. Our pilot study of these contextual measures demonstrated high interrater reliability (>.90).

Data Analysis

We calculated descriptive statistics for park use, park characteristics, and for the park user and resident surveys at both the individual level and aggregate park level. We summed the presence of intimidating groups, fighting, and apparent gang members (each of which was rare alone), to a single variable named “interpersonal safety issues.” We used logistic regression to analyze the relationship between people’s perception of safety and measures of mental health and collective efficacy, controlling for respondents’ characteristics as well as the fixed effects of parks. Park users and residents were analyzed separately because they represent different populations and were recruited differently. We fitted a generalized linear model between park use and park-level predictors, including park characteristics, park conditions, and survey measures aggregated at the park level. We used the negative binomial distribution to account for variance inflation in the number of park users. Time of day and the day of the week were modeled by binary indicator variables (e.g., morning, afternoon, weekday, weekend) to allow for flexible time trends in park use. Repeated measures in each park were handled by the generalized estimating equation (GEE). Observations refer to one complete rotation of recording all the target areas in a park.

Results

Field staff made 818 visits to the 48 parks and counted over 67,000 park users during their observations and surveyed 1,445 park users and 1,592 residents who lived within a mile radius of the parks. Table 1 presents the park characteristics and observed park conditions as well as observed characteristics of park users. The parks were relatively large (M = 8.3 acres), and all had full-time staff and were equipped with multiple facilities, including a gymnasium, a playground area, and a variety of courts and fields. Park users were largely Hispanic and male and, compared with the general population, included a greater proportion of children and teens and fewer seniors. Two thirds of park users were observed being sedentary. On average, 75.4 park users were observed during any given hourly observation, ranging between 0 and 431 persons. Field staff rarely observed people fighting or any intimidating groups, which together, occurred in only 2.8% of all observations and in only 12 of the 48 parks.

Table 1.

Characteristics of 48 Study Parks in Low-Income Neighborhoods.

Variable	M	SD
Acres	8.3	6.5
Population (1-mile radius; 10,000)	4.6	1.8
Poverty rate	25.7	8.1
No. of accessible target areas	26.8	14.4
No. of target areas with supervised activity	0.4	0.8
No. of target areas with organized activity	0.7	1.0
No. of users per observation (round/day)?	75.4	72.9
Weekly use (estimated person hours)^a	7,389	7,139
Observed park users
Variable	n	% of total
Male	44,632	66.1
Female	22,914	33.9
Seniors	1,711	2.5
Adults	34,607	51.2
Teens	11,114	16.5
Children	20,112	29.8
Sedentary	45,408	67.2
Moderate	17,901	26.5
Vigorous	4,235	6.3
Apparent race/ethnicity
African American	8,586	12.7
Latino	52,650	78.2
White	3,057	4.5
Asian/Other	3,076	4.6
Park contextual measures
Variable	% total observations with condition	No. of parks with condition (total = 48)
Interpersonal safety concerns (gangs, intimidating groups, conflicts)	2.8	12
Intoxicated persons	8.4	27
Persons smoking	7.1	27
Vendors in parks	30.0	38
Vendors outside parks	31.0	39
Construction	13.9	19
Homeless individuals	49.8	43

Assuming a park was usable 14 hr a day and for 7 days a week.

We observed only a small number of instances of people appearing intoxicated (8.4% of observations)¹ and smoking (7.1% of observations) in 27/48 parks. In contrast, homeless individuals were seen in nearly all the parks (43/48) and during nearly half the hourly observations.

Table 2, which compares the sociodemographic characteristics and survey responses of park user and neighborhood household survey respondents, shows that park users were younger than the household respondents, had a lower educational attainment, and a higher percentage were male. A lower percentage of park users were African American and White than household respondents, and a larger percentage of park users were Hispanic than household respondents. Response rates were an average of 41% for park users and 83% for household respondents. Park users were 2.8 times more likely than household respondents to visit the park at least once a week (83% vs. 29%, p < .01) and they reported engaging in slightly longer exercise bouts (23 vs. 21 min, p < .01). More park users (86%) thought the parks were very safe or safe compared to household respondents (78%). They also rated their health as slightly higher, but mental health scores did not differ between park users and household respondents. Park users and household respondents had similar perceptions of social cohesion and trust, but park users were more likely to think that other park users would intervene to help out when needed (i.e., higher perceptions of informal social control).

Table 2.

Survey Respondent Characteristics and Measures.

	Park users (n = 1,445)	Household respondents (n = 1,592)	p value
Respondent demographics
% male	48.6	37.7	<.01
Age group (%)
18-29	20.6	19.3	.18
30-39	36.4	22.3	.21
40-49	20.9	21.6	<.01
50-59	13.1	23.1	<.01
>60	9.0	13.8	<.01
Self-reported race/ethnicity
% African American	6.1	10.1	<.01
% Hispanic	87.5	73.3	<.01
% non-Hispanic White	3.4	10.2	<.01
% Asian	1.6	1.2	.35
% Other race or ethnicity	1.4	5.2	<.01
Education
% less than high school	35.1	30.4	<.01
% high school graduate/GED	41.1	34.9	<.01
% greater than high school	23.9	34.8	<.01
Respondent park use and perceptions
% using park once a week or more	83.3	29.1	<.01
No. of days of park use in the past week	2.8	0.9	<.01
No. of weekly exercise sessions	2.5	2.3	.03
Duration of usual exercise session (minutes)	23.1	21.0	<.01
% self-rated health (good to excellent)	81.9	78.1	<.01
% saying park is safe or very safe	86.3	78.3	<.01
Average perception of neighborhood safety (scale 1-4, higher is safer)	2.9	2.9	.97
Social cohesion and trust (scale 1-5, higher is more trust)	3.0	3.0	.33
Informal social control (scale 1-5, higher is more control)	3.0	2.9	<.01
Mental health (Kessler-6; scale 1-5, higher is better mental health)	4.6	4.6	.70

Note. GED = General Educational Development.

Table 3 reports the multivariate model predicting perceived safety in parks. Among household respondents, African Americans were more likely as those of other race/ethnicities to perceive parks as safe (log odds 2.13, p < .04). Household respondents with less than a high school education were least likely to perceive the park as safe (log odds −1.73, p < .01). Household respondents who thought that other park users would intervene to help others were more likely to perceive the park was safe (log odds 1.15, p < .01).

Table 3.

Model Estimates for Respondents’ Perception of Safety in Parks.

	Household respondents			Park users
Variable	Estimates (log odds)	SE	p	Estimates (log odds)	SE	p
Race/ethnicity
African American	2.13	1.06	.04	—^a	—	—
Hispanic	0.25	0.70	.72	−0.41	0.79	.60
Others	1.14	1.03	.27	−0.18	1.00	.86
Non-Hispanic White	—	—	—	—	—	—
Female	0.09	0.42	.84	−0.80	0.28	<.01
Age	0.02	0.02	.37	0.00	0.01	.83
Education
Less than high school	−1.73	0.71	.01	0.24	0.50	.63
High school graduate/GED	−0.10	0.42	.80	0.50	0.33	.12
Greater than high school	—	—	—	—	—	—
Self-rated health
Excellent	0.27	1.33	.84	1.08	0.93	.24
Very good	0.50	1.22	.68	0.69	0.81	.39
Good	0.23	1.16	.84	0.72	0.79	.36
Fair	−0.65	1.16	.58	−0.20	0.75	.79
Poor	—	—	—	—	—	—
Mental health index (Kessler-6)	0.09	0.34	.80	−0.30	0.25	.24
Social cohesion and trust	0.07	0.48	.88	1.41	0.28	<.01
Informal social control	1.15	0.28	<.01	0.29	0.17	.08
Vendors in parks	−0.02	0.07	.76	0.04	0.05	.41
Vendors around parks	0.17	0.07	.02	0.04	0.04	.30
Construction in parks	−0.15	0.07	.03	−0.05	0.04	.27
Homeless in parks	−0.09	0.04	.04	−0.06	0.03	.08
Interpersonal safety issue in parks	−0.49	0.23	.03	−0.28	0.14	.04
Intoxicated persons in parks	−0.22	0.13	.10	−0.10	0.11	.35
Smoking persons in parks	−0.22	0.22	.31	−0.38	0.12	<.01

An insufficient number of African American park users took the survey to estimate the race effect. GED = General Educational Development.

Park conditions associated with greater perceptions of safety among household respondents included the presence of vendors around parks. Conditions associated with lower perceptions of safety included construction, the presence of homeless people, and intimidating groups or fighting. Among park users, women were less likely than men to perceive the park as safe. Additionally, the presence of interpersonal threats, such as seeing intimidating persons, had a significantly negative relationship with the perception of safety. The presence of smokers was also negatively related to park users’ perception of safety.

Table 4 presents the models predicting the number of observed park users. Consistent with findings in the existing literature, supervised activities, organized activities, and accessibility were positively related to park use (all three p values <.01). Activities had a particularly high magnitude of association, as each additional supervised or organized activity was associated with about 25% additional park users. The presence of homeless people in parks was positively associated with park use (27.9% increase in the number of users, p < .01), as was the presence of food vendors in parks (38.7% increase, p < .01) and around parks (21.5% increase, p < .01). Factors negatively related to park use included the presence of intoxicated persons (20.7% decrease, p = .02), and construction in parks (26.4% decrease, p = .04). None of the aggregated survey items (park-level perception of safety, mental health index, social cohesion and trust, and informal social control) had a significant relationship with the number of observed park users. We found similar results when the outcome was total minutes of moderate-to-vigorous physical activity (MVPA) occurring in the park.

Table 4.

Model Estimates for Park Use.

Variables	Estimates of log(M)	SE	p	Magnitude of multiplicative effect (%)
Park characteristics and conditions
Acres	−0.01	0.01	.29	−1.3
Percent households in poverty	−0.01	0.01	.53	−0.5
Population within 1-mile radius	0.04	0.03	.19	3.8
Presence of homeless	0.25	0.07	<.01	27.9
Interpersonal safety issues (gang, intimidating group, conflict)	0.07	0.18	.70	7.1
Persons smoking	0.07	0.10	.48	7.3
Persons intoxicated	−0.23	0.10	.02	−20.7
Areas under construction	−0.31	0.15	.04	−26.4
Food vendors in park	0.33	0.10	<.01	38.7
Food vendors around park	0.19	0.07	<.01	21.5
No. of accessible target areas	0.03	0.01	<.01	3.2
No. of supervised activities	0.22	0.04	<.01	24.8
No. of organized activities	0.22	0.06	<.01	25.2
Aggregated survey respondent measures
Household: Safety perception	0.74	0.77	.34	109.2
Household: Mental health index	0.51	0.56	.36	66.6
Household: Social cohesion and trust	1.74	1.22	.15	467.9
Household: Informal social control	−0.70	0.39	.08	−50.2
Park user: Safety perception	−0.97	0.87	.26	−62.1
Park user: Mental health index	0.32	0.46	.49	37.6
Park user: Social cohesion and trust	−0.64	0.65	.32	−47.3
Park user: Informal social control	−0.09	0.30	.76	−8.8

Because the presence of intoxicated persons was negatively related to park use, we used model similar to that in Table 3, we also examined a model to predict the presence of intoxicated persons in parks (data not shown). The only significant predictor was collective efficacy among park users, such that for every one unit increase in perceived social cohesion and trust (perception that others have the same values and are trustworthy), there was a 20.7% decrease in the likelihood of seeing intoxicated persons in the park (log odds ratio = −0.36, p = .01).

Discussion

We found significant differences in the relative importance of environmental and user characteristics with respect to park use and park-based physical activity, with environmental features having stronger associations (Bedimo-Rung et al., 2005). With the exception of the presence of intoxicated persons, most incivilities and potentially perceived threats in parks were either positively or not at all related to observed use of these low-income neighborhood parks. Most surprising was the finding that the presence of homeless persons was apparently not a barrier to park use. The parks studied are relatively large, about 8 acres, and at any given time, we counted an average of 75 park users. It’s possible that given the large spaces, potential threats or incivilities may not be encountered by park users who stick to specific target areas. For example, if only using the playground, the homeless people in the picnic area may be too far away to be considered a threat. However, homeless persons may deliberately choose parks with more people because of the relative safety of having many individuals around. On the other hand, they may possibly choose parks that are generally perceived as more accessible and pleasant and thus are used more. Similarly, there was a relationship between vendors and the number of park users, and it is more likely that the people in the park attract vendors, rather than the other way around.

Because this is a cross-sectional study, it is not possible to be confident about the direction of associations. Nevertheless, it isn’t plausible that homeless individuals would attract other park user–although food vendors might. However, it is possible that intoxicated persons might drive park users away—as is the likelihood that intoxicated persons might choose parks with few other people, considering that others might not tolerate intoxicated behaviors.

A study of minority park users in Chicago also found that they were highly likely to perceive their local parks as safe, but were also concerned about being in the park after dark or in poorly lit areas (Gobster, 2002). In particular, the Mexican American adults interviews were concerned about gang activity in the parks, but they also recognized the important role of parks as a setting for social and physical activities (Stodolska, Shinew, Acevedo, & Izenstark, 2011). A qualitative study of Mexican American adolescents found that although youths were concerned about crime in their neighborhoods, they adopted strategies that allowed them to take advantage of parks and recreational spaces at times when they perceived the area was safe– for example, during daylight hours and when there were many others present (Stodolska, Shinew, Acevedo, & Roman, 2013).

At the park level, there was no substantial variation in average informal social control and social cohesion and trust, collective efficacy and little variation in average mental health, and neither aggregated measure was associated with the number of park users. The construct of collective efficacy is intended to be a group aggregated variable (Sampson et al., 1997), however, it has been used at the individual level in some studies (Broyles et al., 2011; Lindblad, Manturuk, & Quercia, 2013). At the individual level, we did see associations with perceptions of safety in the hypothesized direction. Greater collective efficacy, in particular the subscale of informal social control, was associated with greater perceived safety. The only relationship between collective efficacy and park use was at the individual level. When collective efficacy was higher, the likelihood of encountering intoxicated persons in the park was lower.

As the study was limited to a single city and the majority of people observed and interviewed were Hispanic, the findings may not be fully applicable to other populations. The low prevalence of seeing people smoking in the parks was impressive—especially because it occurred less frequently than observations of intoxicated persons. At the time of our study, both smoking and drinking alcohol were illegal in the parks, but the ban on smoking was relatively new. The low smoking rate could be a reflection of California having an overall very low rate of smoking compared with other states.

There was a very strong relationship between programmed activities, both supervised and organized, and the number of park users. This was much stronger than any relationship with perceived threats, a finding seen in other studies of youth who are attracted to activities like team sports (Cohen et al., 2010; Perry, Saelens, & Thompson, 2011). In the qualitative study cited above, Mexican American adolescents stated that the presence of supervised activities made the parks safer and increased their own participation in physical activity (Stodolska et al., 2013). The effects of supervision are very large, given that each additional supervised activity was independently associated with 24.8% more park users and each organized activity was associated with 25.3% more park users. There is room for many more supervised and organized activities–each park had an average of 26.8 target areas where supervised activities could potentially occur, yet at any given time, there were fewer than one such activity occurring. It is unknown whether simply offering activities is sufficient to draw more park users. User characteristics themselves may be barriers; for example, a population with few resources may have less leisure time than higher income groups. If residents rely on mass transit, it will likely take them longer to get to a park than they have a private vehicle. If they have more than one job or a family that requires child care, busy schedules may limit park use (Carlson et al., 2010).

The relatively low prevalence of concerns about safety among residents and park users is supported by the limited threats observed in the parks. As the majority of residents and park users considered their local parks as being safe or very safe, it might be more fruitful for park systems to increase park use by focusing resources on programming and stimulating the demand for increased use, rather than on targeting perceived threats.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by NHLBI R01HL114283

Notes

Author Biographies

Deborah A. Cohen is senior scientist at RAND and the principal investigator of several studies on parks and physical activity. She is the author of A Big Fat Crisis—The Hidden Forces Behind the Obesity Epidemic and How We Can End It.

Bing Han is a statistician at the RAND Corporation. His research interest includes biostatistics and its application in public health and other public policy areas.

Kathryn P. Derose is senior policy researcher at RAND and focuses on understanding health inequalities and developing community-based and policy solutions to address them.

Stephanie Williamson is a research programmer at RAND. She has more than 17 years of experience with developing analytic databases and analyzing data on a range of project topics, including workers compensation, student achievement, teacher effectiveness, and reserve recruiting. She is the lead programmer, database architect, and data manager for primary data collection efforts for several studies addressing neighborhood impacts on health.

Terry Marsh is a project associate at RAND with many years’ experience managing field research in urban parks. His research interests span education, nutrition, and physical activity.

Laura Raaen is a research assistant at RAND. Her research interests include parks, physical activity, public health, and epidemiology.

Thomas L. McKenzie, emeritus professor of exercise and nutritional sciences, San Diego State University, is the developer of numerous observation systems to assess physical activity and its contexts in homes, schools, and community settings.

References

Bedimo-Rung

A. L.

Mowen

A. J.

Cohen

D. A.

(2005). The significance of parks to physical activity and public health: A conceptual model. American Journal of Preventive Medicine, 28(2, Suppl. 2), 159-168. doi:10.1016/j.amepre.2004.10.024

Boslaugh

S. E.

Luke

D. A.

Brownson

R. C.

Naleid

K. S.

Kreuter

M. W.

(2004). Perceptions of neighborhood environment for physical activity: Is it “who you are” or “where you live”?. Journal of Urban Health: Bulletin of the New York Academy of Medicine, 81, 671-681.

Broyles

S. T.

Mowen

A. J.

Theall

K. P.

Gustat

Rung

A. L.

(2011). Integrating social capital into a park-use and active-living framework. American Journal of Preventive Medicine, 40, 522-529. doi:10.1016/j.amepre.2010.12.028

Cairney

Veldhuizen

Wade

T. J.

Kurdyak

Streiner

D. L.

(2007). Evaluation of 2 measures of psychological distress as screeners for depression in the general population. Canadian Journal of Psychiatry, 52, 111-120.

Carlson

S. A.

Brooks

J. D.

Brown

D. R.

Buchner

D. M.

(2010). Racial/ethnic differences in perceived access, environmental barriers to use, and use of community parks. Preventing Chronic Disease, 7, A49.

Cohen

D. A.

Han

Derose

K. P.

Williamson

Marsh

Rudick

McKenzie

T. L.

(2012). Neighborhood poverty, park use, and park-based physical activity in a Southern California city. Social Science & Medicine (1982), 75, 2317-2325. doi:10.1016/j.socscimed.2012.08.036

Cohen

D. A.

Inagami

Finch

(2008). The built environment and collective efficacy. Health & Place, 14, 198-208. doi:10.1016/j.healthplace.2007.06.001

Cohen

D. A.

Lapham

Evenson

K. R.

Williamson

Golinelli

Ward

McKenzie

T. L.

(2013). Use of neighbourhood parks: Does socio-economic status matter? A four-city study. Public Health, 127, 325-332.

Cohen

D. A.

Marsh

Williamson

Derose

K. P.

Martinez

Setodji

McKenzie

T. L.

(2010). Parks and physical activity: Why are some parks used more than others? Preventive Medicine, 50(Suppl. 1), S9-12. doi:10.1016/j.ypmed.2009.08.020

10.

Cohen

D. A.

Setodji

Evenson

K. R.

Ward

Lapham

Hillier

McKenzie

T. L.

(2011). How much observation is enough? Refining the administration of SOPARC. Journal of Physical Activity and Health, 8, 1117-1123.

11.

Committee on Environmental Health, & Tester

J. M.

(2009). The built environment: Designing communities to promote physical activity in children. Pediatrics, 123, 1591-1598. doi:10.1542/peds.2009-0750

12.

Craig

C. L.

Marshall

A. L.

Sjöström

Bauman

A. E.

Booth

M. L.

Ainsworth

B. E.

. . . Oja

(2003). International Physical Activity Questionnaire: 12-country reliability and validity. Medicine & Science in Sports & Exercise, 35, 1381-1395.

13.

Dolash

Yin

Sosa

E. T.

(2015). Factors that influence park use and physical activity in predominantly Hispanic and low-income neighborhoods. Journal of Physical Activity and Health, 12, 462-469. doi:10.1123/jpah.2013-0226

14.

Evenson

K. R.

Wen

Golinelli

Rodriguez

D. A.

Cohen

D. A.

(2013). Measurement Properties of a Park Use Questionnaire. Environment and Behavior, 45, 526-547. doi:10.1177/0013916512436421

15.

Frank

L. D.

Saelens

B. E.

Chapman

Sallis

J. F.

Kerr

Glanz

. . . Cain

K. L.

(2012). Objective assessment of obesogenic environments in youth: Geographic information system methods and spatial findings from the Neighborhood Impact on Kids study. American Journal of Preventive Medicine, 42, e47-e55.

16.

Gobster

P. H.

(2002). Managing urban parks for a racially and ethnically diverse clientele. Leisure Sciences, 24, 143-159. doi:10.1080/01490400252900121

17.

Kaczynski

A. T.

Henderson

K. A.

(2007). Environmental correlates of physical activity: A review of evidence about parks and recreation. Leisure Sciences, 29, 315-354.

18.

Kaczynski

A. T.

Koohsari

M. J.

Stanis

S. A.

Bergstrom

Sugiyama

(2014). Association of street connectivity and road traffic speed with park usage and park-based physical activity. American Journal of Health Promotion, 28, 197-203. doi:10.4278/ajhp.120711-QUAN-339

19.

Lindblad

M. R.

Manturuk

K. R.

Quercia

R. G.

(2013). Sense of community and informal social control among lower income households: The role of homeownership and collective efficacy in reducing subjective neighborhood crime and disorder. American Journal of Community Psychology, 51, 123-139. doi:10.1007/s10464-012-9507-9

20.

Parks

S. E.

Housemann

R. A.

Brownson

R. C.

(2003). Differential correlates of physical activity in urban and rural adults of various socioeconomic backgrounds in the United States. Journal of Epidemiology & Community Health, 57, 29-35.

21.

Perry

C. K.

Saelens

B. E.

Thompson

(2011). Rural Latino youth park use: Characteristics, park amenities, and physical activity. Journal of Community Health, 36, 389-397. doi:10.1007/s10900-010-9320-z

22.

Saelens

B. E.

Sallis

J. F.

Black

J. B.

Chen

(2003). Neighborhood-based differences in physical activity: An environment scale evaluation. American Journal of Public Health, 93, 1552-1558.

23.

Saelens

B. E.

Sallis

J. F.

Frank

L. D.

Cain

K. L.

Conway

T. L.

Chapman

J. E.

. . . Kerr

(2012). Neighborhood environment and psychosocial correlates of adults’ physical activity. Medicine & Science in Sports & Exercise, 44, 637-646.

24.

Sampson

R. J.

Raudenbush

S. W.

Earls

(1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918-924.

25.

Shinew

K. J.

Stodolska

Roman

C. G.

Yahner

(2013). Crime, physical activity and outdoor recreation among Latino adolescents in Chicago. Preventive Medicine, 57, 541-544. doi:10.1016/j.ypmed.2013.07.008

26.

Stodolska

Shinew

K. J.

Acevedo

J. C.

Izenstark

(2011). Perceptions of urban parks as havens and contested terrains by Mexican-Americans in Chicago neighborhoods. Leisure Sciences, 33, 103-126. doi:10.1080/01490400.2011.550220

27.

Stodolska

Shinew

K. J.

Acevedo

J. C.

Roman

C. G.

(2013). “I was born in the hood”: Fear of crime, outdoor recreation and physical activity among Mexican-American urban adolescents. Leisure Sciences, 35, 1-15. doi:10.1080/01490400.2013.739867

28.

Sturm

Cohen

(2014). Proximity to urban parks and mental health. Journal of Mental Health Policy and Economics, 17, 19-24.

29.

Sundquist

Hamano

Kawakami

Shiwaku

Sundquist

(2014). Linking social capital and mortality in the elderly: A Swedish national cohort study. Experimental Gerontology, 55, 29-36. doi:10.1016/j.exger.2014.03.007

30.

Tappe

K. A.

Glanz

Sallis

J. F.

Zhou

Saelens

B. E.

(2013). Children’s physical activity and parents’ perception of the neighborhood environment: Neighborhood impact on kids study. International Journal of Behavior Nutrition and Physical Activity, 10, Article 39.

31.

US Dept of Health and Human Services (USDHHS). (2008). Physical activity guidelines for Americans. Washington, DC: Author.

32.

Weiss

C. C.

Purciel

Bader

Quinn

J. W.

Lovasi

Neckerman

K. M.

Rundle

A. G.

(2011). Reconsidering access: Park facilities and neighborhood disamenities in New York City. Journal of Urban Health: Bulletin of the New York Academy of Medicine, 88, 297-310.