Background
Urban settings include a diversity of characteristics that, on the one hand, can be hazardous to human health (e.g., exposure to pollution, motor vehicle traffic, and crime), while on the other hand, can provide health-supportive functions (e.g., sanitation, high-quality drinking water, the experience of nature, and opportunities for social interaction and physical activity) [
1,
2]. In particular, urban parks are important for population health because they can impact human health via multiple pathways [
3‐
5] and they can serve as a publically-available recreational facility for many people over multiple generations. Urban parks provide opportunities for socializing, contemplation, and physical activity. Moreover, urban parks can be designed and managed to link people with each other and with nature across a range of different demographic, socioeconomic, and cultural backgrounds [
6].
The importance of parks for enhancing the quality of human life has long been recognized. Indeed, urban planning practices since the late 1800s have included provision for park space so as to separate industrial and residential land use, and to serve as central gathering places for civic activities [
6]. More recently, local or municipal governments have devoted much of their services and significant proportion of financial resources in the creation, renewal, and upkeep of parks within urban settings. These investments, furthermore, often depend on transfers of resources from higher levels of government and engagement of citizens as tax-payers and volunteers. The provision of functional, attractive, and safe park space is considered to be an important characteristic of healthy communities and cities [
7,
8]. Despite their recognized importance, evidence regarding which characteristics of urban parks influence behaviour, in particular physical activity, is only just beginning to be systematically collated and analyzed from the perspective of promoting population health [
9‐
11].
Regular participation in physical activity, especially at recommended levels, can reduce the risk of cardiovascular disease, diabetes, depression, obesity, and some cancers [
12,
13]. Moreover, physical activity in urban parks and neighbourhoods can promote social connectedness and interaction, which are also positively linked with physical and mental health outcomes [
14,
15]. Existing quantitative and qualitative evidence suggests that parks support physical activity, but not uniformly across the life course [
9‐
11]. Youth activities in parks often involve participating in unstructured activities such as play [
16,
17] while for adults of different ages, walking is a common park-based physical activity [
18]. Sedentary behaviour (e.g., sitting on a park bench) is also a common park-based activity [
19], which may entail walking to, within, and from the park [
20]. Variation in the patterns and levels of park-based activity are likely influenced by the interaction between the characteristics of park users and characteristics of the park environment, as well as by the built and social characteristics of the surrounding neighbourhood [
21‐
23].
Bedimo-Rung et al. [
24] offer a conceptual framework for understanding how park characteristics influence physical activity, and this framework points to a need for synthesizing data derived from multiple methods. Yet most studies do not explore reasons why such relationships exist or integrate any qualitative data; rather, they focus on the influence of built environmental features on physical activity by adjusting statistically for the influence of other contextual and visitor characteristics [
10,
11]. An approach that incorporates both quantitative and qualitative approaches to investigate the determinants of park-based physical activity may provide greater insight into nuanced interactions between park visits, physical features, and socio-demographic characteristics. In addition, whereas qualitative as well as quantitative data about park use can be generated
in situ[
6], qualitative studies of physical activity in urban parks have been dominated by face-to-face and focus group interviews conducted off-site [
9]. Thus, for this multiple case study of four urban parks in a Canadian city, we apply quantitative and qualitative methods of direct observation to study patterns of park-based physical activity. The objective of this study was to document and account for patterns of park-based physical activity in ways that could meaningfully inform urban planning and policies to promote health. The University of Calgary Conjoint Human Research Ethics Board granted ethics approval for this study.
Discussion
Among the four park study sites, we found variations in the patterns of use, characteristics of users, and the types of activities undertaken. In general, the patterns of use in all four parks combined reflected findings regarding the gender and age characteristics of urban park visitors elsewhere [
33‐
36]. However, contrary to previous evidence [
18,
19,
34,
37], we found that sedentary activities were not the most common park activity. Rather, walking and dog-walking were the most common activities for our sample of parks. Our findings also suggest that patterns of park activities are associated with the physical attributes within each of the parks and the surrounding neighbourhoods. Few other studies have integrated multiple quantitative and qualitative approaches (i.e., structured observation, fieldnotes, and photography) to interpret the interrelationships between the characteristic of parks and surrounding neighbourhoods and park use and activity [
38].
Despite being unable to collect data for the entire day, we found that patterns of park use were consistent with patterns of use found elsewhere [
33,
34,
39], whereby that the majority of visits take place in the afternoons and evenings. Patterns of weekday and weekend (Saturday and Sunday) use differed between parks but not in a consistent way. Across the four parks, male visitors were more common than female visitors, while for three of the four parks, users were more likely to be adults than children or youth – a pattern that is consistent with evidence collected using systematic observation [
33‐
36]. While we were unable to quantify the interaction between gender and activity types within the parks due to sample size, elsewhere men have been found to be more physically active within park settings than have women [
19,
35,
36]. This pattern is similar to differences in physical activity found between men and women in population-based studies [
40,
41]. Creating more opportunities for physical activity within parks could contribute to more park-based physical activity among women, thereby reducing gender disparities in physical activity. To encourage more park visits and more park-based physical activity among women, design and management strategies could include improvements associated with safety (e.g., lighting and surveillance) as well as greater opportunities for social interaction [
9].
Park visitors are often more likely to be adults than children or youth [
18,
19,
34,
42]. In one of our parks (i.e., Martindale) however, over one-half of all users observed were children or youth. This park did not include physical attributes, such as playgrounds, typically associated with children’s use and physical activity within parks [
21]. Moreover, within this same park, participation in “play” activity was low compared with other activities which is noteworthy as we expected to see more “play” in settings with high numbers of children. Our qualitative data suggest that the youth who visited this park were unaccompanied by adults and that they visited a central secluded wooded area to ‘hang-out’ and socialize. Furthermore, monuments within the Martindale park were used for play activity, such as climbing. This finding stands in contrast with the West Hillhurst park, which included a playground, where one-third of visitors were children or teenagers, and where one-third of activity observed was “play”. Our findings suggest that playgrounds, while encouraging children to be physically active, are not the only reason why children visit parks. For older children and teenagers, park attributes that provide places for friends to meet and socialize are likely just as important [
9]. Introducing age-appropriate playground equipment, creating and maintaining sports fields and courts and open spaces, as well as the provision of areas for socializing that balance privacy with safety, might encourage more children and teenagers to be physically active by visiting parks [
9,
33].
Sedentary behaviour is commonly observed in parks [
18,
19,
34,
37]. Kaczynski et al. [
19] found that just over half of observed park users participated in sedentary activity while Floyd et al. [
18] and Cohen et al. [
37] found that almost two-thirds of park users were sedentary. Contrary to this evidence, our findings suggest that for the four parks included in our study, sedentary activity was less common than other activities. Nevertheless, observed participation in sedentary activity differed between the parks and the reasons for these differences are in part explained by the qualitative data. Located within Taradale and Martindale parks are storm water drains and pools of stagnate water that facilitate mosquito breeding, which is particularly severe during the early months of the Calgary summer. The mosquito populations within these two parks were particularly numerous during the data collection (“unbearable” as pointed out by LB and KS while conducting the observations in these locations), likely deterring people from spending extended periods of sedentary time (i.e., sitting, standing, remaining stationary) within these locations. Furthermore, the parks were relatively small in area and the layout of the open space was not particularly suitable for organized or unorganized sports activities, which often attract (sedentary) spectators [
37]. Seating was also entirely absent (in Martindale) or in short supply (in Taradale). Nevertheless, even those observed to be sedentary in parks likely derived some health benefit from the physical activity accumulated whilst traveling to and from the park. Active transportation is a common mode of transportation for reaching public recreational opportunities [
43]. Creating and redesigning parks to include attributes and amenities that support both passive and active pursuits, therefore, could increase levels of physical activity in urban populations [
44,
45].
People may visit parks more frequently when accompanied [
46,
47]. We found that almost two-thirds of park visitors were with another person and that about one-half of children visiting parks were accompanied by parents or guardians. Compared with other parks, West Hillhurst included the highest proportion of group visits. Because of its playground, “playing” was also the most common activity in West Hillhurst park relative to the other parks. Together, these findings might suggest that some parents and caregivers are physically active while accompanying their children to the playground, but are themselves inactive in parks while their children play. The physical attributes within this park including benches and shade close to the playground and the limited amount of pathway might facilitate sedentary activity among these caregivers. Besenyi et al. [
21] found that for adults and seniors higher levels of park-based physical activity energy expenditure occurred on paths and courts compared with other areas (open spaces, picnic areas), while for children, higher levels of park-based physical activity energy expenditure occurred in playgrounds. Floyd et al. [
33] found that less shade was associated with a greater likelihood of park users walking and participating in vigorous-intensity activity versus sedentary behaviour. Shores and West [
22] also found sedentary activities to be more common in parks with shelter and picnic areas. One suggestion for park planners might be to include attributes that will support physical activity for adults (e.g., fitness zones, walking paths) near playgrounds so parents or caregivers can still supervise their children while they are physically active themselves.
Dogs, in our study, contributed positively to patterns of physical activity. After walking without a dog, dog-walking and other dog-related pursuits, such as playing fetch, were the most commonly observed activities. These findings extend a growing body of research in health promotion linking dog-walking with physical activity [
48]. This line of inquiry is highly relevant to population-level patterns of physical activity because in Canada, dogs reside in one out of every three urban households [
49], and similar statistics have been reported in other Western countries [
50,
51]. Dog-walking is also becoming a common form of physical activity in some non-Western countries, such as China and Japan [
52,
53]. Nevertheless, high levels of dog-ownership can also impact negatively on physical activity for both dog-owners and non-dog-owners, given that ill-controlled dogs and litter from dog-waste can deter people from visiting parks [
51], so policies and management strategies need to take dog and owner behaviour into account.
Our method of systematic observation differed from methods used elsewhere (i.e., SOPARC, SOPLAY). Unlike other approaches, we attempted to record park user information and activities for the entire park and not for separate activity zones within the park. This approach was facilitated because of our focus on neighbourhood parks that were typically smaller than other open spaces (e.g., community parks, regional/national parks, sports fields). Because we centered our observations on persons, not on multiple activity areas in the same park, double counting of visitors was minimized. When multiple activity areas serve as the basis for observations, any visitors who moved between activity areas during observation periods could be double-counted [
39]. Whenever possible, we observed each park user for a full two minutes so as to document multiple activities (if undertaken within such a short time-frame) during the same park visit. The time frame of two minutes was selected to balance the need to capture the range of activities undertaken by a single individual and the need to include as many cases as possible. In the Taradale, Martindale, and West Hillurst parks, most visitor observations lasted for two minutes. In Meadowlark however, less than one-half of all visitors were observed for two minutes. Meadowlark included a short pathway that connected a residential area and large mall both of which were surrounded by major roads with high volume traffic. Many walkers and cyclists observed in the Meadowlark park were passing through en route to other destinations, thus the fact that fewer visitors were observed for the entire two minutes reflects the activity patterns within, and the design of, the park. During times when the volume of visitors was low, observation of visitors may have begun immediately as they entered the park – thus their activity would likely have been recorded as walking or cycling but underestimating other types of activity (e.g., sedentary activity or play). Similarly, visitors who were already in the park and were being sedentary, playing, or undertaking dog-related activity likely entered the park via walking or cycling, so our results may underestimate walking and cycling. Whereas our findings are a cross-sectional representation of summer park activities in four parks only, they suggest that the relationships between park policies and physical activity are complex.
Despite limitations, our case study approach provided a rich description of the activities and settings and their interaction within the four parks. Indeed, such a detailed description would not have been possible using a single methodological approach or data source. The unique findings from our case study highlight that while policy-related physical attributes in some parks appear to encourage certain types of behaviour (e.g., the presence of playgrounds and frequency of play amongst children), in other parks, the relationship between context and behaviour was less than straight-forward. Thus, the physical and social characteristics as well as the history of the community in which a park is situated are potentially important influences on park use. The case study approach might provide useful information for park developers and planners in creating or modifying parks, for example, to encourage specific types of activities or to encourage more use among particular types of visitors [
6]. Furthermore, this approach is highly relevant to innovations in park planning and management, which often begin with pilot or demonstration projects that require consideration of ambient policy, sociocultural and geographical contexts [
6].
Competing interests
The authors declare that they have no competing interest.
Authors’ contributions
GRM contributed to the study design, and led the quantitative analysis and manuscript writing. MR contributed to the study design, and led the qualitative analysis and contributed to the manuscript writing. AM contributed to the study design and the manuscript writing. KS collected data, and contributed to the qualitative analysis and to the manuscript writing. LB collected data and contributed to the qualitative and quantitative analysis. All authors contributed to the interpretation of results and provided final approval for the manuscript.