This study examined associations between frequency of visiting and being active in certain outdoor public recreation spaces with social connectedness among adolescents. No associations were observed for frequency of visitation to the nearest public park, trails/path, beach/lake/river/creek, or sports features (e.g., basketball court, soccer field). Additionally, no associations were observed for frequency of physical activity in participants’ street, on a local street or bike/foot path, or at a nearby park/reserve with social connectedness.
Contrary to our hypothesis, visiting outdoor public recreation spaces in a usual week was not related to social connectedness among adolescents. Participants reported frequency of visitation to the nearest outdoor public recreation spaces to their homes. While at least half of the sample reported visiting the locations at least once in a usual week, it could be that adolescents also visited outdoor public recreation spaces located farther away from home that attracted their peers to “hang out” and socialise. Previous qualitative evidence among adolescents has indicated that parks in a convenient location near school, friends’ houses, shops, and public transport would encourage visitation (31), and adolescents have reported that they would be prepared to visit high-quality parks located farther away from home (31, 47). Future studies should therefore consider examining adolescents’ visitation to the closest and the most frequently visited outdoor public recreation spaces in their neighbourhood.
Additionally, rapid advancements in technology have recently led to more ways to socialise (e.g., social media) than in previous years (48, 49), and it is possible that these different ways and nature of connecting socially may have impacted our findings. For example, adolescents may be commonly visiting outdoor public recreation spaces with friends but mostly interacting socially through social media on their mobile devices as opposed to face-to-face. It is also unknown what activities they were engaging in whilst in the settings, and it is possible that these activities may not have involved social interaction. A further reason for the lack of association between visiting outdoor public recreation spaces and social connectedness may be that we did not assess other public settings that are also important for adolescents’ social connectedness, such as plazas, civic squares and shopping strips (15, 16, 50). In addition, we did not measure duration of visits to outdoor public recreation spaces, so it could be that adolescents in our sample were not spending sufficient time at the locations for there to be an association with social connectedness.
Previous qualitative research has found that while the presence of other people attracts adolescents to visit urban parks, minimal interaction between different user groups was often reported (51). This suggests that co-presence with other user groups may not translate into meaningful contact, which may have been the case in the present study. Thus, although outdoor public recreation spaces provide opportunities for incidental interactions (52, 53), it could be that adolescents who visited their nearest public parks, beach/lake, trails, and sports features at least once in a usual week were not experiencing the types of social interactions necessary for fostering social connectedness. Moreover, Gibson’s Affordance Theory postulates that environments afford various behaviours and actions (54), and the perceived opportunities for use are relational to the needs and interests of users in the environment (55, 56). Adolescents’ different needs and interests may explain our findings. Given that outdoor public recreation spaces support many different activities (not limited to social interaction and physical activity), future research may benefit from examining how affordances and context of use of these settings influence adolescents’ social connectedness.
Our findings also showed that the frequency of performing physical activity in the examined outdoor public recreation spaces was not significantly associated with social connectedness, and there are several plausible explanations for this. Firstly, at least half of the sample did not visit a nearby park, their street, or a local street or path at least once per week, so it may be that they were not active frequently enough at these places for there to be an association with social connectedness. Additionally, regarding streets and paths, it could be that these areas may be used more often for active transportation (57, 58) – walking, running, and/or cycling for travel (58) – than for active recreation. It is possible that active travel on streets/paths (even if accompanied) may be less conducive to the social interactions needed to foster adolescents’ social connectedness. Further, evidence has indicated that adolescents engage in a range of physical activities in parks, such as organised sports, playing games and/or on play equipment, walking, and running (32, 34). Given that adolescents value opportunities for being active while socialising with peers (59, 60), it could be that the specific types of physical activities undertaken, as well as whether they involve accompaniment and interaction with others (e.g., organised sports versus walking alone), matter for supporting social connectedness. Future research should consider exploring associations between types of physical activity undertaken in outdoor public recreation spaces with social connectedness among adolescents.
The Social Connectedness Scale was used to assess social connectedness in this study, and the scores of adolescents in our sample were comparable to scores in other studies, which have used modified versions of the same scale with adolescents (61) and young adults (43, 62). However, there is currently no universal instrument for assessing social connectedness and other measures are available (7, 10, 63–65). It has also been recognised that adolescents develop social relationships across multiple environments (66, 67), and social connectedness has been differentiated from connectedness in different social domains/contexts (e.g., school, family, community, peer) (5, 10, 68). Thus, the lack of associations observed may have been due to the scale employed in this study being a global measure of connectedness rather than social connectedness in the community, and adolescents’ use of outdoor public recreation spaces might be more closely linked to community social connectedness, which is an area for future research. Moreover, when seeking to understand people-place relationships, place-related constructs (e.g., sense of community, sense of place, place attachment) have commonly been used (69–71). Future research may also benefit from examining place constructs in relation to adolescents’ use of outdoor public recreation spaces.
To our knowledge, this is the first study to examine associations of frequencies of visitation to and physical activity in outdoor public recreation spaces with adolescents’ social connectedness. Additional strengths include the examination of multiple outdoor public recreation spaces and the roughly equal split of males and females in the analytical sample. However, as a cross-sectional study that relied on self-report, causal inferences between variables cannot be drawn, and the data may be subject to recall or social desirability bias (72, 73). Future studies may consider ecological momentary assessment to sample and examine participants’ social experiences and behaviours in real time (74). Further, responses were dichotomised as visiting less than once vs at least once in a usual week. Different associations may have been found for those adolescents who visit on most/all days of the week. It could be that different associations may have been found for those adolescents who visit regularly (e.g., most/all days of the week). It is also possible that participants considered both indoor and outdoor sports facilities when completing the items regarding the nearest basketball court and other fields/courts. Future research may also benefit from using objective measures for assessing visitation and physical activity in outdoor public recreation spaces in addition to assessing the duration and context of use.
Furthermore, there was a high proportion of missing responses for the predictor variables (particularly visitation to different outdoor public recreation spaces). While complete case analyses are commonly used to address missing data in epidemiological research (75), this approach can bias findings and lead to losses in statistical power and precision (75, 76). The analyses of patterns of missingness performed showed that data were at least ‘Missing at Random’ (MAR) (75). Specifically, adolescents with a higher social connectedness score were less likely to have missing data compared to those with lower social connectedness. Consequently, our findings may have been biased due to the lack of power to detect associations resulting from the reduced analytical sample (75, 76). Additionally, when data are not ‘Missing Completely At Random’ (MCAR), complete case analyses can bias regression coefficients. This bias increases with the difference between means of the observed and missing cases, and with the proportion of missing cases (77).