1 Research Background
Children are the future of cities and constitute a fundamental basis for sustainable social development. The Child Friendly Cities Initiative (CFCI) proposed by UNICEF in 1996, together with the subsequent Convention on the Rights of the Child, both focus on the necessity to use policy-driven and design-based strategies to protect the rights of children to participation and health in cities
[1–
2]. According to the United Nations’ statistics in 2020, there were approximately 298 million children in China, which was 21.1% of the overall population; of these, about 187 million lived in the urban regions, making up almost two-thirds (62.9%) of the total child population in the country
[3]. However, over the past three decades, rapid urbanization in China has progressively compressed the spaces available for children’s activities. Outdoor activity levels among school-aged children have declined markedly, contributing to a sustained increase in childhood obesity rates
[4–
5].
The CFCI has, in fact, been implemented in more than 400 cities around the world
[6]. Meanwhile, a growing amount of research attention has gradually shifted to children’s behavioral patterns in cities, their personal perceptions of the built environment, and the complex interactions between behaviors and environment
[7]. Existing evidence indicates that multiple underlying factors influence children’s outdoor activity participation. Among the most salient are age- and gender-related differences, various modes of guardian participation, and distinct patterns of time allocation
[8–
13]. Guardians play an essential role in children’s outdoor activities, as they not only are caregivers of children’s safety, but also shape their behavioral patterns
[14]. A growing body of research has highlighted that everyday parent–child interactions have gradually become an important and widely discussed topic in child behavior studies and environmental psychology
[15]. Moreover, as the trend of population aging continues, the proportion of older adults who serve as caregivers is also steadily increasing
[16–
17]. In addition, children’s daily outdoor activity duration and frequency exhibit strong temporal dependence. There are relatively consistent differences between weekdays and weekends, across different seasons and holidays, and in relation to guardian’s time allocation
[14,
18]. Clarifying the complex interplay among these factors is thus essential for refining child-friendly community environmental design and for systematically improving the overall quality of children’s outdoor activities.
Compared with more formal types of spaces, such as parks, plazas, and specially designed children playground facilities
[19], the so-called “loose spaces”
[20], which are part of daily life, including streets and alleys, the space around shops, and open activity areas, tend to be more successful in attracting children’s active participation and sustained use
[21–
23]. Nevertheless, comprehensive research on the actual behaviors of children and their guardians and how their social and spatial relations operate in these open spaces remains limited, especially in terms of understanding the interaction mechanisms that work where age, gender, and temporal dimensions intersect
[7,
14,
18,
22–
23]. In response to this gap, this research will pay close attention to child–guardian outdoor activities patterns in community loose spaces and examine how spatial distribution of such participation varies across different temporal contexts. Specifically, this study seeks to address the following research questions:
1) Do children’s age and gender differ in outdoor activity participation (e.g., the choice of activity spaces)?
2) Across different time periods, is children’s outdoor activity participation (e.g., activity range, spatial clustering) influenced by caregivers?
3) Does guardians’ gender affect childcare time allocation and spatial use preferences?
By examining the spatio-temporal behavioral characteristics of children’s and their guardians’ everyday activities, this study provides new empirical evidence to inform the design of child-friendly communities, which contributes more practical strategies for promoting children’s participation in outdoor activities.
2 Research Content
2.1 Study Area
The study was conducted in the FCC Central City Community, located in Fangshan District of Beijing. Owing to its distinct location and socio-economic characteristics, the community is a typical peri-urban neighborhood within a megacity. It accommodates a large cross-district commuting population and exhibits a pronounced temporal pattern characterized by high dynamic density during morning and evening peak hours, contrasted with relatively low levels of daytime occupancy
[24]. Based on these characteristics, the site south of Building 2 was selected as the study area (Fig. 1). It covers approximate 772 m
2 and was built gradually between 2017 and 2019. The site has numerous recreational amenities that appeal to children and their guardians, providing a foundation for analyzing outdoor activity patterns in loose spaces within communities.
2.2 Research Methods
2.2.1 On-site Observation and Spatio-Temporal Behavior Mapping
This study employed on-site observation in combination with spatio-temporal behavior mapping (STBM)
[25–
26] to examine the spatio-temporal behavioral characteristics of children and their guardians. Specifically, STBM method systematically records key attributes of observed participants, including gender, age, activity type, and time period, covering both weekdays and weekends, through continuous observation and structured data collection procedures
[27–
28]. The research process comprised four stages (Fig. 2).
1) Pilot study: assess the feasibility of the observation tools and recording protocol.
2) Data collection: involve systematic observation and recording of children’s and guardians’ behaviors.
3) GIS database establishment: geocode (spatialize) and integrate the observed behavioral data.
4) Data analysis: use ArcGIS to characterize spatio-temporal distributions and to examine correlational relationships among behavioral variables.
First, during the pilot study stage, a site survey was conducted and a detailed plan drawing was produced. Researchers also assessed the surrounding conditions, delineated the boundaries of the study area, and identified specific observation points (Fig. 3).
Second, at the data collection stage, systematic observations were carried out on March 9, 11, 15, 17, and 19, 2023, including three weekdays and two weekend days to capture variations in use patterns. Observations were conducted four times a day—morning (08:00–09:00), noon (11:00–12:00), afternoon (14:00–15:00), and evening (17:00–18:00). Within each time slot, panoramic photographs of the site were taken at a ten-minute interval, obtaining seven pictures per time slot. As no elevated vantage point was available, a tripod was used to capture multiple fixed-point shots, ensuring visual consistency and continuity across all observation periods.
For the GIS database establishment, all photographs were imported into ArcGIS for systematic data cleaning, classification, and coding. The image records were grouped according to gender, age, and activity type (i.e. guardian-accompanied playing and children’s free playing). To enhance the accuracy of age-classification, a machine-learning classifier based on WEKA (version 3.9.4) was employed, using visual cues from the photos (e.g., facial features, body posture, clothing style) to assist age recognition
[29–
30]. According to Jean Piaget’s theory of cognitive developmental stages
[31], participants were divided into two age groups: 0–7 years old and 8–18 years old. The younger group corresponds to early childhood, characterized by greater dependence on guardians; whereas the older group represents school-aged children with higher levels of independence, whose outdoor activity patterns are more strongly shaped by school schedules and institutionalized time structures.
Lastly, during the data analysis stage, quantitative statistical analysis was integrated with qualitative evidence (including STBM visualizations and questionnaire data) to systematically evaluate spatio-temporal behavioral differences between children and their guardians.
2.2.2 Scale Model Construction
To further investigate the influence of the built environment in shaping the everyday activities of children and their guardians, a questionnaire survey was conducted utilizing the Neighborhood Environment Walkability Scale (NEWS). The validity and reliability of this instrument have been widely recognized in previous studies
[32], and it has been extensively applied in urban design research as well as in practice
[33–
34]. The questionnaire survey was conducted offline, with primary caregivers (parents, grandparents, nannies, etc.) as respondents. The questionnaires were distributed primarily during afternoon and evening hours, with particular emphasis on the evening period, when outdoor activities involving children and their guardians were most frequent, thereby maximizing the the number of responses. The questionnaire comprised three main sections: 1) general information on the child and household, including the gender and age of children and guardians; 2) children’s behavioral characteristics within the community environment; 3) overall perception measurement of the community environment, encompassing five dimensions, namely transportation and travel, travel motivations, recreational activities, supporting facilities and services, and perceptions and preferences
[34]. All questionnaire items were rated using a five-point Likert scale, where 1 represents the lowest level of evaluation (very poor) and 5 represents the highest level of evaluation (very good).
During the data processing phase, the Analytic Hierarchy Process (AHP) was employed to structure and synthesize the multiple items of NEWS into a hierarchical analytical framework
[35]. To ensure the robustness of the weighting procedure, two urban design experts from Beijing Institute of Technology were invited to independently assess the relative importance of relevant factors. Subsequently, independent-samples t-tests were conducted to systematically examine the effects of gender differences on guardian’s behaviors, providing quantitative evidence of differences between male and female guardians in children’s everyday activities.
3 Results
3.1 Differences in Outdoor Activity Participation by Children’s Age and Gender
Table 1 illustrates the distribution of children’s outdoor activity participation across different time periods. Overall, outdoor activities were more frequent among younger children compared with the older ones. The total of 915 younger children were observed, comprising 92% of the total sample; and gender distribution in this age group was nearly equal, comprising 448 boys and 467 girls. Further analysis revealed a significant gender difference among the older group. Specifically, the number of boys engaging in outdoor activities was 1.8 times higher than that of girls (47 boys and 26 girls), indicating that boys in this age group were more active in activity engagement within the community.
Besides, there are significant differences in the number of boys and girls participating in outdoor activities at different times of the day. The number of boys and girls who engaged in outdoor activities during the morning and midday times was relatively balanced. However, a pronounced difference emerged in the afternoon and evening, with the number of girls increasing noticeably in the afternoon, whereas the number of boys rose significantly in the evening.
To further reveal the spatial distribution characteristics of gender differences, the research employed STBM to visualize children’s activity participation within the site (Fig. 4). The results indicated that the activity ranges of some boys extended beyond the boundaries of the central recreational facilities, and they were more frequently concentrated in peripheral areas, such as the plaza and shop entrances (Zones A to E), whereas some girls tended to congregate and play with more open and unobstructed spaces (Zone F). Notably, the proportion of boys in these peripheral areas was significantly higher than that of girls.
3.2 Spatio-Temporal Distribution of Caregivers’ and Children’s Outdoor Activities
The STBM results reveal distinct difference in spatial distribution of children’s activities and their guardian’s accompaniment at different time periods. In the morning, specifically, the number of children and their guardians was relatively small, and their spatial distribution was in a sparse dotted pattern, primarily concentrated along the southern and northern edges of the site (Fig. 5). At noon, spatial overlap between children and their guardians was relatively marked, primarily concentrated on the north and south of the site, which implies close interaction between children and their guardians during this time period. In the afternoon, although the spatial distribution of children’s activities did not change significantly, the spatial overlap between children and their guardians decreased sharply than that in noon. Similar to the noon period, more children engaging in independent play were recorded during the afternoon, indicating that guardians allowed children to engage more in independent activities during the daytime. By the evening, the spatial distributions of children and their guardians highly overlapped, with guardians’ distribution completely covering the children’s activity area, indicating that guardian supervision was most intensive during the evening period.
3.3 Gender Differences Among Guardians
Analysis of the data across all time periods shows that, although the number of male guardians increased significantly at noon and especially in the evening, female guardians still predominated the caregiving during children’s activity engagement (Table 2). Furthermore, the number of female guardians is generally higher than that of male guardians, particularly pronounced at noon and in the evening, which also indicate that females usually undertake greater caregiving responsibilities in children’s everyday activities. Further comparison between weekday and weekend data shows that the numbers of both male and female guardians increased substantially on weekends. However, despite the rise in male guardian participation, their numbers remained lower than those of female guardians.
Moreover, to further explore the potential impact of gender differences among guardians on caregiving behaviors, this study used the AHP method to incorporate different factors that might affect guardian behaviors, and invited experts to conduct pairwise comparisons of each factor and assign weights accordingly (Table 3).
Besides, independent-samples t-tests were conducted to examine the relationship between guardian’s gender and each factor (Table 4). The results indicate that there are statistically significant effects on gender differences across various dimensions. Firstly, there was a significant gender difference in transportation and mobility (t = 3.177, p = 0.002). Male guardians reported a higher mean score of 3.509 (SD = 0.741) than females’ 3.243 (SD = 0.618). Combining the higher-weighted item, “My daily commute to work is long, and I have little time to spend with my child after getting back home,” this finding suggests that male guardians are more prone to agree that longer commuting time reduces the time spent with children after work. This implies a more limited allocation of time between work and family responsibilities among male guardians, which in turn influences their participation in children’s everyday activities. Second, a significant gender difference was also observed in the dimension of travel motivations (t = −3.462, p = 0.001). Specifically, female guardians reported a higher mean score of 3.920 (SD = 0.710) than male guardians’ score of 3.613 (SD = 0.665), indicating that female were more involved in this dimension. Further examination of higher-weighted items, such as “In my family, the responsibility for taking children out for activities is mainly borne by women” and “I usually take my child with me when go groceries shopping for the family,” revealed the gender difference in children’s everyday travelling, accompaniment, and care-related activities in the family.
Altogether, female guardians play a crucial and sustained role in accompanying children’s everyday activities, and the division of household labor regarding children’s outdoor activities and recreation is primarily taken by the female. This further evidenced that in most families, women are generally considered as the primary guardians for children, especially within the context of traditional culture.
4 Discussion: Guidance for Design Practice and Policy
4.1 Child–Guardian Interaction Patterns Under Children’s Individual Differences
This study demonstrated that age differences among children lead to variations in their perceptions, needs, and behaviors within the community environment, which are consistent with previous research by scholars such as Amanda Bateman and Taina Laaksoharju
[36–
38]. Field observations reveal that children aged 0–7 years tend to be engaged in activities within semi-enclosed spaces with open view. Such spaces usually equipped with basic play amenities characterized by low complexity and low risk (e.g., slides, swings, rocking horses) and evenly-distributed, low-density facilities, making it easy for guardians to provide prompt and close care. This also supports Piaget’s theory that early childhood spatial cognition depends on specific physical boundaries
[31], which aligned with the characteristics of children at this stage whose cognitive and social abilities are not fully developed and whose perception of environment remains relatively simple
[39–
41]. In comparison, field observations suggest that children aged 8 to 18 years showed significantly greater activity demands, preferred games with rules and complexity, and larger spaces and more innovative recreational facilities. This indicates that children’s needs for the physical environment are constantly changing at different stages of their growth, and as spatial exploration and autonomy increase, their dependence on close companionship from guardians will decrease
[15]. These age differences also influence when and where guardians decide to take children out, which then shapes the way they engage and the decision-making patterns
[23].
Moreover, this study also revealed the different impact of physical environment facilities on children’s participation patterns at different time periods. This finding contrasts with the view proposed by Inaki Larrea
[42] and others, who suggest that affordance availability has a continuous and all-day promoting effect on children’s activity participation. In this context, affordance availability refers to material characteristics of the environment that can be perceived by children and transformed into play behaviors, for instance, flat open surfaces that enable various activities such as ball games. The results of this study indicate that, across all observation periods, the number of children aged 0–7 years in outdoor activities was significantly higher than that of children aged 8–18 years, especially from noon to evening (Table 1). This is primarily attributed to that younger children are at the stage characterized by sensory development and environment exploration, and thus their everyday activities tend to be more inclined to play in open outdoor spaces. In contrast, children aged 8 to 18 years are greatly constrained by school schedules and institutionalized time structures, resulting in the comparatively low overall level of children’s outdoor activities
[31]. It is noteworthy that the number of outdoor activity children was significantly less in the morning than in the other three time periods. This can be explained by the fact that mornings often coincide with peak periods for guardians’ work commuting and household labor, which result in limited opportunities for children to engage in outdoor activities. Taken together, these findings suggest that different time periods and age factors interact in the participation of children’s and guardians’ outdoor activities, mutually shaping temporal preferences, spatial use patterns, and levels of participation among different groups within the community environment.
Analysis of children’s activity distributions, together with field observations, indicates that boys tend to engage more frequently in exploratory and uncertain activities, demonstrating a stronger behavior inclination toward adventure and spatial exploration. For instance, boys are more likely to engage in loose spaces
[20], such as planted buffer zones, non-paved surfaces, and sloping areas. These spaces provided a wider range of activity and greater flexibility in interacting with the surroundings, while maintaining relatively manageable safety. In contrast, girls’ activities are more concentrated in spaces with clearly defined boundaries, such as fence facilities and areas with explicit safety demarcations. This pattern is consistent with the findings of Roger Hart, as well as Gill Valentine and John McKendrick amongst others on children’s spatial experiences
[11,
43]. Underlying these differences is the cultural construction of gender roles in spatial behavior
[44], namely, the societal expectation believed that “boys should be brave, exploratory, and willing to take risks,” whereas “girls should be cautious and protected,” which is reinforced through family education and neigbourhood interactions, thus potentially influencing children’s understanding and everyday use of space
[45–
46]. This difference reveals how socio-cultural factors play a crucial role in shaping gender behavior patterns, working in conjunction with the characteristics of the physical environment to shape children’s scope of participation, activity strategies, and spatial perception in community public spaces. This finding also underscores that the social construction of gender roles should be fully considered in community’s spatial design to influence children’s activity participation. For example, providing more open and flexible areas can support exploratory and adventurous activities often favored by boys, while offering engaging environment can promote girls’ play experiences within a safe area. Through this, it could enhance the diversity and flexibility of the space, while addressing guardians’ needs for visual supervision and safety control.
4.2 Spatio-Temporal Characteristics and Bidirectional Effects of Child–Guardian Interaction Patterns
This study has found that noon and evening are the peak hours for children’s outdoor activities and for guardians taking them out for activity participation. The spatio-temporal characteristics is shaped by multiple interacting factors. First, the abundant sunshine and warm weather in March create comfortable conditions, which significantly enhance the willingness of guardians and children to participate in outdoor activities. Second, influenced by socio-cultural factors, noon and evening periods regularly coincide with daily family activities such as grocery shopping and school-related travel. The accessibility and convenience of community amenities further influence the duration and frequency with which children and guardians occupy the community public space.
In addition, findings from on-site observations and questionnaire surveys showed that the caregiving behavior of guardians varied significantly at different time periods. At noon, guardians tended to maintain a moderate observation distance, creating a relatively free space for children to move around. This pattern was facilitated by the permeability and connectivity of the site, which ensured visual accessibility while preserving a degree of spatial enclosure, enabling guardians to effectively supervise children and encouraging them to independently engage with such spaces
[47]. In the afternoon, guardians more frequently occupied resting facilities or peripheral areas within the site, and their behavior pattern shifted from moderate companionship
[48] to low-intervention companionship
[49]. This not only reflects the balance between the guardians’ own needs for rest and the children’s play needs in their everyday engagement, but also highlights the role of spatial configuration in guiding guardians’ behavioral decisions. In the evening, as visibility decreased and the number of people increased, the lighting facilities in area became insufficient, which resulted in the more concentrated distribution of guardians’ behaviors, with a significant increase in interaction with children. This change indicates that safety is the dominant factor influencing guardians’ evening activity decisions, prompting a more proactive companionship in children’s night-time activities
[50] and intervention strategies aimed at mitigating potential risks.
The above findings not only demonstrate the interplay between spatial features, environmental factors, and behavioral patterns, but also suggest that child–guardian interactions are not characterized by one-sided dependence. Rather, they constitute an adaptive, intergenerational interaction pattern that emerges through everyday practices and mutual adjustment
[17,
51]. From a design perspective, the provision of spaces with a certain degree of enclosure can greatly enhance children’s sense of safety and support the continuity of their activities, while simultaneously improving guardians’ visual access and caregiving efficiency. Thus, two adjacent but independent areas can be connected through continuous green belts and landscape fence to form transitional edges
[52]. These transitional spaces not only achieve a balance between visual accessibility and sense of security, but also encourage children’s participation and their exploratory drives within the site.
4.3 Gender-Role Differentiation in Chinese Families: Women’s Dominant Role in Caregiving
This study found that there are significant gender differences between male and female guardians in performing household labor. By contrast, female guardians generally undertake a greater share of caregiving duties, and even when experiencing the same commuting pressure as men, female tend to fulfill their caregiving duties to a greater extent. Thus, the current study believed that the relationship between gender differences in commuting time and family caregiving cannot be explained solely by individual choices, but more determined by the gender role expectations shaped by social culture. This difference reflects the long-standing stereotypes and institutionalized perceptions of the division of household labor between men and women
[53], whereby women are commonly perceived as the “inherent” providers of family care, while men are more associated with economic provision and social roles.
This gender difference in caregiving behavior presents new challenges for the design of community recreational spaces. First, given the primary role of female guardians in daily childcare, they spend more time in community spaces and engage in more diverse activities. Improving basic supporting facilities, such as seating, resting areas, and shaded spaces, can thus enhance their comfort during caregiving. Second, considering that male guardians’ caregiving activities often occur in non-working hours, particularly in evenings and on weekends, the design of community recreational spaces should also take temporal patterns into account. For instance, existing research indicates that fathers are more likely to build close relationships with children through sports-related activities
[54–
55]. Therefore, increasing exercise areas or fitness facilities suitable for adults and children can promote parent–child interaction, thereby encouraging active participation by male guardians, and balancing the gender division of household labor.
5 Conclusions
This study takes a site in suburban Beijing as an example to explore the behavioral patterns of children and their guardians in outdoor activities within community loose spaces. The findings demonstrate that children’s spatial needs change significantly with age. Specifically, children aged 0–7 years show a strong reliance upon guardians and preference for safe, semi-enclosed, small-scale spaces, whereas those aged 8–18 years exhibit greater independence and a need for more complex activity spaces. This not only influences children’s activity engagement, but also shapes the way guardians accompany and guide them. Furthermore, children’s gender differences in outdoor activities are significant that boys tend to concentrate their activities in edge areas of community space, while girls more frequently engaged in play within defined spatial areas. Simultaneously, interactions between children and guardians show temporal differences, with noon and evening emerging as peak periods. These interactions dynamically change over time, reflecting the combined influence of family structure, daily routines, and socio-cultural factors.
Besides, in contrast to previous studies that primarily focused on children as independent subjects, the current study incorporates guardians into the empirical framework of child–environment interactions through the combining STBM with quantitative analysis. This study proposes a contextualized design strategy based on spatio-temporal behavioral characteristics, which strengthens the guiding significance of the research for urban design practices, while providing an empirical and scientific basis for policy-making in child-friendly community construction, especially in terms of spatial equity and guardian participation, thereby contributing to more inclusive and socially responsive urban environments.
However, this study also has certain limitations. First, the study is based on a single-community case, which constrains the representativeness and broader applicability of the findings. Second, both on-site observations and questionnaire surveys were conducted within a relatively short time periods, which is insufficient to reflect the seasonal or long-term dynamic characteristics of children’s activity participation. Third, although the measurement scales incorporated gender and guardian role factors, they still fall short in considering other potential variables, such as socioeconomic status and educational background. Future research could address these limitations by adopting multi-case comparisons across different community types and urban contexts to enhance the applicability. Moreover, integrating top-down policymaking with bottom-up user feedback would help better understand how institutional frameworks and everyday practices interact in shaping child-friendly environments. Such an approach would contribute to promoting spatial equity and social inclusion, more effectively responding to the diverse needs of families and communities, and ultimately supporting the healthy development of children through more inclusive and participatory community loose space design.
PDF
(2756KB)
