Home Yard and Neighbourhood Physical Environment Latent Class Profiles and Preschooler’s Physical Activity Behaviour: Findings from the PLAYCE Study

Abstract

Home yard and neighbourhood physical environment typologies were identified using data from 1,634 preschool children. Relationships between neighbourhood typologies and preschooler physical activity behaviour were investigated. Physical activity was measured using accelerometry and parent-reported outdoor play. Traffic exposure, street connectivity, public transport, residential density, land use mix, vegetation, blue space and home yard size were measured objectively. Parent perceptions of the neighbourhood were collected using the Neighbourhood Environment Walkability Scale for Youth. Latent profile analysis identified six neighbourhood profiles: ‘No blue space’, ‘Higher blue space’, ‘Higher public transport & destination mix’, ‘Connected residential’, ‘Established’ and ‘Semi-rural’. Living in ‘Higher blue space’ and ‘Semi-rural’ profiles was associated with more minutes of total daily physical activity compared with living in ‘Higher public transport & destination mix’. Further research to understand how urban form varies within and between neighbourhoods by population group and geographic location is needed to guide urban planning and policy.

Keywords

neighbourhood home physical activity preschool latent profile analysis

Introduction

Daily physical activity including outdoor play is critical for preschool children’s physical and mental health and development (Carson et al., 2017; Lambert et al., 2019; Lee et al., 2021; Tremblay et al., 2015; Veldman et al., 2021). Many countries along with the World Health Organisation have developed 24-Hour Movement Guidelines for the Early Years (Australian Government Department of Health, 2019; Tremblay et al., 2017; World Health Organization, 2019). These guidelines recommend children aged 2 to 5 years accumulate at least 3 hr of physical activity per day including energetic play (moderate-to-vigorous physical activity), with 3–5-year-olds needing at least 60 min of energetic play per day. However, many young children do not meet these guidelines, negatively impacting their current and future health and development (Active Healthy Kids Australia, 2022).

The social-ecological model (Stokols, 1996) considers the effect of an individual’s social and physical environment on health behaviours such as physical activity. Three principles underpin social-ecological theory: (1) Health and illness are determined by multiple environmental and personal factors; (2) A person’s home is interrelated to other life settings; and (3) There is an emphasis on a multiple-level and multidisciplinary perspective of health. The level of compatibility between a person and their surroundings is seen as an important predictor of health. The social-ecological model also accepts that the interaction between environment and behaviour can be in multiple directions (Stokols, 1996). Several individual (e.g., child sex and age), social (e.g., parenting practices) and physical environment (e.g., park access) factors have been correlated with preschool children’s physical activity behaviour (De Craemer et al., 2012; Hinkley et al., 2008).

Given the importance of the physical environment to early child health and development (Christian et al., 2015; Villanueva et al., 2016), a comprehensive understanding of how features of the neighbourhood built environment uniquely influence young children’s physical activity behaviour is needed to inform inclusive urban policy and design. An umbrella review (Carlin et al., 2017) of the physical environment determinants of physical activity behaviour across the life course identified that preschool children’s total physical activity was positively associated with home yard space (Robinson et al., 2024), outdoor play equipment at home (Armstrong et al., 2019) and access to neighbourhood parks and playgrounds (Carlin et al., 2017; Gemmell et al., 2022). For example, objectively measured home yard size was positively associated with outdoor play in a large sample of 1,558 preschool children in the PLAYCE cohort study (Robinson et al., 2024). The finding was consistent across high and low household level socio-economic groups (Robinson et al., 2024). Overall, there are relatively few studies of the relationship between neighbourhood-level built environment features, other than park access, and preschooler physical activity behaviour (Carlin et al., 2017). Further studies using objective measures of both the built environment and movement behaviour are required to confirm whether the relationships concur with those found in older school-aged children.

For preschool children who have very little independent mobility, parent’s perceptions of the built environment may play a critical role in shaping their physical activity behaviour. As such, studies conducted in adults show perceptions of the neighbourhood built environment may have a greater influence on physical activity behaviour compared with objective measures (Christian, Knuiman, et al., 2017; Giles-Corti et al., 2013). In a recent study parent perceptions of the neighbourhood environment moderated the relationship between objectively measured neighbourhood environment attributes and preschooler physical activity (Nathan et al., 2023). For example, objectively measured street connectivity was associated with fewer minutes of energetic play when parents perceived pedestrian and traffic safety to be below average (Nathan et al., 2023). These findings highlight the importance of including both parent perceptions and objective measures of the built environment in studies of preschooler physical activity behaviour.

Further, research on the influence of the neighbourhood built environment on physical activity behaviour should consider the combined and interrelated effects of multiple built environment attributes. An example is the walkability index, which includes the combined effects of three built environment attributes, street connectivity, land use mix and residential density (Giles-Corti et al., 2014). However, studies rarely consider the wider aggregated effects of multiple built environment attributes in a neighbourhood. The built environment attributes of a neighbourhood may or may not be correlated and could have synergistic or opposing effects on physical activity behaviour. Furthermore, the influence of different combinations of built environment attributes on physical activity behaviour is likely to be geographically specific. For example, the built environment characteristics of a low-density city such as Perth, Western Australia is very different to a high-density European city such as Copenhagen, Denmark.

It is important to examine the neighbourhood as a whole and compare the makeup of different types of neighbourhoods and their influence on preschooler physical activity behaviour. An innovative method to classify different types of neighbourhoods based on a broad set of neighbourhood attributes is latent profile analysis (LPA; Weden et al., 2011). Such cluster-based statistical techniques can be used to describe patterns of neighbourhood typologies based on a combination of measures of the built environment. Data-driven approaches reflect the pattern of attributes ‘on the ground’ and may therefore provide a fuller picture of neighbourhood environments. Many such studies have been conducted in physical activity with adults with some in adolescents and school-aged children but none to date appear to have been conducted with preschool children. Of the handful of studies conducted in school-aged children the findings have been mixed (DeWeese et al., 2018; Kurka et al., 2015; Timperio et al., 2017; Tu et al., 2016). A study examining patterns of food and physical activity environments using latent class analysis found that children living in neighbourhoods characterised by high residential density, more physical activity facilities, less parks and more healthy and unhealthy food outlets were more likely to walk to destinations (DeWeese et al., 2018). To our knowledge, no studies of neighbourhood typologies using both perceived and objective measures of the built environment and preschooler physical activity have been reported. Thus, this study sought to (i) identify neighbourhood typologies using both parent perceived and objective measures of the home yard and neighbourhood physical environment, and (ii) examine the association with preschool children’s device-measured physical activity and parent-reported outdoor play.

Methods

Study Population and Setting

This study used baseline data (2015–2018) from the Play Spaces and Environments for Children’s Physical Activity (PLAYCE) cohort study (Christian et al., 2016; Christian et al., 2025). The PLAYCE cohort study examined changes in young children’s movement behaviours, the influence of setting-specific (childcare/school, home, neighbourhood) physical, social, and policy environmental factors, and the impact on child health and development outcomes. Children, aged 2 to 5 years, were recruited from childcare centres, within the Perth metropolitan area of Western Australia. Parents provided written informed consent for themselves and their child to participate in the research. Ethics approval was granted by The University of Western Australia Human Research Ethics Committee (RA/4/1/7417 and 2020/ET000353).

Perth, the capital city of Western Australia, is inhabited by approximately 2.3 million residents. The city is characterised by urban sprawl spanning 125 kilometres north to south (6,420 square kilometres), constrained by the Indian Ocean to the west and the Darling scarp to the east. The population density across Perth varies greatly from area to area, however the average population density is 359.7 persons per square kilometre. Perth experiences a Mediterranean climate with hot dry summers and cool wet winters (Australian Bureau of Statistics, 2024).

Device-Measured Physical Activity and Parent-Report Outdoor Play

Children’s physical activity was measured using ActiGraph GT3X+ accelerometers (ActiGraph Corporation, Pensacola, FL USA) worn on the hip during waking hours for 7 days. Raw accelerometer data (30 Hz) was processed using a random forest machine learning model for preschool-aged children developed by Ahmadi, Pavey, and Trost (2020) to estimate daily time sedentary (sitting and lying down), light intensity activities and games (slow walking or ‘pottering about’, standing, standing arts and crafts), walking, running and moderate-to-vigorous activities and games (active games with balls, riding scooters/tricycles).

The random forest physical activity classification model (Ahmadi, Pavey, & Trost, 2020) used in the current study was trained and cross-validated in the foregoing manner: (1) Accelerometer data were collected on 31 preschool-aged children while completing a 20-minute free-living active play session (e.g., family home, park, local green space); (2) Accelerometer data was segmented into 15-second non-overlapping windows and 22 time and frequency domain features in the signal vector magnitude were extracted; (3) Windows were annotated with ground truth labels specifying one of five broad activity classes: sedentary, light-intensity activities and games, walking, running, and moderate-to-vigorous intensity activities and games. The children were video recorded during the free play sessions with a hand-held Go-Pro Hero 5 (GoPro, Inc., San Mateo, CA, USA) camera for subsequent direct observation coding. A two-step coding scheme was implemented in which each child’s movement behaviour was first coded for physical activity intensity using the Children’s Activity Rating Scale (CARS; Puhl et al., 1990) and then activity type based on a list of 23 developmentally appropriate physical activities for preschool-aged children; and (4) The random forest algorithm was trained on the fully annotated dataset, and the resultant models predictions were compared to ground truth using leave-one-out cross-validation.

Cross-validation data (Ahmadi & Trost, 2022) indicated that the random forest classification model predicted walking with acceptable accuracy. Accuracy [F-scores] for each activity class was as follows: 87.7% for SED, 88.2% for light-intensity activities and games, 78.5 % for moderate-to-vigorous activities and games, 79.4% for walking, and 82.6% for running, providing an overall weighted average F-score of 85.9% accuracy. The random forest model exhibited significantly higher agreement with preschooler measured physical activity intensity than cut-point methods and exhibited evidence of equivalence with directly observed time spent sedentary, light-intensity and moderate-to-vigorous physical activity (Ahmadi & Trost, 2022). Further details on the random forest physical activity classification model used have been previously published (Christian et al., 2024).

Non-wear periods were identified by summing the time periods in which the standard deviation of the accelerometer signal vector magnitude was <13 mg for ≥30 consecutive minutes (Ahmadi, Nathan, et al., 2020). Data validation was based on at least 3 weekdays and 1 weekend day with a minimum 8 hr wear time per day (Christian et al., 2016). Daily ‘energetic play’ (moderate-to-vigorous intensity physical activity) was calculated as the sum of minutes of walking, running, and moderate-to-vigorous activities and games. Daily total physical activity was calculated as the sum of minutes of light intensity activities and games and energetic play.

To complement device-based measures of physical activity and provide a more context-specific measure of young children’s physical activity behaviour parent-reported outdoor play was included. Parents completed the PLAYCE survey which included an existing established measure of outdoor play (Burdette et al., 2004). Parents were asked how much time their child spends playing outdoors on a typical day in two locations: in the yard or street around their house; and at a park, playground, or outdoor recreation area. The day was divided into three time periods: wake-up time until noon, noon until 6 pm, and 6 pm until bedtime. Within each time period, the amount of outdoor play was reported using a 5-point scale (0, 1–15, 16–30, 31–60, over 60 min) and were coded 0 through 4. Responses for each question were summed resulting in a score ranging from 0 to 24 (Burdette et al., 2004). The parent survey also collected information on child sex, age and parent education level (secondary school, higher education certificate, or university).

Home Yard and Neighbourhood Physical Environment Variables

Two home yard and nine neighbourhood physical environment variables were calculated using geographic information systems (GIS). These variables were calculated for the same years physical activity was measured (i.e., 2016–2017). Parent perceptions of the neighbourhood physical environment (six variables) were collected using the PLAYCE parent survey.

Outdoor Home Yard Environment

Total yard size (m²) was calculated using data provided by the Western Australian Land Information Authority (Landgate, 2017). It included any portion of the cadastre not part of the building footprint including both the front and back yard. Yard greenness was assessed using four-band (Red, Green, Blue, Infrared) high resolution stereo-ortho-imagery with vegetation classified using a Normalised Vegetation Index (NDVI) thresholding approach (>0.2). Data was provided by the Commonwealth Scientific and Industrial Research Organisation’s (CSIRO) Urban Monitor Program (Caccetta et al., 2012). Vegetation cover (m²) was calculated as the sum of grass (height: <50 cm), shrub (height: 50 cm to 3 m), and tree (height: >3 m) cover (Duncan et al., 2019) then calculated as a percentage of total yard size.

Neighbourhood Physical Environment

All GIS measures of the neighbourhood physical environment, except blue space, were based on a 1,600 m road network buffer (service area), centred around the child’s geocoded residential address. An adult-based road network buffer was chosen due to the very high level of dependency preschool children have on their parents/carers in relation to where they are physically active in their neighbourhood.

Neighbourhood vegetation was calculated in the same way as home yard vegetation: percentage of vegetation (grass, shrub, and tree) of the total service area. The percentage of total area of blue spaces (oceans, river ways, canals, estuaries, lakes, and swamps) within a 1,600 m Euclidean buffer (as the crow flies) from the child’s residential address was calculated (George et al., 2024). Euclidian buffers were used as blue spaces are more often accessed through trails, paths and adjacent beaches or green spaces (George et al., 2024).

All other neighbourhood built environment measures have been previously described (Christian, Ball, et al., 2017; Duncan & Boruff, 2023; Villanueva et al., 2013). Briefly, street connectivity was measured as the number of three-way nodes and number of one-way intersections (i.e., cul de sacs) per square kilometre in a service area. Low road traffic exposure was calculated as the percentage of the length of access (low traffic: ≤3,000 vehicles per day) roads of the total length of all roads in a service area. Public transport stops were measured as the total count of transport stops within a service area. Net residential density was measured as the number of residential dwellings per hectare of total land use area within each service area.

Two measures of land use mix were calculated using data from the Western Australian Land Use and Employment Survey (WA Department of Planning Lands and Heritage, 2022). A measure of service destination mix was calculated by counting the presence of eight different types of services (post office, bank, pharmacy, general practice/medical centre, dentist, community centre/hall, childcare centre, public library; range 0–8). A measure of food outlet mix was calculated by counting the presence of five types of food stores (supermarket, deli, speciality food including meats, fish, fruit, bread, or confectionary, petrol station, and other food stores including health food store; range 0–5).

Parental Perceptions of the Neighbourhood Physical Environment

The PLAYCE parent survey included items from a modified version of the Neighbourhood Environment Walkability Scale for Youth (NEWS-Y) to assess parent perceptions of the neighbourhood environment (Rosenberg et al., 2009). The modified version comprised six of the nine NEWS-Y subscales: (i) pedestrian and traffic safety, (ii) crime safety, (iii) land use mix access (e.g., access to shops, parks, public transport), (iv) street connectivity, (v) walking and cycling facilities, and (vi) neighbourhood aesthetics (Vanwolleghem et al., 2016). Thirteen additional items, which specifically related to young children, such as access to local play equipment, safety of walking with young children during the day and presence of footpaths on both sides of the street, were included across the subscales. Each item was scored on a four-point Likert scale (strongly disagree to strongly agree) with some items reverse-coded so that all items were in the same conceptualised direction. Higher scores represented more positive perceptions of the neighbourhood environment (Robinson et al., 2022).

Analysis

All 17 home yard and neighbourhood physical environment variables were first standardised using z-scores. Latent profiles of the home yard and neighbourhood physical environment and the association with physical activity were determined using the three-step latent profile analysis approach with a distal outcome (Bakk et al., 2013). Standardised home yard and neighbourhood physical environment variables were entered into the latent profile analysis. Models with 1 through 10 profiles were examined to determine the optimal number of profiles. Model selection was based on the log-likelihood (LL); the Akaike Information Criterion (AIC), Consistent Akaike Information Criterion (CAIC), Bayesian Information Criterion (BIC), and sample-size adjusted Bayesian Information Criterion (SABIC) information criteria; entropy, interpretability, posterior probabilities, and profile size. Lower information criterion (LL, AIC, CAIC, BIC, SABIC) indicate better fit, while higher entropy (>0.8) and higher average latent posterior probabilities (>0.9) indicate better fit (Sinha et al., 2021). Characteristics of the emerging profiles were considered to determine whether profiles were qualitatively and quantitatively distinguishable and made conceptual sense within the Perth, Western Australia geographical context. To aid interpretation, participants geocoded residential addresses were mapped onto the Perth metropolitan area by their assigned profile (colour coded). Children were included in latent profile analyses if they had valid age and sex and parent education data and home yard or neighbourhood physical environment variables (n = 1,634); the minimum number of home yard or neighbourhood physical environment variables a child had was five.

Second, children were assigned the most appropriate profile using proportional assignment adjusted for uncertainty when modelling profiles using the Bolck-Croon-Hagenaars (BCH) correction method (Bakk et al., 2013; Bakk & Vermunt, 2016). Third, linear regressions were used to examine the association between assigned latent profiles of the home yard and neighbourhood physical environment and preschooler’s daily minutes of energetic play and total physical activity (as well as the components of sedentary, light intensity activities and games, walking, running, moderate-to-vigorous activities and games) and outdoor play score, adjusting for child age and sex and socio-economic status as measured by parent education. For each physical activity variable, estimated mean differences (95% CI) between profiles and daily mean minutes for each profile are reported. The sample size for analysis of the device-based outcomes was n = 1,108 (after excluding participants who did a not meet accelerometer valid wear time criteria (Ahmadi, Nathan, et al., 2020) and n = 1,545 for parent-reported outdoor play. Data were prepared in Stata v17.0 and the three-step latent profile analysis was undertaken in LatentGOLD 6.0 using the cluster and step3 procedures (Statistical Innovations Inc., 2016).

Results

Overall, 52.1% of preschool children were boys and the median age was 3.3 years. Over half of parents had a university degree (56.5%), 27.7% had a higher education certificate and 15.8% had completed secondary school (Supplementary Table 1). On average, preschoolers engaged in 291.8 (standard deviation (SD) 63.2) minutes sedentary time, 334.3 (SD 46.5) minutes of light intensity activities and games, 17.5 (SD 8.3) minutes walking, 4.3 (SD 2.9) minutes running, 17.7 (SD 10.4) minutes of moderate-to-vigorous activities and games, 39.4 (SD 14.2) minutes of energetic play and 373.7 (SD 51.0) minutes of total physical activity per day. The mean outdoor play score was 9.2 (SD 4.7; Supplementary Table 1). Table 1 presents the summary descriptives of the home yard and neighbourhood physical environment.

Table 1.

Characteristics of the Home and Neighbourhood Physical Environment.

Home and neighbourhood physical environment characteristic	n	Mean (SD)	Median (IQR)
Outdoor home yard environment
Yard size (m²)	1,618	777.0 (2,970.2)	342.6 (257.2)
Yard vegetation (% of yard size)	1,618	30.2 (19.7)	29.0 (31.2)
Neighbourhood physical environment ^a
Low road traffic exposure (% of all roads)	1,624	69.0 (11.0)	69.2 (13.4)
Three-way intersection density (n per km²)	1,624	67.9 (24.2)	62.9 (24.6)
One-way intersection density (n per km²)	1,624	13.0 (7.4)	11.6 (10.1)
Number of public transport stops	1,624	37.4 (22.5)	33.0 (29.0)
Net residential dwelling density (n per hectare)	1,621	17.7 (13.3)	16.2 (5.4)
Count of types of service destinations^b	1,624	4.4 (2.5)	5.0 (4.0)
Count of types of food outlets^c	1,624	3.4 (1.5)	4.0 (3.0)
Blue space (% of service area)^d	1,624	5.7 (10.0)	0.0 (7.3)
Vegetation (% of service area)	1,623	25.0 (7.4)	24.1 (7.3)
Parent perceptions of the neighbourhood physical environment ^e
Pedestrian and traffic safety	1,493	2.9 (0.5)	2.9 (0.6)
Crime safety	1,483	2.5 (0.7)	2.4 (1.0)
Land use mix access	1,516	2.9 (0.5)	3.0 (0.8)
Street connectivity	1,511	2.8 (0.6)	2.7 (1.0)
Walking and cycling facilities	1,508	2.6 (0.7)	2.5 (0.8)
Neighbourhood aesthetics	1,513	3.2 (0.5)	3.2 (0.8)

Note. ^aNeighbourhood physical environment features were measured using a 1,600 m road network area around participant’s residential address.

Service destinations: count of the presence of eight types of services: post office, bank, pharmacy, general practice/medical centre, dentist, community centre/hall, childcare centre, and public library.

Food outlets: count of the presence of five types of food stores: supermarket, deli, speciality food (including meats, fish, fruit, bread, or confectionary), petrol station, and other food stores (including health food store).

Blue space: total area of oceans, river ways, canals, estuaries, lakes, and swamps within a 1,600 m Euclidean buffer around participant’s residential address.

Modified Neighbourhood Environment Walkability Scale for Youth (NEWS-Y) sub-scales: four-point Likert scale ranging from 1 = strongly disagree to 4 = strongly agree.

Latent Profiles of Preschooler’s Home Yard and Neighbourhood Physical Environments

Number of Profiles

Information criterion improved with increasing number of profiles, however the improvements to fit were minimal beyond six profiles (Supplementary Table 2). Entropy declined to three profiles then increased to 10 profiles (except for a decline at seven profiles) but generally remained high (>0.9) and indicative of high levels of classification prediction. Models with six or more profiles included at least one profile with fewer than 5% of the sample, and models with seven, nine, or 10 profiles included a profile with fewer than 2% of the sample. Models with four, five, and six profiles were investigated further for interpretability. The model with six profiles was selected as the best fitting model.

Characteristics of Profiles

Each profile was labelled based on its distinguishing physical environment feature/s relative to the other profiles and knowledge of the study location (Perth, Western Australia). Standardised scores for each profile are displayed in Figure 1, and unstandardised means are displayed in Table 2 and Figure 2. A spatial map of participants by profile is in Figure 3.

Figure 1.

Standardised home yard and neighbourhood physical environment variables scores by latent profile.

Table 2.

Characteristics of the Six Latent Profiles of Preschooler Home and Neighbourhood Physical Environments.

Profile number	1	2	3	4	5	6
Profile name	No blue space	Higher public transport and destination mix	Higher blue space	Connected residential	Established	Semi-rural
% of sample	(30.2%)	(23.2%)	(21.7%)	(17.0%)	(4.1%)	(3.8%)
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)
Outdoor home yard environment
Yard size (m²)	397.3 (163.6)	333.3 (169.1)	410.0 (163.1)	229.7 (80.8)	1,784.4 (2,871.5)	10,057.1 (11,618.6)
Yard vegetation (% of yard size)	31.0 (18.2)	33.2 (20.2)	33.1 (18.7)	17.5 (15.2)	35.6 (24.7)	39.6 (22.7)
Neighbourhood physical environment ^a
Low road traffic exposure (% of all roads)	65.1 (9.2)	68.8 (6.6)	68.1 (9.4)	79.7 (9.0)	60.1 (15.1)	66.0 (20.4)
Three-way intersection density (n per km²)	61.9 (13.5)	62.2 (14.5)	63.9 (14.8)	102.6 (23.9)	61.5 (24.9)	22.7 (13.7)
One-way intersection density (n per km²)	16.4 (7.9)	8.1 (3.9)	16.7 (7.9)	11.2 (4.4)	9.3 (4.1)	7.8 (5.6)
Number of public transport stops	32.9 (15.1)	62.8 (18.1)	30.4 (12.7)	22.5 (11.7)	56.3 (32.5)	5.2 (6.9)
Net residential dwelling density (n per hectare)	14.9 (3.3)	21.9 (4.8)	15.0 (2.5)	17.9 (3.4)	40.8 (56.2)	3.0 (2.6)
Count of types of service destinations^b	4.1 (2.2)	6.8 (1.1)	3.9 (2.1)	2.7 (1.9)	6.3 (2.1)	0.6 (1.0)
Food outlets^c	3.2 (1.3)	5.0 (0.0)	3.1 (1.1)	2.3 (1.2)	4.1 (0.9)	1.0 (0.2)
Blue space (% of area)^d	0.0 (0.1)	3.8 (6.4)	16.1 (13.3)	4.1 (5.9)	11.8 (13.6)	3.9 (8.6)
Vegetation (% of area)	24.2 (5.1)	25.8 (4.6)	27.3 (5.7)	18.2 (4.4)	32.1 (11.8)	38.3 (14.1)
Parent perceptions ^e
Pedestrian and traffic safety	2.9 (0.5)	2.8 (0.5)	2.9 (0.4)	3.0 (0.4)	2.7 (0.5)	2.6 (0.6)
Crime safety	2.4 (0.6)	2.5 (0.6)	2.7 (0.7)	2.5 (0.7)	2.7 (0.6)	2.8 (0.7)
Land use mix access	2.9 (0.5)	3.1 (0.4)	2.9 (0.5)	3.0 (0.5)	3.0 (0.5)	2.0 (0.6)
Street connectivity	2.6 (0.5)	3.0 (0.6)	2.7 (0.5)	3.1 (0.5)	2.9 (0.6)	2.3 (0.7)
Walking and cycling facilities	2.5 (0.6)	2.8 (0.6)	2.4 (0.7)	2.8 (0.6)	2.8 (0.7)	1.7 (0.7)
Neighbourhood aesthetics	3.0 (0.6)	3.3 (0.5)	3.3 (0.5)	3.2 (0.5)	3.3 (0.5)	3.5 (0.4)

Note. ^aNeighbourhood physical environment features were measured using a 1,600 m road network area around participant’s residential address.

Blue space: total area of oceans, river ways, canals, estuaries, lakes, and swamps within a 1,600 m Euclidean buffer around participant’s residential address.

Modified Neighbourhood Environment Walkability Scale for Youth (NEWS-Y) sub-scales: 4-point Likert scale ranging from 1 = strongly disagree to 4 = strongly agree.

Figure 2.

(A) Unstandardised home yard and neighbourhood physical environment GIS derived variables means by latent profile and (B) unstandardised home yard and neighbourhood physical environment GIS derived variables (service and food outlet types) and parent perception variable means by latent profile.

Figure 3.

Spatial distribution of preschoolers residence by their home and neighbourhood physical environment profile.

The first profile was labelled ‘No blue space’ and represented 30.2% of preschoolers and was characterised most notably by the absence of any blue space in the neighbourhood. Across all other variables, this profile had values close to the median. The second profile, ‘Higher public transport & destination mix’ (23.2% of preschoolers) was characterised by the highest number of public transport stops and count of service and food outlet destination types, relative to other profiles. The third profile, ‘Higher blue space’ (21.7%), was similar to profile one with values close to the median for most variables, however it had the most blue space compared with the other profiles. The fourth profile, ‘Connected residential’ (17.0%) had the smallest yard size, the highest proportion of low traffic roads, the highest three-way intersection density, and low counts of different types of services and food outlets. The fifth profile, ‘Established’ (4.1%), had large yards, a high number of public transport stops, the highest residential density, and high counts of different types of services and food outlets. Profile five was similar to profile two, except for having larger home yard outdoor areas and higher residential density. The final profile, ‘Semi-rural’ (3.8%), had very large home yard outdoor areas, low three-way intersection density, low residential density, few public transport stops, the highest proportion of neighbourhood vegetation, and few different types of service and food outlets. Profile six also had the lowest parent perception scores for land use mix access, street connectivity, and walking and cycling facilities. Overall, the profiles differed least on home yard vegetation (except for profile four ‘Connected residential’) and parent perception subscales (except for profile six ‘Semi-rural’).

Association Between Latent Profiles of Preschooler’s Home Yard and Neighbourhood Physical Environment and Physical Activity

After adjusting for socio-demographic factors, latent profiles of preschooler’s home yard and neighbourhood physical environment were significantly associated with daily minutes of total physical activity (Wald test p = .029). Children living in the ‘Higher blue space’ and ‘Semi-rural’ profiles had greater daily total physical activity than children in the ‘Higher public transport & destination mix’, by an estimated 13.5 (95% CI 4.1, 22.9) and 26.5 (95% CI 7.2, 45.8) minutes per day, respectively (Table 3). There were no other significant differences between the profiles for daily total physical activity. Children living in the ‘No blue space’, ‘Higher blue space’, ‘Connected residential’ and ‘Semi-rural’ profiles had greater daily minutes of light intensity activities and games than children in the ‘Higher public transport & destination mix’ (all p < .05; range 9–24 min/day; Supplementary Table 3). Profiles were not associated with daily minutes of energetic play (p = .190), walking (p = .220), running (p = .290), moderate-to-vigorous physical activities and games (p = .120), sedentary time (p = .094) or outdoor play score (p = .140).

Table 3.

Coefficients (β, 95% Confidence Intervals) and Mean Daily Minutes of Total Physical Activity and Energetic Play and Outdoor Play Score for Each Latent Profile of Preschooler’s Home and Neighbourhood Physical Environment.

	Total physical activity^a (mins/day)		Energetic play^b (mins/day)		Outdoor play^b (score)
Home and neighbourhood physical environment profile	B (95% CI)	Mean	B (95% CI)	Mean	B (95% CI)	Mean
1. No blue space	8.0 (−0.3, 16.3)	373.8	−1.2 (−3.5, 1.2)	38.5	−0.3 (−0.9, 0.3)	9.2
2. Higher public transport and destination mix	Ref	365.8	Ref	39.6	Ref	9.5
3. Higher blue space access	13.5 (4.1, 22.9)*	379.3	1.3 (−1.3, 3.9)	41.0	−0.3 (−1.0, 0.4)	9.2
4. Connected residential	9.1 (−0.6, 18.8)	374.9	−0.7 (−3.3, 1.9)	38.9	−0.9 (−1.7, −0.2)	8.6
5. Established	5.6 (−12.3, 23.6)	371.5	−3.0 (−7.3, 1.3)	36.7	0.2 (−1.2, 1.6)	9.7
6. Semi-rural	26.5 (7.2, 45.8)*	392.3	2.5 (−3.0, 8.0)	42.1	−1.1 (−2.4, 0.2)	8.4

Note. ^aTotal physical activity and energetic play n = 1,108 children who had valid accelerometer data, age, sex, parent education, and latent profile information.

Outdoor play n = 1,545 who had outdoor play score, age, sex, parent education, and latent profile information. Score range 0 to 24.

p < .05: Ref = profile with lowest average daily minutes of total physical activity.

Discussion

This study sought to identify neighbourhood typologies using parent perceived and objective measures of the home yard and neighbourhood physical environment, and the association with preschool children’s device-measured physical activity and parent-reported outdoor play. Six profiles of preschool children’s home yard and neighbourhood physical environment were identified including ‘No blue space’, ‘Higher blue space’, ‘Higher public transport & destination mix’, ‘Connected residential’, ‘Established’ and ‘Semi-rural’. Preschooler home yard and neighbourhood physical environment profiles were associated with total daily physical activity, in particular light intensity physical activities and games, but not energetic play or outdoor play. Specifically, children in the ‘Higher blue space’ and ‘Semi-rural’ profiles accumulated between 13 and 26 more minutes of total daily physical activity compared with children in the ‘Higher public transport & destination mix’ neighbourhood profile. These findings were largely driven by the contribution light intensity activities and games (e.g., slow walking or ‘pottering about’, standing, standing arts and crafts) make to daily minutes of total physical activity (almost 90%).

Differences in the characteristics of these profiles may explain the observed difference in preschooler total daily minutes of physical activity. The ‘Higher blue space’ and ‘Semi-rural’ profiles had fewer public transport stops and fewer types of service and food outlet destinations compared with the ‘Higher public transport & destination mix’ profile. In addition, the ‘Higher blue space’ profile had greater access to blue spaces and the ‘Semi-rural’ profile had higher levels of neighbourhood vegetation and home yard space, compared with the ‘Higher public transport & destination mix’ profile. Previous research has also shown neighbourhoods with access to blue space are positively associated with light intensity activities and games in preschoolers (George et al., 2024). In addition, a review of the physical environment determinants of physical activity behaviour has shown home yard space is positively associated with young children’s physical activity (Carlin et al., 2017). Our findings lend further support to the social-ecological framework that home yard space and nature-based neighbourhood attributes such as vegetation and water are salient influences on preschooler s’ physical activity behaviour, particularly engagement in light intensity activities and games.

These findings in general support the handful of studies of the relationships between neighbourhood typologies and physical activity in school-aged children (DeWeese et al., 2018; Kurka et al., 2015; Timperio et al., 2017). For example, among school-aged children living in Victoria, Australia, neighbourhood typologies characterised by few playgrounds and sport venues, low intersection density and mixed land use were associated with less moderate-to-vigorous physical activity, compared to other neighbourhood typologies (Timperio et al., 2017). Furthermore, in a study combining GIS, Global Positioning Systems (GPS) and accelerometry data from 168 preschoolers, parks, playgrounds, childcare centres and non-home residences in the neighbourhood were found to have higher levels of daily moderate-to-vigorous physical activity (i.e., they were moderate-to-vigorous physical activity ‘hot spots’; Bai et al., 2023). While it is apparent neighbourhood typologies contain built environment attributes (i.e., parks and playgrounds) known to be associated with children’s physical activity when examined individually, additional studies examining the influence of data driven neighbourhood typologies on preschoolers’ movement behaviours are warranted to confirm these findings, especially in cities with different geographies. Future research should also investigate longitudinal relationships between neighbourhood typologies and movement behaviours across childhood.

Despite efforts to use comparable methodologies in studies of neighbourhood typology and physical activity behaviour, comparison of findings based on the different neighbourhood typologies identified may not be possible. This is due to neighbourhood typologies being specific to the geographical location of the study with limited generalisability to other cities. For instance, it is possible to compare the current study’s findings with other studies conducted in low density cities such as those found in the US, but it would be problematic to compare with high density (and often much older) cities more common throughout Europe. Furthermore, the contribution of built environment attributes such as blue space access may not be relevant for neighbourhood typology studies not conducted in coastal areas.

The six neighbourhood typologies identified in the current study for the most part are specific to families of preschool children living in the Perth, Western Australian metropolitan urban landscape. Specifically, the ‘Higher blue space’ profile represents a typical Perth neighbourhood located along the sprawling Perth coastline or Swan River. In contrast the ‘No blue space’ profile represents a typical Perth neighbourhood without access to blue space (i.e., the beach or Swan River). The ‘Higher public transport & destination mix’ profile was characterised by better land use mix and public transport access and for the most part are located in ‘activity centres’. Activity centres in Perth were designed to have compact urban form and mixed land uses (employment, housing, goods and services) and support active transport (WA Department of Planning Lands and Heritage, 2023). The ‘Established’ profile had the highest residential density with larger yards, better public transport and more land use mix, with most situated in the older and more established (relative to other suburbs) areas of Perth. The ‘Connected residential’ profile included neighbourhoods located in the newer growth areas of Perth characterised by new homes built lot-line to lot-line on smaller blocks, with low traffic and highly connected street networks but little access to shops and services. These areas provide more affordable housing for young families. The last profile, ‘Semi-rural’ included neighbourhoods located on the urban fringe and were characterised by larger blocks with more vegetation and yard space and thus lower residential density. They also had less access to public transport and lower street connectivity and land use mix. While the findings show that neighbourhood profile characteristics such as home yard space and access to green and blue space are associated with preschooler physical activity, it is important to consider the urban planning implications of such findings. It is not feasible particularly for low density cities with high levels of urban sprawl to continue to provide high levels of private (home yard) space. Thus, it is essential with increasing densification that additional safe and attractive public open space is made available close to homes, especially in areas with higher proportions of households with children.

Overall, it is not clear why device-measured energetic play and parent-reported outdoor play were not associated with preschool children’s home yard and neighbourhood physical environment profile. The parent-reported outdoor play checklist measures time spent playing outdoors on a typical day in the yard or street around the house; and at a park, playground, or outdoor recreation area. It is possible this measure is limited because it is parent-reported and/or too specific to the location of outdoor play (i.e., home and park/playground) and thus did not capture outdoor play undertaken in other locations such as blue spaces or that outdoor play is occurring in places outside the neighbourhood. Our previous research using GPS to identify the spatial location of device-based moderate-to-vigorous physical activity in preschoolers showed a key location was other households outside of the neighbourhood (i.e., the homes of relatives and friends; Bai et al., 2023). Further research matching device-based movement behaviour data with GPS data to specifically match behaviours to their environments will help to confirm where young children are more physically active and the profile of the neighbourhood built environments they reside in.

Finally, despite parent perceptions of the neighbourhood built environment being included in each of the six neighbourhood profiles identified, they did not appear to be a defining characteristics of one profile over another. This may in part be due to the limited number of response options used in the Y-NEWS as well as the relatively small amount of variation in the responses provided by parents. Furthermore, the Y-NEWS may be less sensitive at identifying differences in parents’ perceptions of the built environment in the context of preschool aged children or that parent perceptions somewhat moderate the relationship between objective measures of the built environment and young children’s physical activity (Nathan et al., 2023).

Strengths and Limitations

Strengths of this study were its relatively large sample size, focus on an understudied population group (preschool age children), latent profile analysis approach to determine the different combinations of neighbourhood built environment attributes in a geographical area (and the relationship with young children’s physical activity behaviour) and the inclusion of 17 objectively measured (GIS) and parent perceptions of the home yard and neighbourhood physical environment created at the individual-child-level. It was however limited in terms of the generalisability of findings which may only be relevant to other low density coastal cities. Further research in other countries is needed to confirm similarities and identify differences in young children’s profiles of their home yard and neighbourhood physical environment and the relationship with preschool children’s movement behaviours. In addition, other objectively measured and or parent perceptions of the home yard and neighbourhood physical environment not measured may be important to include in future studies (e.g., playground access, quality and use of different neighbourhood physical environment attributes). Future research should also investigate the longitudinal relationship between neighbourhood built environment profiles and movement behaviours across childhood, the multi-level nature of these relationships as well as the role that preschool children’s own level of independence and decision making has on the interaction between young children’s physical activity behaviour and the neighbourhood built environment.

Conclusion

Six profiles of preschool children’s home yard and neighbourhood physical environment were identified with children living in ‘Higher blue space’ and ‘Semi-rural’ profiles accumulating significantly more minutes of total daily physical activity compared with children characterised by the ‘Higher public transport & destination mix’ profile. The findings highlight how urban form can vary across neighbourhoods and by study population and geographic location. Findings from such studies can provide more localised and relevant evidence to support urban planning decisions to support young children (and their families) to establish and maintain positive health behaviours such as physical activity. Overall, neighbourhood profiling approaches can provide a ‘real world’ and geographically specific and locally relevant approach to examining the influence of the built environment on child movement behaviour.

Supplemental Material

sj-docx-1-eab-10.1177_00139165251375176 – Supplemental material for Home Yard and Neighbourhood Physical Environment Latent Class Profiles and Preschooler’s Physical Activity Behaviour: Findings from the PLAYCE Study

Supplemental material, sj-docx-1-eab-10.1177_00139165251375176 for Home Yard and Neighbourhood Physical Environment Latent Class Profiles and Preschooler’s Physical Activity Behaviour: Findings from the PLAYCE Study by Hayley Christian, Emma Adams, Stewart Trost, Jasper Schipperijn, Kevin Murray, Bryan Boruff, Andrea Nathan, Gareth Stratton and Trina Robinson in Environment and Behavior

Footnotes

Acknowledgements

The BEACHES Project is a joint initiative between The Kids Research Institute Australia, The University of Western Australia, and Swansea University, with collaborators from Curtin University, Monash University, The Queensland University of Technology, University of Southern Denmark, WA Department of Local Government, Sports and Cultural Industries, WA Department of Health, WA Department of Transport, WA Local Government Association, Australian Childcare Alliance, Nature Play Australia, Heart Foundation, The PLAY Spaces and Environments for Children’s Physical Activity (PLAYCE) cohort study, Cancer Council WA, Goodstart Early Learning, and Hames Sharley. Bridget Beesley, Dr John Duncan and Dr Claire Boulange are gratefully acknowledged for their work on the geospatial measures and map.

The PLAYCE cohort study research team would like to thank our project team members who helped to collect and prepare the data, participating ECEC services investigators (Prof Michael Rosenberg, A/Prof Leanne Lester, Prof Gina Trapp, Dr Clover Maitland and Dr Ashleigh Thornton) and partners: the National Heart Foundation, Western Australian (WA) Department of Health, WA Department of Local Government, Sport and Cultural Industries, WA Department of Transport, WA Local Government Association, Cancer Council WA, Nature Play WA, Australian Childcare Alliance (WA), UWA Childcare, Goodstart Early Learning, Maragon Early Learning, Sonas Early Learning & Care, Mercy Care, Great Beginnings Early Education, Jellybeans Child Care & Kindy, Buggles Early Learning and Kindy and ArborCarbon who provided in-kind support. We thank the families who participated in the PLAYCE cohort study for their time and commitment.

ORCID iDs

Hayley Christian

Stewart Trost

Jasper Schipperijn

Trina Robinson

Ethical Considerations

Ethics approval was granted by The University of Western Australia Human Research Ethics Committee (RA/4/1/ 7417 and 2020/ET000353). Parents provided written informed consent for themselves and their child to participate.

Author Contributions

HC: conceptualisation, methodology, investigation, writing – original draft, writing – review and editing, supervision, project administration, funding acquisition. EA: methodology, formal analysis, writing – original draft, writing – review and editing. SGT: conceptualisation, methodology, writing – review and editing, funding acquisition. JS: conceptualisation, writing – review and editing, funding acquisition. KM: conceptualisation, methodology, writing – review and editing, funding acquisition. BB: conceptualisation, methodology, writing – review and editing, funding acquisition. AN: conceptualisation, methodology, investigation, writing – review and editing. GS: conceptualisation, writing – review and editing, funding acquisition. TR: data curation, project administration, writing – review and editing. All authors read and approved the final manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is part of the BEACHES Project which is a joint initiative between The Kids Research Institute Australia, The University of Western Australia and Swansea University. The BEACHES Project is funded by the UKRI-NHMRC Built Environment Prevention Research Scheme (GNT1192764 and MR/T039329/1) and partially supported by the Australian Government through the Australian Research Council’s Centre of Excellence for Children and Families over the Life Course (CE200100025). The PLAYCE cohort study was funded by the Western Australian Health Promotion Foundation (Healthway; 24219 and 32018) and part-funded by UKRI-NHMRC Built Environment Prevention Research Scheme (GNT1192764 and MR/T039329/1). This work is supported by the Western Australian Future Health Research and Innovation Fund (Grant ID WACSOSP2023-2024/4). Hayley Christian is supported by an Australian National Heart Foundation Future Leader Fellowship (102549) and The Kids Research Institute Australia – Ascend Senior Research Fellowship. Emma Adams is supported by an Australian Government Research Training Program Fees Offset and Australian Government Research Training Program Stipend, a UWA and Graduate Women (WA) Research Scholarship, and a Stan and Jean Perron Top Up Scholarship.

Declaration of Conflicting Interests

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

Data Availability Statement

The datasets used during the current study are available through application to the PLAYCE study.

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Hayley Christian is a Professor at The Kids Research Institute Australia and The University of Western Australia. She leads the Child Physical Activity, Health and Development team. Her research focuses on improving children’s physical activity, health and development through interventions focused on the child, the family, and social and built environment.

Emma Adams is a PhD Candidate in The University of Western Australia’s School of Population and Global Health and a Research Officer in the Child Physical Activity, Health and Development team at The Kids Research Institute Australia. Her research focuses on understanding the patterns and predictors of physical activity, with an emphasis on using longitudinal and device-based data.

Stewart Trost is a Professor in the Child Health Research Centre at The University of Queensland. His research interests include device-based assessment of physical activity and sedentary behaviour, early life prevention of childhood obesity and chronic disease, therapeutic exercise programs for children with chronic health conditions, and the relationships between physical activity and other health behaviours.

Jasper Schipperijn is a Professor in the Department of Sports Science and Clinical Biomechanics at the University of Southern Denmark. His research interests include understanding the relationship between the built environment and human behaviour, and in particular evaluating the effect of changes to the environment on behaviour.

Kevin Murray is an Associate Professor in the School of Population and Global Health at The University of Western Australia. His research interests include the application of statistics to medical and health sciences research.

Bryan Boruff is an Associate Professor in the School of Agriculture and Environment at The University of Western Australia. His expertise lies in the application of Geographic Information Systems (GIS) and Remote Sensing technologies to the study of environmental hazards.

Andrea Nathan is a Senior Research Officer in the Child Physical Activity, Health and Development team at The Kids Research Institute Australia. Her interests are in developing and evaluating multi-level, intergenerational interventions to improve health and wellbeing in children and older adults.

Gareth Stratton is a Professor in Paediatric Exercise Science in the Department of Sport and Exercise Science at the School of Engineering and Applied Science, Swansea University and an adjunct Professor at The University of Western Australia. His academic interests include children’s maturation and physical activity and physical activity, fitness and health.

Trina Robinson is a Research Officer in the Child Physical Activity, Health and Development team at The Kids Research Institute Australia. Her research interests focus on the influence of the urban environment on children’s physical and mental health and development.

References

Active Healthy Kids Australia. (2022). Reboot! Reimagining physically active lives: 2022 Australian report card on physical activity for children and young people. https://www.activehealthykids.org/australia/

Ahmadi

M. N.

Nathan

Sutherland

Wolfenden

Trost

S. G.

(2020). Non-wear or sleep? Evaluation of five non-wear detection algorithms for raw accelerometer data. Journal of Sports Sciences, 38(4), 399–404. https://doi.org/10.1080/02640414.2019.1703301

Ahmadi

M. N.

Pavey

T. G.

Trost

S. G.

(2020). Machine learning models for classifying physical activity in free-living preschool children. Sensors, 20(16), 4364. https://doi.org/10.3390/s20164364

Ahmadi

M. N.

Trost

S. G.

(2022). Device-based measurement of physical activity in pre-schoolers: Comparison of machine learning and cut point methods. PLoS One, 17(4), e0266970. https://doi.org/10.1371/journal.pone.0266970

Armstrong

G. P.

Maitland

Lester

Trost

S. G.

Trapp

Boruff

Al Marzooqi

M. K.

Christian

H. E.

(2019). Associations between the home yard and preschoolers’ outdoor play and physical activity. Public Health Research and Practice, 29(1). https://doi.org/10.17061/phrp2911907

Australian Bureau of Statistics. (2024). Regional population, 2022–23. Retrieved January 24, 2024, from https://www.abs.gov.au/statistics/people/population/regional-population/latest-release

Australian Government Department of Health. (2019). Australian 24-hour movement guidelines for the early years (birth to 5 years): An integration of physical activity, sedentary behaviour and sleep. https://www.health.gov.au/topics/physical-activity-and-exercise/physical-activity-and-exercise-guidelines-for-all-australians/for-infants-toddlers-and-preschoolers-birth-to-5-years

Bai

Schipperijn

Rosenberg

Christian

(2023). Neighborhood places for preschool children’s physical activity: A mixed-methods study using global positioning system, geographic information systems, and accelerometry data. Journal of Physical Activity & Health, 20(8), 781–791. https://doi.org/10.1123/jpah.2022-0462

Bakk

Tekle

F. B.

Vermunt

J. K.

(2013). Estimating the association between latent class membership and external variables using bias-adjusted three-step approaches. Sociological Methodology, 43(1), 272–311. https://doi.org/10.1177/0081175012470644

10.

Bakk

Vermunt

J. K.

(2016). Robustness of stepwise latent class modeling with continuous distal outcomes. Structural Equation Modeling, 23(1), 20–31. https://doi.org/10.1080/10705511.2014.955104

11.

Burdette

H. L.

Whitaker

R. C.

Daniels

S. R.

(2004). Parental report of outdoor playtime as a measure of physical activity in preschool-aged children. Archives of Pediatrics and Adolescent Medicine, 158(4), 353–357. https://doi.org/10.1001/archpedi.158.4.353

12.

Caccetta

Collings

Devereux

Hingee

McFarlane

Traylen

Zhou

(2012). Urban Monitor: Enabling effective monitoring and management of urban and coastal environments using digital aerial photography (EP129592). https://publications.csiro.au/publications/publication/PIcsiro:EP129592

13.

Carlin

Perchoux

Puggina

Aleksovska

Buck

Burns

Cardon

Chantal

Ciarapica

Condello

Coppinger

Cortis

D’Haese

De Craemer

Di Blasio

Hansen

Iacoviello

Issartel

Izzicupo

, . . . Boccia

(2017). A life course examination of the physical environmental determinants of physical activity behaviour: A “Determinants of Diet and Physical Activity” (DEDIPAC) umbrella systematic literature review. PLoS One, 12(8), e0182083-e0182083. https://doi.org/10.1371/journal.pone.0182083

14.

Carson

Lee

E.-Y.

Hewitt

Jennings

Hunter

Kuzik

Stearns

J. A.

Unrau

S. P.

Poitras

V. J.

Gray

Adamo

K. B.

Janssen

Okely

A. D.

Spence

J. C.

Timmons

B. W.

Sampson

Tremblay

M. S.

(2017). Systematic review of the relationships between physical activity and health indicators in the early years (0–4 years). BMC Public Health, 17(5), 854. https://doi.org/10.1186/s12889-017-4860-0

15.

Christian

H. E.

Adams

E. A.

Moore

H. L.

Nathan

Murray

Schipperijn

Trost

S. G.

(2024). Developmental trends in young children’s device-measured physical activity and sednetary behaviour. International Journal of Behavioural Nutrition and Physical Activity, 21, 97. https://doi.org/10.1186/s12966-024-01645-z

16.

Christian

Ball

S. J.

Zubrick

S. R.

Brinkman

Turrell

Boruff

Foster

(2017). Relationship between the neighbourhood built environment and early child development. Health & Place, 48, 90–101. https://doi.org/10.1016/j.healthplace.2017.08.010

17.

Christian

Knuiman

Divitini

Foster

Hooper

Boruff

Bull

Giles-Corti

(2017). A longitudinal analysis of the influence of the neighborhood environment on recreational walking within the neighborhood: Results from RESIDE. Environmental Health Perspectives, 125(7), e077009. https://doi.org/10.1289/ehp823

18.

Christian

Maitland

Enkel

Trapp

Trost

S. G.

Schipperijn

Boruff

Lester

Rosenberg

Zubrick

S. R.

(2016). Influence of the day care, home and neighbourhood environment on young children’s physical activity and health: Protocol for the PLAYCE observational study. BMJ Open, 6(12), e014058. https://doi.org/10.1136/bmjopen-2016-014058

19.

Christian

Nathan

Trost

S. G.

Schipperijn

Boruff

Adams

E. K.

George

Moore

H. L.

Henry

(2025). Profile of the PLAY Spaces & Environments for Children’s Physical Activity, Sedentary Behaviour and Sleep (PLAYCE) cohort study, Western Australia. Journal of Activity, Sedentary and Sleep Behaviors, 4, 7. https://doi.org/10.1186/s44167-025-00078-8

20.

Christian

Zubrick

S. R.

Foster

Giles-Corti

Bull

Wood

Knuiman

Brinkman

Houghton

Boruff

(2015). The influence of the neighborhood physical environment on early child health and development: A review and call for research. Health & Place, 33, 25–36. https://doi.org/10.1016/j.healthplace.2015.01.005

21.

De Craemer

De Decker

De Bourdeaudhuij

Vereecken

Deforche

Manios

Cardon

. (2012). Correlates of energy balance-related behaviours in preschool children: A systematic review. Obesity Reviews, 13(s1), 13–28. https://doi.org/10.1111/j.1467-789X.2011.00941.x

22.

DeWeese

R. S.

Ohri-Vachaspati

Adams

M. A.

Kurka

Han

S. Y.

Todd

Yedidia

M. J.

(2018). Patterns of food and physical activity environments related to children’s food and activity behaviors: A latent class analysis. Health & Place, 49, 19–29. https://doi.org/10.1016/j.healthplace.2017.11.002

23.

Duncan

J. M. A.

Boruff

(2023). Monitoring spatial patterns of urban vegetation: A comparison of contemporary high-resolution datasets. Landscape and Urban Planning, 233, 104671. https://doi.org/10.1016/j.landurbplan.2022.104671

24.

Duncan

J. M. A.

Boruff

Saunders

Sun

Hurley

Amati

(2019). Turning down the heat: An enhanced understanding of the relationship between urban vegetation and surface temperature at the city scale. The Science of the Total Environment, 656, 118–128. https://doi.org/10.1016/j.scitotenv.2018.11.223

25.

Gemmell

Ramsden

Brussoni

Brauer

(2022). Influence of neighborhood built environments on the outdoor free play of young children: A systematic, mixed-studies review and thematic synthesis. Journal of Urban Health, 100(1), 118–150. https://doi.org/10.1007/s11524-022-00696-6

26.

George

Murray

Trost

S. G.

Boruff

Christian

(2024). Associations between natural blue and green space and preschool children’s movement behaviours. People and Nature, 6(4), 10682. https://doi.org/10.1002/pan3.10682

27.

Giles-Corti

Bull

Knuiman

McCormack

Van Niel

Timperio

Christian

Foster

Divitini

Middleton

Boruff

(2013). The influence of urban design on neighbourhood walking following residential relocation: Longitudinal results from the RESIDE study. Social Science & Medicine, 77, 20–30. https://doi.org/10.1016/j.socscimed.2012.10.016

28.

Giles-Corti

Macaulay

Middleton

Boruff

Bull

Butterworth

Badland

Mavoa

Roberts

Christian

(2014). Developing a research and practice tool to measure walkability: A demonstration project. Health Promotion Journal of Australia, 25(3), 160–166. https://doi.org/10.1071/HE14050

29.

Hinkley

Crawford

Salmon

Okely

A. D.

Hesketh

(2008). Preschool children and physical activity: A review of correlates. American Journal of Preventive Medicine, 34(5), 435–441. https://doi.org/10.1016/j.amepre.2008.02.001

30.

Kurka

J. M.

Adams

M. A.

Todd

Colburn

Sallis

J. F.

Cain

K. L.

Glanz

Frank

L. D.

Saelens

B. E.

(2015). Patterns of neighborhood environment attributes in relation to children’s physical activity. Health & Place, 34, 164–170. https://doi.org/10.1016/j.healthplace.2015.05.006

31.

Lambert

Vlaar

Herrington

Brussoni

(2019). What is the relationship between the neighbourhood built environment and time spent in outdoor play? A systematic review. International Journal of Environmental Research and Public Health, 16(20), 3840. https://doi.org/10.3390/ijerph16203840

32.

Landgate. (2017). Spatial Cadastral Database (SCDB). https://www.landgate.wa.gov.au/location-data-and-services/surveying/cadastral/

33.

Lee

E.-Y.

Bains

Hunter

Ament

Brazo-Sayavera

Carson

Hakimi

Huang

W. Y.

Janssen

Lee

Lim

Silva

Tremblay

(2021). Systematic review of the correlates of outdoor play and time among children aged 3–12 years. International Journal of Behavioral Nutrition and Physical Activity, 18, 1–17. https://doi.org/10.1186/s12966-021-01097-9

34.

Nathan

Schipperijn

Robinson

George

Boruff

Trost

S. G.

Christian

(2023). The moderating role of parent perceptions in relationships between objectively measured neighbourhood environment attributes and pre-schooler’s physical activity: Findings from the PLAYCE study. Health & Place, 81, 103030. https://doi.org/10.1016/j.healthplace.2023.103030

35.

Puhl

Greaves

Hoyt

Baranowski

(1990). Children’s activity rating scale (CARS): Description and calibration. Research Quarterly in Exercise and Sport, 61(1):26–36.

36.

Robinson

Dhamrait

Murray

Boruff

Duncan

Schipperijn

Christian

(2024). Association between preschooler outdoor play and home yard vegetation as measured by high resolution imagery: Findings from the PLAYCE study. Health & Place, 85, 103178. https://doi.org/10.1016/j.healthplace.2024.103178

37.

Robinson

Nathan

Murray

Christian

(2022). Perceptions of the neighbourhood built environment are associated with the social and emotional development of young children. International Journal of Environmental Research and Public Health, 19(11), 6476. https://www.mdpi.com/1660-4601/19/11/6476

38.

Rosenberg

Ding

Sallis

J. F.

Kerr

Norman

G. J.

Durant

Harris

S. K.

Saelens

B. E.

(2009). Neighborhood environment walkability scale for youth (NEWS-Y): Reliability and relationship with physical activity. Preventive Medicine, 49(2), 213–218. https://doi.org/10.1016/j.ypmed.2009.07.011

39.

Sinha

Calfee

C. S.

Delucchi

K. L.

(2021). Practitioner’s guide to latent class analysis: Methodological considerations and common pitfalls. Critical Care Medicine, 49(1), E63–E79. https://doi.org/10.1097/CCM.0000000000004710

40.

Statistical Innovations Inc. (2016). LatentGOLD 5.1 [Computer software]. Statistical Innovations Inc.

41.

Stokols

(1996). Translating social ecological theory into guidelines for community health promotion. American Journal of Health Promotion, 10(4), 282–298. https://doi.org/10.4278/0890-1171-10.4.282

42.

Timperio

Crawford

Ball

Salmon

(2017). Typologies of neighbourhood environments and children’s physical activity, sedentary time and television viewing. Health & Place, 43, 121–127. https://doi.org/10.1016/j.healthplace.2016.10.004

43.

Tremblay

M. S.

Chaput

J.-P.

Adamo

K. B.

Aubert

Barnes

J. D.

Choquette

Duggan

Faulkner

Goldfield

G. S.

Gray

C. E.

Gruber

Janson

Janssen

Jaramillo Garcia

Kuzik

LeBlanc

MacLean

Okely

A. D.

, . . . Carson

(2017). Canadian 24-hour movement guidelines for the early years (0–4 years): An integration of physical activity, sedentary behaviour, and sleep. BMC Public Health, 17(Suppl. 5), 874. https://doi.org/10.1186/s12889-017-4859-6

44.

Tremblay

M. S.

Gray

Babcock

Barnes

Bradstreet

C. C.

Carr

Chabot

Choquette

Chorney

Collyer

Herrington

Janson

Janssen

Larouche

Pickett

Power

Sandseter

E. B.

Simon

Brussoni

(2015). Position statement on active outdoor play. International Journal of Environmental Research and Public Health, 12(6), 6475–6505. https://doi.org/10.3390/ijerph120606475

45.

A. W.

Mâsse

L. C.

Lear

S. A.

Gotay

C. C.

Richardson

C. G.

(2016). Exploring the mediating roles of physical activity and television time on the relationship between the neighbourhood environment and childhood obesity. Canadian Journal of Public Health, 107(2), e168–e175. https://doi.org/10.17269/cjph.107.5268

46.

Vanwolleghem

Schipperijn

Gheysen

Cardon

De Bourdeaudhuij

Van Dyck

(2016). Children’s GPS-determined versus self-reported transport in leisure time and associations with parental perceptions of the neighborhood environment. International Journal of Health Geographics, 15(1), 1–16. https://doi.org/10.1186/s12942-016-0045-9

47.

Veldman

S. L. C.

Paw

Altenburg

T. M.

(2021). Physical activity and prospective associations with indicators of health and development in children aged 5 years: A systematic review. International Journal of Behavioral Nutrition and Physical Activity, 18(1), 1–6. https://doi.org/10.1186/s12966-020-01072-w

48.

Villanueva

Badland

Kvalsvig

O’Connor

Christian

Woolcock

G. B. A.

Giles-Corti

Goldfeld

(2016). Can the neighborhood built environment make a difference in children’s development? Building the research agenda to create evidence for place-based children’s policy. Academic Pediatrics, 16(1), 10–19. https://doi.org/10.1016/j.acap.2015.09.006

49.

Villanueva

Pereira

Knuiman

Bull

Wood

Christian

Foster

Boruff

B. J.

Beesley

Hickey

Joyce

Nathan

Saarloos

Giles-Corti

(2013). The impact of the built environment on health across the life course: Design of a cross-sectional data linkage study. BMJ Open, 3(1), e002482. https://doi.org/10.1136/bmjopen-2012-002482

50.

WA Department of Planning Lands and Heritage. (2022). Land use and employment surveys. https://www.wa.gov.au/government/document-collections/land-use-and-employment-surveys

51.

WA Department of Planning Lands and Heritage. (2023). State planning policy 4.2: Activity centres. https://www.wa.gov.au/system/files/2023-07/spp_4.2-activity-centres.pdf

52.

Weden

M. M.

Bird

C. E.

Escarce

J. J.

Lurie

(2011). Neighborhood archetypes for population health research: Is there no place like home? Health & Place, 17(1), 289–299. https://doi.org/10.1016/j.healthplace.2010.11.002

53.

World Health Organization. (2019). Guidelines on physical activity, sedentary behaviour and sleep for children under 5 years of age. https://www.who.int/publications/i/item/9789241550536

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB