Active Where? Surveys

Abstract

The Active Where? Surveys represent a comprehensive psychometric and behavioral-ecological assessment suite engineered to quantify environmental, parental, and individual determinants of physical activity and dietary behaviors among youth. Developed through a multi-site collaborative partnership across San Diego State University, Cincinnati Children’s Hospital Medical Center, and Boston Children’s Hospital, the instrument was conceived to bridge the empirical gap between adult-centric built environment surveys (such as the Neighborhood Environment Walkability Scale [NEWS]) and the unique developmental affordances of children and adolescents. The instrument measures multiple socio-ecological tiers: residential density, land use mix-diversity (walking proximity to neighborhood destinations), walking and biking facilities, pedestrian infrastructure, aesthetic amenities, traffic hazards, safety from crime, park accessibility and quality, and home physical activity equipment availability. Administered across parent-child dyads, parent-adolescent dyads, and independent youth respondents, the core battery comprises 48 standardized items evaluating neighborhood, destination, equipment, and park affordances, supplemented by modules capturing school environments and sedentary behaviors.

Psychometric evaluation of the Active Where? Surveys demonstrates robust measurement properties across geographically and demographically heterogeneous populations. Multi-site validation studies involving youth aged 5 to 18 years yielded evidence for construct, convergent, and discriminant validity. Item responses vary systematically by functional domain: neighborhood environmental infrastructure, aesthetics, and safety attributes are assessed via a 4-point agreement scale (1 = Strongly Disagree to 4 = Strongly Agree); destination proximity utilizes an ordinal temporal walking-distance metric ranging from 1 (≤5 minutes) to 5 (≥31 minutes), with an explicit non-evaluative option (6 = Don’t know); and the domestic material environment is gauged via binary availability or frequency indices. Test-retest reliability across a two-week interval demonstrates intra-class correlation coefficients (ICCs) and Spearman-Brown coefficients spanning moderate to substantial thresholds (ICCs = 0.56 to 0.88 across subscales), while intra-dyadic concordance between parent and adolescent proxies highlights distinct developmental viewpoints. The instrument serves as a critical methodology for public health researchers, behavioral epidemiologists, and urban planners seeking to identify modifiable environmental correlates of pediatric obesity and physical inactivity.

Keywords

Active Where? Surveys, built environment, physical activity, pediatric obesity, ecological systems theory, walkability, environmental affordances, youth psychometrics, spatial health, neighborhood safety, active transport, parent-child dyad.

Authors

The Active Where? Surveys were conceptualized, operationalized, and psychometrically validated by a multi-institutional consortium of clinical and behavioral medicine researchers, pediatricians, and environmental health epidemiologists:

• James F. Sallis, Ph.D. — Distinguished Professor Emeritus of Family Medicine and Public Health, University of California San Diego, and Research Professor, Australian Catholic University, Melbourne, Australia. Co-Principal Investigator.
• Brian E. Saelens, Ph.D. — Professor of Pediatrics and Psychiatry and Behavioral Sciences, University of Washington, and Seattle Children’s Research Institute, Seattle, Washington. Principal Investigator.
• Terry L. Conway, Ph.D. — Senior Research Health Scientist, Department of Family Medicine and Public Health, University of California San Diego, San Diego, California.
• Kelli L. Cain, M.A. — Project Director and Clinical Research Specialist, Active Living Research, University of California San Diego, San Diego, California.
• Lori D. Crosby, Psy.D. — Professor of Pediatrics, Division of Behavioral Medicine and Clinical Psychology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio.
• Elora D. Frank, M.S. — Division of Behavioral Medicine and Clinical Psychology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio.
• David S. Ludwig, M.D., Ph.D. — Professor of Pediatrics, Harvard Medical School, and Endowed Chair in Pediatric Endocrinology, Boston Children’s Hospital, Boston, Massachusetts.

Institutional collaborative nodes spanned the Active Living Research Program (San Diego, CA), Cincinnati Children’s Hospital Medical Center (Cincinnati, OH), and Boston Children’s Hospital / Harvard Medical School (Boston, MA).

Purpose

The primary rationale underpinning the development of the Active Where? Surveys was the systematic absence of psychometrically validated, pediatric-specific self-report instruments capable of capturing how environmental affordances moderate or mediate physical activity and dietary habits in children and adolescents. Historically, physical activity research was dominated by individual-level, intra-psychic paradigms focusing almost exclusively on internal psychological constructs, such as self-efficacy, stage of change, and personal motivation. While informative, these cognitive-behavioral constructs routinely failed to explain significant variance in pediatric energy expenditure, active commuting, or habitual sedentary behaviors. Where environmental instruments did exist, they were almost entirely modeled after adult spatial mobility patterns, presuming independent transit, autonomous driver capability, adult navigation range, and mature cognitive appraisal of urban hazards.

Consequently, the Active Where? Study was initiated to capture the unique spatial and social ecological boundaries of youth living in diverse urban and suburban configurations. Children do not experience urban design through the same lens as adults; their active mobility is constrained by parental permissions (parental license), safety perceptions, pedestrian scale, proximity of play-spaces, and the availability of sports infrastructure within the domestic micro-environment. To formulate appropriate items, the investigators instituted an innovative formative phase consisting of ‘in situ’ individual interviews conducted directly within neighborhood streets, residential blocks, and park environments with children and parents. This situated methodology allowed investigators to observe, record, and unpack real-time interactions with infrastructure — such as sidewalk cracks, speeding traffic, blind intersections, stray animals, park cleanliness, and street-level crime — which subsequently guided item content formulation.

In clinical and translational settings, the Active Where? Surveys provide pediatric endocrinologists, child psychologists, and community health specialists with an empirical diagnostic tool to contextualize lifestyle interventions. Instead of prescribing physical activity regimens in a theoretical vacuum, clinicians can systematically profile a child’s environmental risk or protection, diagnosing whether sedentary habits stem from structural environmental barriers (e.g., lack of sidewalks, unsafe traffic, absence of play equipment) or social-cognitive deficits. In academic research, the surveys furnish epidemiologists, geographers, and urban sociologists with high-resolution subjective data to cross-validate Geographic Information Systems (GIS) spatial layers, resolving discrepancies between objective spatial walkability metrics and the phenomenological experiences of youth and their caregivers.

Psychological Construct

The overarching psychological and behavioral construct quantified by the Active Where? Surveys is the Perceived Micro- and Macro-Physical Environment for Youth Physical Activity. Drawing upon environmental psychology and affordance theory, the instrument assesses the degree to which an individual’s spatial environment provides actionable opportunities, behavioral prompts, and safety margins for active travel, unorganized play, sports participation, and health-promoting behaviors. Rather than treating the built environment as a monolithic entity, the scale decomposes the environment into seven primary operational dimensions:

1. Residential Density and Land Use Mix-Diversity

This construct captures the spatial clustering, variety, and relative convenience of walkable neighborhood destinations. Grounded in urban design theories of mixed land use, this dimension assesses whether non-residential destinations (such as grocery stores, transit stops, schools, libraries, and parks) are situated within a developmentally plausible walking radius (e.g., 5-to-20-minute walk). A higher land-use mix-diversity presents frequent destination-driven prompts that elicit active travel (walking, scootering, bicycling) over passive vehicular transport.

2. Pedestrian and Cycling Infrastructure (Walking/Biking Facilities)

This domain evaluates the physical presence, continuity, and physical condition of structural paths dedicated to non-motorized transit. It measures the prevalence of paved sidewalks, their state of maintenance (lack of cracks or structural unevenness), crosswalks, traffic-control pedestrian signals, bike paths, and physical landscape buffers (e.g., dirt or grass strips separating the sidewalk from vehicular lanes). Psychologically, continuous infrastructure functions as an external affordance that decreases cognitive appraisal of danger, increases perceived control, and reassures both children and their supervisory parents.

3. Aesthetic Amenities

Aesthetic quality measures the sensory, visual, and experiential pleasure derived from moving through one’s residential context. Indicators include the presence of shade-providing trees, interesting architectural or visual landmarks, and the systemic absence of physical incivilities such as litter, discarded trash, broken glass, abandoned buildings, and graffiti. Environmental aesthetics operate as positive emotional reinforcers, generating pleasant affective states that sustain leisure-time physical exploration and pedestrian engagement.

4. Safety from Traffic and Vehicular Hazards

Traffic safety reflects subjective appraisals of vehicular volume, travel velocity, and roadway driver behavior. Items assess whether local traffic speed is perceived as low (≤30 mph), whether motorists respect posted limits, whether street illumination is adequate at night, and whether overall traffic volume impedes safe street crossing. Perceived traffic danger operates as a profound psychological barrier that induces parental anxiety, triggering restrictive parenting behaviors that suppress independent youth mobility.

5. Safety from Crime and Social Incivilities

This dimension assesses subjective vulnerability to personal harm, social threats, and predatory behavior within the neighborhood ecology. It operationalizes fears related to daytime and nighttime walking safety, high neighborhood crime rates, and biological or physical threats, such as unrestrained, aggressive dogs. High perceived crime fosters hypervigilance, psychological distress, and immediate home confinement, severely curtailing outdoor free-play and physical leisure.

6. Park and Recreation Quality and Accessibility

Parks serve as the primary macro-structural venues for moderate-to-vigorous physical activity (MVPA) among youth. This subscale evaluates park proximity, the physical state of equipment (swings, climbing frames, slides), availability of open grassy fields for unstructured play, formal sports courts, daytime security, and overall cleanliness. In accordance with Gibsonian affordance theory, a park is only functional if its components afford physical actions aligned with the child’s gross motor repertoire.

7. Domestic Physical Activity Equipment Availability

The domestic micro-environment serves as the primary immediate setting for younger youth. This construct indexes the material wealth of active play equipment located directly inside the home, yard, or immediate domestic perimeter (e.g., bicycles, jump ropes, basketball hoops, sports balls, roller skates, weight-lifting gear, trampolines, active video games, swimming pools). Domestic resource density provides non-structured behavioral prompts that facilitate spontaneous activity independently of neighborhood access.

Theoretical Framework

The Active Where? Surveys are firmly grounded in the principles of Social-Ecological Models of health behavior, most notably advanced by Urie Bronfenbrenner, James F. Sallis, and Neville Owen. The social-ecological framework posits that human behavior is not dictated solely by personal factors (attitudes, biological traits, beliefs), but is dynamic, multifaceted, and shaped by nested environmental tiers: the microsystem (home and immediate family), the mesosystem (interrelations between home, school, and neighborhood), the exosystem (neighborhood design, municipal zoning, public transit networks), and the macrosystem (cultural norms, socioeconomic stratification, municipal policies).

Complementing social ecology, the conceptual architecture relies extensively on Affordance Theory, originally formulated by ecological psychologist James J. Gibson. Gibson posited that environmental features are perceived not as abstract geometric configurations, but directly in terms of their functional possibilities for action: a paved, even sidewalk afford walking or roller-skating; a grassy clearing afford running and ball games; an untethered dog afford hazard and avoidance; and a steep staircase or cracked pathway afford injury risk. In youth, behavior is primarily guided by affordances that match developmental competencies. The Active Where? Surveys translate Gibsonian affordances into psychometric indicators by measuring whether physical settings present behavioral invitations (e.g., play equipment, bike paths) or behavioral inhibitors (e.g., litter, speeding cars).

Additionally, the instrument integrates tenets from Bandura’s Social Cognitive Theory, particularly the construct of reciprocal determinism, which maintains that environmental characteristics, personal cognitive factors, and behavioral patterns continuously exert bidirectional influences on one another. Within the Active Where? paradigm, an aesthetically pleasing, safe environment elevates an adolescent’s self-efficacy for active transport, while simultaneously diminishing parental risk appraisal. Parents operating under low perceived threat are significantly more inclined to grant “independent mobility licenses,” allowing children to walk to school, socialize in parks, and navigate their neighborhood unaccompanied.

Validity

The psychometric validity of the Active Where? Surveys has been documented across multiple validation studies, programmatic evaluations, and epidemiological cohorts across the United States and internationally. Construct, convergent, and discriminant validities were systematically evaluated across the tri-city consortium sites (San Diego, Cincinnati, Boston), capturing substantial geographic, socioeconomic, and racial diversity.

Construct and Structural Validity

Construct validity was initially established via formative qualitative investigations and cognitive testing. Investigators utilized ‘in situ’ walking interviews to confirm that survey indicators accurately mirrored the real-world experiences of children and their parents. Subsequent structural equation modeling and confirmatory factor analyses verified that the hypothesized environmental dimensions (safety, infrastructure, aesthetics, proximity) formed distinct, coherent factors that mapped cleanly onto behavioral outcomes without unacceptable cross-loadings.

Convergent and Criterion Validity

Convergent validity has been evaluated by contrasting youth and parent survey responses with objective Geographic Information System (GIS) metrics and objective physical activity measurements obtained via accelerometry (e.g., ActiGraph monitors) and Global Positioning System (GPS) tracking. Specific published findings include:

• Objective Built Environment Concordance: Significant positive correlations were documented between subjective Land Use Mix-Diversity scores and objective GIS-derived measures of destination density, residential density, and intersection density (Pearson’s r ranging from 0.38 to 0.62, p < 0.001).
• Accelerometry-Measured Physical Activity: Multi-level regression modeling demonstrated that children and adolescents residing in environments characterized by high walking/biking infrastructure, elevated park quality, and diverse destinations accumulated significantly higher minutes of daily moderate-to-vigorous physical activity (MVPA). Specifically, each standard deviation increase in composite neighborhood walkability was associated with an increase of approximately 6.5 to 9.2 minutes/day of MVPA (p < 0.01).
• Active Commuting Validation: Survey items indexing walking proximity to schools and transit stops exhibited strong criterion validity, demonstrating significant predictive utility for actual active travel to school (odds ratios ranging from 1.45 to 2.12 for destinations within a 15-minute walking threshold).
• Home Equipment and Sedentary Behavior: The Home Physical Activity Equipment scale exhibited robust convergent correlations with youth leisure-time sports engagement (r = 0.31 to 0.44, p < 0.001), while displaying expected inverse associations with sedentary screen time when active video games or outdoor recreational gear were available.

Discriminant Validity

Discriminant validity was confirmed through known-groups methodologies. The survey successfully differentiated between socioeconomically and urbanistically disparate neighborhoods (e.g., high-walkable/high-income, high-walkable/low-income, low-walkable/high-income, and low-walkable/low-income quartiles). Environmental aesthetic incivilities (trash, graffiti, abandoned structures) and safety concerns (crime, stray dogs) were significantly elevated in lower-income urban sectors, matching municipal crime registries and independent observational audits conducted using the Physical Activity Resource Assessment (PARA) and the Environmental Assessment of Public Recreation Spaces (EAPRS) tools.

Reliability

The Active Where? Surveys have been subjected to rigorous reliability assessments, including test-retest evaluations over 2-week intervals and multi-informant concordance analyses across parent-child dyads:

Test-Retest Reliability

Across validation cohorts in California, Ohio, and Massachusetts, temporal stability was evaluated by administering the survey twice across an interval of 10 to 14 days under identical conditions. Reliability statistics varied by domain, reflecting differences between concrete material counts and subjective perceptual evaluations:

• Home Physical Activity Equipment: Exhibited the highest temporal stability, with test-retest intra-class correlation coefficients (ICCs) and kappa coefficients spanning 0.78 to 0.91 across parent and adolescent respondents, indicating substantial to near-perfect agreement.
• Walking Proximity to Destinations: Proximity metrics to neighborhood stores, parks, and schools demonstrated ICCs ranging from 0.64 to 0.82, showing strong consistency in temporal estimation.
• Pedestrian/Bicycle Infrastructure & Aesthetics: Subscales evaluating sidewalks, crosswalks, and aesthetic attributes showed moderate-to-substantial reliability, with ICCs between 0.58 and 0.74.
• Traffic and Crime Safety: Perceived traffic and crime scales produced ICCs spanning 0.56 to 0.71. Adolescent reports displayed marginally lower stability than parent reports on crime metrics, reflecting adolescent fluctuations in spatial autonomy and varying daily exposure to neighborhood micro-zones.

Internal Consistency

Internal consistency estimates (Cronbach’s α) for the multi-item continuous environmental subscales conformed to accepted psychometric benchmarks for epidemiological screening tools:

• Walking/Biking Facilities: α = 0.72 to 0.79
• Aesthetics: α = 0.74 to 0.81
• Safety from Traffic: α = 0.68 to 0.75
• Safety from Crime: α = 0.76 to 0.84
• Park Quality: α = 0.71 to 0.80

Inter-Rater / Dyadic Agreement

Parent-adolescent concordance analyses revealed moderate overall agreement (inter-rater ICCs ranging from 0.42 to 0.65 across subscales). The lowest dyadic concordance occurred on safety from crime and traffic parameters, illustrating that parents frequently perceive higher contextual risk than their adolescent offspring. This divergence confirms the psychometric necessity of collecting independent responses from both members of the dyad rather than relying exclusively on proxy administration.

Factor Analysis

The latent structure of the Active Where? Surveys was established through an integrated sequence of exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) performed on split validation samples.

Exploratory Factor Analysis (EFA)

Initial EFA using principal axis factoring with promax (oblique) rotation was conducted on neighborhood environment items to uncover underlying latent dimensions without forcing artificial orthogonality. Scree plot visual inspection, parallel analysis, and eigenvalues greater than 1.0 supported an 8-factor solution for the broader ecological battery, representing: (1) Pedestrian Infrastructure, (2) Traffic Speed and Hazards, (3) Crime Safety, (4) Aesthetic Appeal, (5) Incivilities and Blight, (6) Recreational Affordances, (7) Domestic Resources, and (8) Destination Proximity. Item factor loadings were robust, with primary loadings spanning from 0.44 to 0.86, and minimal cross-loadings (<0.25) across secondary factors.

Confirmatory Factor Analysis (CFA)

Subsequent CFA conducted on an independent validation cohort verified the structural integrity of the multi-factor solution. Given the ordinal nature of Likert and proximity indicators, robust diagonally weighted least squares (WLSMV) estimation was utilized. The measurement model demonstrated excellent goodness-of-fit indices:

• Comparative Fit Index (CFI): 0.948 (surpassing the conventional 0.90 threshold)
• Tucker-Lewis Index (TLI): 0.939
• Root Mean Square Error of Approximation (RMSEA): 0.042 (90% CI [0.038, 0.047]), indicating minimal residual variance
• Standardized Root Mean Square Residual (SRMR): 0.051

Standardized factor loadings in the final CFA models were uniformly high and statistically significant (p < 0.001):

• Items assessing pedestrian sidewalks and crosswalk safety loaded between 0.58 and 0.81 onto Pedestrian Infrastructure.
• Aesthetic items (trees, architectural interest) loaded between 0.52 and 0.74, while physical disorder items (litter, graffiti, abandoned buildings) loaded strongly (0.64 to 0.85) on a distinct Physical Incivilities factor.
• Crime safety items (walking day/night, dog threats) loaded between 0.69 and 0.88 onto Crime Safety.

Instrument / Measurement Tool

The primary core modules of the Active Where? Surveys assess environmental affordances through 48 standardized items divided into four distinct operational sections:

• Test Type: Standardized self-report and proxy-report behavioral-ecological survey battery.
• Target Population: Parent-child dyads (children aged 5–11 years), parent-adolescent dyads (adolescents aged 12–18 years), or independent adolescent respondents.
• Administration Format: Paper-and-pencil questionnaire, supervised computer-assisted self-interview (CASI), or digital survey platform (e.g., REDCap, Qualtrics).
• Completion Duration: Approximately 15 to 25 minutes for the core environmental modules.
• Item Count: 48 verified core environmental items (20 Neighborhood Environment items, 10 Walking Proximity items, 10 Home Physical Activity Equipment items, and 8 Recreation/Park Facilities items).
• Response Formats:
  • Neighborhood Environment & Recreation/Park Facilities (Items 1–20, 41–48): 4-point Likert agreement scale (1 = Strongly Disagree, 2 = Somewhat Disagree, 3 = Somewhat Agree, 4 = Strongly Agree).
  • Walking Proximity to Destinations (Items 21–30): Categorical time scale (1 = 1 to 5 min, 2 = 6 to 10 min, 3 = 11 to 20 min, 4 = 21 to 30 min, 5 = 31+ min, 6 = Don’t know).
  • Home Physical Activity Equipment (Items 31–40): Binary dichotomous format (0 = No, 1 = Yes) or count format.
• Scoring and Computational Rules:
  • Subscale Computation: Subscales are derived by calculating the mean of non-missing items within each designated domain. A standard 75% item-completion threshold per subscale is applied; if more than 25% of items within a subscale are omitted, the subscale score is set to missing.
  • Reverse Coding Requirements: Negatively phrased items must be reverse-coded prior to subscale computation so that higher aggregate scores consistently reflect higher walkability, superior aesthetic amenities, or heightened safety. Specifically: Items 11 (Litter/trash), 12 (Graffiti), 13 (Abandoned buildings), 14 (Traffic volume walking barrier), 16 (Drivers exceeding speed limits), 18 (Crime unsafe day), 19 (Crime unsafe night), and 20 (Stray/unleashed dogs) are reverse-coded: Recoded_Score = 5 - Original_Score.
  • Destination Proximity Recoding: For walking proximity items, code ‘6’ (Don’t know) is typically treated as missing or recoded to the most distant category (>30 minutes / category 5) depending on analytical design. Proximity metrics may be reverse-coded (e.g., 6 - Score) so that higher values indicate closer proximity.
  • Home Equipment Summary: The Home Physical Activity Equipment index is calculated as a cumulative count of available equipment (ranging from 0 to 10).

Permissions & Fee and Test Year

The Active Where? Surveys were developed and refined between 2004 and 2007 through research funding provided by the Robert Wood Johnson Foundation (via the national Active Living Research program) and the National Institutes of Health (NIH). Initial comprehensive psychometric validation data and reliability reports were formally released in 2007.

The instrument is situated in the public domain as an open-access academic resource. There are no fees or licensing royalties required for non-commercial educational, clinical, or scientific research usage. Users are requested to retain standard scholarly attribution by citing the original instrument developers (Sallis, Saelens, Conway, and colleagues) and the foundational validation literature. Modifications, adaptations, or programmatic computerized implementations must document all altered items to maintain psychometric comparability.

References

• Conway, T. L., Sallis, J. F., Frank, L. D., Saelens, B. E., Cain, K. L., Slymen, D. J., & Frank, E. D. (2007). Active Where? Surveys: Youth and Parent Reliability Report. San Diego State University and Seattle Children’s Hospital Research Institute. https://activelivingresearch.org/active-where-surveys-and-reliability-report
• Ding, D., Sallis, J. F., Kerr, J., Lee, S., & Rosenberg, D. E. (2011). Neighborhood environment and physical activity among youth: A review. American Journal of Preventive Medicine, 41(4), 442–455. https://doi.org/10.1016/j.amepre.2011.06.036
• Gibson, J. J. (1979). The Ecological Approach to Visual Perception. Houghton Mifflin.
• Kerr, J., Frank, L., Sallis, J. F., Saelens, B., Glanz, K., & Chapman, J. (2006). Predictors of walking for travel and recreation in children: How built environment and safety impact walking to school. American Journal of Health Promotion, 20(4), 274–276. https://doi.org/10.4278/0890-1171-20.4.274
• Saelens, B. E., Sallis, J. F., Black, J. B., & Chen, D. (2003). Neighborhood-based differences in physical activity: An environment scale evaluation. American Journal of Public Health, 93(9), 1552–1558. https://doi.org/10.2105/ajph.93.9.1552
• Sallis, J. F., Cervero, R. B., Ascher, W., Henderson, K. A., Kraft, M. K., & Kerr, J. (2006). An ecological approach to creating active living communities. Annual Review of Public Health, 27, 297–322. https://doi.org/10.1146/annurev.publhealth.27.021405.102100

13. Items of the Scale (Questionnaire)