Physical Activity Environment
MEASURES REGISTRY USER GUIDE
SECTION
5
Examples of Measures with Reliability and/or Validity Evidence
This section highlights a sample of physical activity environment measures included in the Measures Registry for each setting and method that have evidence for reliability and/or validity. It is important to note that this information is not based on a systematic review, and other measures not listed here also have strong measurement properties. Table 1 lists examples of physical activity environment measures by method and setting.
Table 1: Examples of commonly used built environment measures with evidence of reliability and/or validity
GIS-based Measures
Community design and transportation system. GIS-based community design variables with evidence of construct validity include residential density, land use mix, and access to parks and recreation facilities. Transportation variables include street connectivity and access to transit. Walkability indices with evidence of construct validity include those by Cervero and Kockelman,74 Ewing et al.,75 and Frank et al.76 The latter index involves a summation of z-score variables, whereas the other two indices are computed using principal components analysis. Brownson et al. provide a detailed list of GIS-based variables and associated data sources.68 Walk Score is an example of an index with evidence of validity, but it is unique in that walkability scores are provided on a publicly available website free of charge. The user simply types in an address, and a Walk Score ranging from zero to 100 is provided. The Walk Score takes into account distance to amenities such as schools, parks, businesses, and other common destinations, as well as measures of street connectivity and residential density.77 The Walk Score creators have made an effort to align their algorithm with academic research, and several studies have found that Walk Score correlates well with measures of street connectivity, residential density, density of retail destinations, and access to other destinations.78-80 However, limited evidence exists regarding associations between Walk Score and physical activity, and validity of other metrics provided by Walk Score, such as Bike Score, Transit Score, and Crime Grade.
Observational Measures
Streetscapes. Streetscape audit tools with evidence of reliability and validity include the Microscale Audit of Pedestrian Streetscapes (MAPS),81 Systematic Pedestrian and Cycling Environment Scan (SPACES),82 Irvine-Minnesota Inventory (IMI),83 and Pedestrian Environment Data Scan (PEDS).84 MAPS includes a comprehensive scoring system, a limitation of other tools,81 as well as a short version designed for practitioners.85 Some streetscape audit tools were designed for specific settings, such as around schools (e.g., Texas Childhood Obesity Prevention Policy Evaluation School Environmental Audit Tool),86 and worksites (e.g., Workplace Walkability Audit),87 and in street alleys (e.g., Systematic Pedestrian and Cycling Environmental Scan for Alleys [SPACES for alleys]).88
Trails, parks, and recreation. The Path Environment Audit Tool (PEAT) has evidence of reliability and validity for assessing trails.89 Park assessment tools with evidence of reliability and validity include the Environmental Assessment of Public Recreation Spaces (EAPRS)57 and Community Park Audit Tool (CPAT).90 Both are comprehensive, but CPAT may be more feasible for use by practitioners. The Physical Activity Resource Assessment (PARA) can be used to assess trails, parks, and recreation areas, and has the advantage of being brief and easy to complete.91 A more comprehensive tool for assessing recreation environments is the Recreation Facility Audit Tool (REFAT), which was adapted from the PARA and has evidence of construct validity.92
Schools and child care. A comprehensive school environment audit tool with evidence of reliability and validity was developed by Jones et al.58 Audit tools also exist for the childcare environment, such as the Environment and Policy Assessment and Observation (EPAO).93
Homes. Home audit tools with evidence of reliability include the Home Food and Activity Assessment94 and Healthy Homes Survey.95 The latter is unique because it can be conducted through a telephone survey, and involves a parent doing an audit of his or her home.
Workplaces and other buildings. Physical activity environment assessment tools for worksites have been reviewed by Hipp et al.96 The Checklist of Health Promotion Environments at Worksites (CHEW)97 and Environmental Assessment Tool (EAT)98 are comprehensive tools for assessing worksites that have evidence for reliability and validity. Evidence of reliability and validity also exists for tools that assess specific building types, such as congregate living facilities for older adults (Audit of Physical Activity Resources for Seniors [APARS]).99
Rural. The Rural Active Living Assessment (RALA) includes a street segment audit tool that has evidence of reliability and validity.100
Questionnaires
Community design, transportation system, and streetscapes. Several report tools are commonly used and have evidence of reliability and validity for assessing neighborhood walkability factors. These include the Neighborhood Environment Walkability Scale (NEWS),51 an abbreviated NEWS,101 St. Louis Environmental Instrument,102 and Neighborhood Physical Activity Questionnaire (NPAQ).103 The Physical Activity Neighborhood Environment Scale (PANES) has acceptable reliability and validity properties but differs from the aforementioned tools because it is brief (17 items).73
Trails, parks, and recreation. Trails, parks, and recreation environments are typically assessed using audit tools. Slater et al. developed an adolescent questionnaire for assessing park environments.104,105 For trails, the Research on Urban Trail Environments (ROUTES) trail use report tool includes items assessing trail environment characteristics.106
Schools. The ActiveWhere? study developed and evaluated reliability and validity for a set of questions on the availability of several physical activity-related school facilities.107 The School Physical Activity Policy Assessment (S-PAPA) is a detailed report tool with reliability and validity that is primarily used for assessing school physical activity practices and policies, but also includes a small number of questions on the physical environment.108 For assessing child care environments, the Nutrition and Physical Activity Self-Assessment for Child Care (NAP SACC) has evidence of reliability and validity.109
Homes. Report tools developed by Hume at al.110,111 and Rosenberg et al.112 have evidence of reliability and validity for assessing home environments related to youth physical activity and sedentary behaviors.
Workplaces and other buildings. The Worksite and Energy Balance Survey (WEBS) includes 72 items on physical activity as well as nutrition environments in the worksite and has evidence for reliability and validity.113 A briefer tool with evidence of reliability and validity is the Worksite Supportive Environments for Active Living Survey (SEALS).114 Also with evidence of reliability and validity, the Office Environment and Sitting Scale specifically assesses the sitting environment in the worksite.115
Rural. The Rural Active Living Assessment (RALA) includes interview questions that are completed by community members or key informants. The questions cover availability of recreation facilities, programs, and policies, and have evidence of reliability and validity.100 Also with evidence of reliability and validity, the Rural Active Living Perceived Environmental Support Scale (RALPESS) can be used to assess participants’ perceptions of their environment. The tool covers: (1) church facilities, (2) town center connectivity, (3) indoor areas, (4) around the home/neighborhood, (5) town center physical activity resources, (6) school grounds, and (7) outdoor areas.116
Assessing Use Of Specific Environments And Physical Activity In Specific Environments
Questionnaires and observational measures. Tools measuring use of specific environments are not presented in the Measures Registry. Examples of questionnaires to measure these constructs can be found on Dr. James Sallis’ website. Counting users, pedestrians, and/or bicyclists within specific environments can provide an objective way of assessing use of environments, and counting can be done manually or with automated devices.117 Many cities and counties use automated counters to track pedestrian and bicycle traffic, and some parks track users in a similar manner. Issues to consider with manual counting include when and how frequently to collect assessments. Both manual and automated counting strategies may need to take into account non-independence (i.e., people being counted at multiple sites).
Commonly used tools for assessing physical activity in specific environments include the observational audit tools System for Observing Play and Recreation in Communities (SOPARC)118 and System for Observing Play and Leisure Activity in Youth (SOPLAY).119 These tools include checklists of environmental features to describe the environments being captured, and can be found in the Measures Registry.
GPS. Several studies have used person-worn GPS tracking devices in physical activity research120 to capture the environments that an individual encounters across the day rather than simply the environment around a specific origin (e.g., home neighborhood). These methods allow investigation of the environments in which physical activity occurs as well as how exposure to various spatial contexts may influence physical activity. Several tools exist for making the use of GPS data in built environment research feasible (e.g., PALMS).121
Gaps and Limitations of Existing Measures
As is apparent in Table 1, few to no strong measures exist for some methods and settings (e.g., home audits, park self-report measures). In some settings multiple measures exist, making it difficult to choose (e.g., streetscape audit and report tools). Several existing measures with evidence for reliability and validity are lengthy and require complex training and/or scoring procedures because they were designed for research. Such tools may not fit the needs of practitioners because of their cost, complexity, and/or user burden. Much progress still needs to be made before both research- and practice-quality tools exist for all methods and settings.