Introduction 介绍

Physical activity provides numerous health benefits for children, including higher cardiorespiratory fitness, muscular fitness, bone health, and cardiometabolic health, as well as having a beneficial effect on mental health [1, 2]. To achieve these health benefits, it is recommended that children and adolescents aged 6 through 17 years should engage in at least 60 min of moderate-to-vigorous physical activity daily [3]. However, in the United States (U.S.), fewer than one quarter (24%) of children and adolescents meet these guidelines, leading to a need for strategies to increase daily physical activity among youth [4].
体育活动为儿童提供了许多健康益处，包括提高心肺健康、肌肉健康、骨骼健康和心脏代谢健康，以及对心理健康产生有益影响 [ 1， 2]。为了实现这些健康益处，建议 6 至 17 岁的儿童和青少年每天至少进行 60 分钟的中等强度到剧烈的体育活动 [ 3]。然而，在美国 （U.S.），只有不到四分之一 （24%） 的儿童和青少年符合这些指南，因此需要制定策略来增加青少年的日常身体活动 [ 4]。

For all age groups, active travel has the potential to be a meaningful contributor to overall physical activity [5]. For children, promoting active commuting to school (ACS) can be an effective strategy to increase physical activity [6]. Investing in urban design features and transportation policies that support active travel and public transit use are 2 of the 8 intervention approaches that work for promoting physical activity [7, 8]. There was a time when ACS was common in the U.S., and thus, an important contributor to physical activity among children. However, car-centric urban design decisions and car-dependency have resulted in very low levels of participation in ACS, with only 10.9% of children age 5–17 walking or bicycling to school in 2017 compared to 47.7% in 1969 [9, 10]. This is even true for children who live within a reasonable active travel distance to school [10]. Additional reasons for the dramatic decline in ACS include that the distance from school that students live increased due to school preference and open transfer policies, urban sprawl, school siting guidelines, and small school closure [11,12,13]. Additionally, children’s independent mobility has decreased over the past decades, so fewer children are allowed to travel to/from school without adult supervision [14].
对于所有年龄段的人来说，积极的旅行有可能对整体身体活动做出有意义的贡献 [ 5]。对于儿童来说，促进积极通勤上学 （ACS） 可能是增加身体活动的有效策略 [ 6]。投资于支持积极出行和公共交通使用的城市设计特色和交通政策是促进体育活动的 8 种干预方法中的 2 种 [ 7， 8]。曾几何时，ACS 在美国很常见，因此是儿童身体活动的重要因素。然而，以汽车为中心的城市设计决策和对汽车的依赖导致 ACS 的参与率非常低，2017 年只有 10.9% 的 5-17 岁儿童步行或骑自行车上学，而 1969 年为 47.7% [ 9， 10]。对于生活在合理的主动上学距离内的儿童来说也是如此 [ 10]。ACS 急剧下降的其他原因包括，由于学校偏好和开放式转学政策、城市扩张、学校选址指南和小型学校关闭，学生与学校的距离增加 [ 11， 12， 13]。此外，在过去几十年中，儿童的独立行动能力下降，因此在没有成人监督的情况下上下学的儿童越来越少 [ 14]。

Parental perception of traffic safety is one of the primary reasons why even children living within walking or bicycling distance to school do not engage in ACS, and though these perceptions play a strong role on whether children actively commute to school or not, it is unlikely to improve perceptions if the reality of the environment is not safe and conducive to walking or biking [15]. So, measuring and improving built environments to become safe for ACS should always be the first step to actually optimize the real safety of the environment before trying to improve parental perceptions of the safety of the environment.
父母对交通安全的看法是为什么即使是住在步行或骑自行车上学距离内的孩子也不参与 ACS 的主要原因之一，尽管这些看法对儿童是否积极上学起着重要作用，但如果环境的现实不安全且不利于步行或骑自行车，则不太可能改善感知 [ 15]。因此，在尝试改善父母对环境安全的看法之前，测量和改进建筑环境以使其对 ACS 来说是安全的，这应始终是真正优化环境真正安全的第一步。

Many built environment elements are known to be associated with active travel among children, and most evidence is from macro-level environment features, including neighborhood residential density, road-network connectivity, and land-use mix, though there have been mixed findings about the associations between macro-level features and active transportation among children [15,16,17]. In addition to the macro-level features that encourage active travel, micro-scale factors are physical built environment design features along streets or segments and can provide qualities like comfort, safety, attractiveness that pedestrians and cyclists seek [18]. There are many physical components that make up the micro-scale built environment, including whether there are trees and shade along the route, the presence of safe crossings at intersections, stoplights that work, or whether the sidewalk is continuous and well-maintained [19, 20]. Micro-scale features are relevant for children traveling relatively small distances, such as to or from school, as they have the potential to influence perceptions of traffic safety, the major barrier to ACS as reported by parents. More specifically, lack of sidewalks, presence of sidewalk obstructions, and intersection features have been identified as top concerns for parents along routes their children could actively travel [21, 22] Safe Routes to School (SRTS) programs, often funded through an existing federal policy aimed at promoting safe active travel to school, often target micro-scale features near schools within engineering projects [23]. Despite the on-going and existing implementation of SRTS engineering projects near schools [24], there is little evidence to evaluate the specific microscale features that may help promote increases in ACS among children living within “active travel distance” to schools. This evidence would inform interventions like SRTS programs of modifiable urban design strategies and identify school environments that need to be prioritized.
众所周知，许多建筑环境元素与儿童的积极出行有关，大多数证据来自宏观环境特征，包括社区住宅密度、道路网络连通性和土地利用混合，尽管关于宏观特征与儿童积极交通之间的关联存在混合发现[15,16,17]。除了鼓励积极出行的宏观特征外，微观尺度因素是沿街道或路段的物理建筑环境设计特征，可以提供行人和骑行者所寻求的舒适、安全、吸引力等品质 [ 18]。构成微型建筑环境的物理组成部分很多，包括沿途是否有树木和树荫，十字路口是否有安全交叉路口，红绿灯是否有效，或者人行道是否连续且维护良好 [ 19， 20]。微尺度特征与出行距离相对较短的儿童（例如上学或放学）相关，因为它们有可能影响对交通安全的看法，而交通安全是家长报告的 ACS 的主要障碍。更具体地说，缺乏人行道、存在人行道障碍物和十字路口特征已被确定为父母在孩子可以积极出行的路线上最关心的问题 [ 21， 22] 安全上学路线 （SRTS） 计划，通常由旨在促进安全主动上学的现有联邦政策资助，通常针对工程项目中学校附近的微尺度特征 [ 23]。 尽管在学校附近正在进行和已经实施了 SRTS 工程项目 [ 24]，但几乎没有证据可以评估可能有助于促进生活在学校“主动旅行距离”内儿童 ACS 增加的特定微尺度特征。这些证据将为 SRTS 计划等干预措施提供可修改的城市设计策略的信息，并确定需要优先考虑的学校环境。

There remains a need for a reliable and practical tool that can be used by researchers, practitioners, and community members to evaluate SRTS interventions, to document current conditions around schools to inform policy actions, to identify priority areas for investment, or to examine the moderating effect of the micro-level street environment on other types of interventions that promote ACS. There are several existing street audit tools to measure micro-scale characteristics specific to the school neighborhood environment or travel routes to school [25,26,27,28]. The Texas Childhood Obesity Prevention Policy Evaluation (T-COPPE) School Environment Audit Tool, a reliable instrument, captures the modifiable attributes of the built environments within school neighborhoods, but the authors did not determine a specific street sampling method for determining the school environment, which limits the utility of the tool [26]. Lee et al. (2020) adapted the T-COPPE audit tool and combined it with geographic information systems (GIS) measures of the macro-level built environment to develop a school walkability index within a 0.4 km Euclidean buffer [29]. However, the range of segments audited in this study was 15–87 segments within the buffer, indicating that audits for schools located in areas with high street connectivity would be resource intensive. Similarly, the Irvine-Minnesota Inventory (IMI) walkability audits assessed several micro-scale factors, including those related to traffic safety, accessibility, pleasurable settings, crime safety, density of housing, and diverse destinations [28]. However, this audit assessed individual children’s walking routes to school, making it resource intensive to gather enough information from enough home-school routes to make assessments about a school’s micro-scale environment. Lastly, Jones et al., (2010) developed a reliable and validated 44-item audit tool to assess aspects of the elementary school grounds but did not capture the street network around the schools [26]. As a result, this tool cannot be utilized within research seeking to understand how micro-scale environments impact active commuting to school behavior which takes place on streets near schools.
仍然需要一种可靠且实用的工具，研究人员、从业人员和社区成员可以使用它来评估 SRTS 干预措施，记录学校周围的现状，为政策行动提供信息，确定优先投资领域，或检查微观街道环境对促进 ACS 的其他类型的干预措施的调节作用。有几种现有的街道审计工具可以测量特定于学校社区环境或上学路线的微观特征 [ 25， 26， 27， 28]。德克萨斯州儿童肥胖预防政策评估 （T-COPPE） 学校环境审计工具是一种可靠的工具，它捕获了学校社区内建筑环境的可修改属性，但作者没有确定确定确定学校环境的特定街道采样方法，这限制了该工具的实用性 [ 26]。Lee et al. （2020） 采用了 T-COPPE 审计工具，并将其与宏观建筑环境的地理信息系统 （GIS） 测量相结合，在 0.4 km 欧几里得缓冲区内开发了学校步行指数 [ 29]。然而，本研究中审计的路段范围为缓冲区内的 15-87 个路段，这表明对位于商业街连接地区的学校的审计将是资源密集型的。同样，尔湾-明尼苏达州清单 （IMI） 步行性审计评估了几个微观因素，包括与交通安全、可达性、愉悦环境、犯罪安全、住房密度和不同目的地相关的因素 [ 28]。 然而，这项审计评估了每个儿童上学的步行路线，这使得从足够多的家庭学校路线中收集足够的信息以评估学校的微观环境成为资源密集型的。最后，Jones et al.， （2010） 开发了一种可靠且经过验证的 44 项审计工具来评估小学场地的各个方面，但没有捕获学校周围的街道网络 [ 26]。因此，该工具不能用于寻求了解微观环境如何影响发生在学校附近街道上的积极通勤行为的研究。

The Micro-scale Audit of Pedestrian Streetscapes (MAPS) direct observation tool was developed in 2013 to assess micro-level features of neighborhood walkability for general populations (i.e., without a specific focus on children). The MAPS tool has been previously validated as being related to active transport in all age groups [30, 31], but this version of the tool was developed for usage around individual’s homes on routes to various destinations, not the school environment. MAPS was the first audit tool to provide a systematic scoring system for summarizing data, and the intention of the tool was to be modified for different environments, countries, populations, and outcomes [31]. Pocock et al. (2020) adapted MAPS to develop the Microscale Audit of Pedestrian Streetscapes Global–School Neighbourhood (MAPS Global-SN) tool for adolescent school environments in New Zealand [27]. The MAPS Global-SN tool also requires GIS to determine the observation route, limiting the use of the tool those who have access to the software and required skills.
行人街景微尺度审计 （MAPS） 直接观察工具于 2013 年开发，用于评估普通人群（即不特别关注儿童）社区步行性的微观特征。MAPS 工具之前已被验证与所有年龄段的主动交通相关 [ 30， 31]，但此版本的工具是为在通往不同目的地的路线上的个人家中使用而开发的，而不是学校环境。MAPS 是第一个提供系统评分系统来总结数据的审计工具，该工具的目的是针对不同的环境、国家、人群和结果进行修改 [ 31]。Pocock et al. （2020） 采用 MAPS 为新西兰青少年学校环境开发了全球行人街景微观审计 （MAPS Global-SN） 工具 [ 27]。MAPS Global-SN 工具还需要 GIS 来确定观测路线，从而限制了有权访问该软件并具备所需技能的用户使用该工具。

Overall, there remains a need for a school-specific, micro-scale observation instrument that is centered around the theme of ACS,provides reliable data, and could be used by researchers, practitioners, urban planners, schools, and community members to understand the micro-scale elements of the school neighborhood environment in the U.S. This school-specific tool would be particularly useful for evaluation of SRTS engineering projects, which are often focused on targeting micro-scale factors within school environments. Therefore, the objective of the present paper is to describe the adaptation of MAPS for use in assessing micro-scale school environments in an urban context and evaluate inter-observer reliability of the adapted tool, which can then be used to determine which micro-scale built environment features are associated with children’s ACS to support future planning and design standards for the school environment.
总体而言，仍然需要一种以 ACS 主题为中心的特定于学校的微型观测仪器，提供可靠的数据，并可供研究人员、从业者、城市规划者、学校和社区成员使用，以了解美国学校社区环境的微观元素。这个特定于学校的工具对于评估 SRTS 工程项目特别有用，这些项目通常侧重于针对学校环境中的微观因素。因此，本文的目的是描述 MAPS 在城市环境中用于评估微尺度学校环境的适应性，并评估适应性工具的观察者间可靠性，然后可用于确定哪些微尺度建筑环境特征与儿童的 ACS 相关联，以支持未来的学校环境规划和设计标准。

Methods 方法

Setting and study design 设置和研究设计

This study used baseline data collected in 2018–2019 from elementary schools involved in the Safe Travel Environment Evaluation in Texas Schools (STREETS) study. The STREETS study is a five-year natural experiment that assesses the impact of Safe Routes to School infrastructure projects funded by a 2016 bond initiative from the City of Austin on children’s physical activity and ACS. The setting for this study was Austin, Texas, U.S.A, the capital city of the state, with a population of 907,779 in 2016, the year the bond initiative was passed, and which has been undergoing rapid urban expansion since the early 2000s [32, 33].
本研究使用了 2018-2019 年从参与德克萨斯州学校安全旅行环境评估 （STREETS） 研究的小学收集的基线数据。STREETS 研究是一项为期五年的自然实验，旨在评估由奥斯汀市 2016 年债券倡议资助的“安全上学路线”基础设施项目对儿童身体活动和 ACS 的影响。本研究的背景是美国德克萨斯州奥斯汀，该州的首府，2016 年人口为 907,779 人，即债券倡议通过的那一年，自 2000 年代初以来一直在经历快速的城市扩张 [ 32， 33]。

The STREETS study utilized a serial cross-sectional design to assess population-level changes in the prevalence of ACS in elementary schools in Central Texas. A subset of the schools recruited into the serial cross-sectional study were also recruited to be a part of a quasi-experimental, prospective cohort study to examine changes in child physical activity levels and psychosocial outcomes. This study used data from the 36 elementary schools recruited into the cohort study. Full methods of the STREETS study have been presented elsewhere [24].
STREETS 研究利用一系列横断面设计来评估德克萨斯州中部小学 ACS 患病率的人群水平变化。被招募到系列横断面研究的学校子集也被招募作为准实验、前瞻性队列研究的一部分，以检查儿童身体活动水平和社会心理结果的变化。这项研究使用了来自队列研究中招募的 36 所小学的数据。STREETS 研究的完整方法已在别处介绍 [ 24]。

Micro-scale Audit of Pedestrian Streetscapes for Safe Routes to School (MAPS-SRTS) tool development
安全上学路线的行人街景微尺度审计 （MAPS-SRTS） 工具开发

The Micro-scale Audit of Pedestrian Streetscapes for Safe Routes to School (MAPS-SRTS) tool was adapted from the MAPS direct observation (or audit) instrument [31] and the MAPS-Abbreviated instrument, a shorter version of MAPS tool [30]. The MAPS tool consisted of 120 micro-scale environmental items (e.g. presence and width of sidewalk, presence of sidewalk obstructions, shade coverage of sidewalks, and presence and quality of marked crosswalks) that potentially influence physical activity, and there were four sections used to measure different components of the streetscape: overall route, street segments, crossings, and cul-de-sacs. A street segment was defined as a section of street or road between two crossings on one side of the street (one block). A crossing occurred when the rater would travel through an intersection, whether a painted pedestrian crossing existed or not. Crossings were located between two segments and were coded any place two roads intersected.
安全上学路线的行人街景微尺度审计 （MAPS-SRTS） 工具改编自 MAPS 直接观察（或审计）工具 [ 31] 和 MAPS 缩写工具，MAPS 工具的简化版本 [ 30]。MAPS 工具由 120 个可能影响身体活动的微尺度环境项目（例如人行道的存在和宽度、人行道障碍物的存在、人行道的阴影覆盖以及标记人行横道的存在和质量）组成，有四个部分用于测量街景的不同组成部分：总体路线、路段、十字路口和死胡同。街段被定义为街道一侧（一个街区）的两个十字路口之间的一段街道或道路。当评分者穿过十字路口时，无论是否存在涂漆的人行横道，都会发生过马路。十字路口位于两个路段之间，并在两条道路相交的任何地方进行编码。

In the MAPS tool, items collected for each section were summarized into subscales with either positive or negative valence scores. Audit data were collected by trained observers who followed a 0.25 mile route from each participant’s home address towards a pre-determined destination, which was typically a cluster of commercial locations or a park. The items and subscales from the MAPS tool had moderate to excellent inter-rater reliability (ICC values ≥ 0.41 and ≥ 0.60, respectively) [31]. MAPS-Abbreviated included 54 items from MAPS, which were selected by investigating the item-level partial correlations with physical activity across age groups (i.e., children, adolescents, younger adults, older adults), with a specific focus on walking or bicycling for transportation [30]. Items significantly correlated with active transport were included, but those with low frequency, limited policy relevance, or required too much labor to rate were dropped. Similar to the protocol for the full-length version of the MAPS tool, MAPS-Abbreviated collected audit data using 0.25-0.45-mile routes from participant homes towards predetermined destinations. The MAPS-Abbreviated and MAPS total scores have been reported to be strongly correlated with each other (r = 0.94) and with physical activity outcomes among children [30].
在 MAPS 工具中，为每个部分收集的项目被总结为具有正或负效价分数的子量表。审计数据由训练有素的观察员收集，他们沿着从每个参与者的家庭住址到预定目的地的 0.25 英里路线，通常是一组商业地点或公园。MAPS 工具中的项目和分量表具有中等到极好的评分者间信度（ICC ≥值分别为 0.41 和 ≥ 0.60）[ 31]。MAPS-Abbreviated 包括 MAPS 中的 54 个项目，这些项目是通过调查不同年龄组（即儿童、青少年、年轻人、老年人）的身体活动的项目级部分相关性来选择的，特别关注步行或骑自行车作为交通工具[ 30]。与主动运输显著相关的项目被纳入，但那些频率低、政策相关性有限或需要太多劳动力才能评级的项目被删除。与 MAPS 工具全长版的协议类似，MAPS-Abbreviated 使用从参与者家到预定目的地的 0.25-0.45 英里路线收集审计数据。据报道，MAPS 缩写评分和 MAPS 总分彼此密切相关 （r = 0.94） 并与儿童的身体活动结果密切相关 [ 30]。

Adaptations for MAPS-SRTS
MAPS-SRTS 的调整

MAPS-SRTS has several modifications from the MAPS and MAPS-Abbreviated tools for the purpose of assessment of the neighborhood environment immediately adjacent to and around schools, emphasizing features considered most relevant to the behavior of active traveling to/from school. Modifications were made to the (1) structure and content of the audit tool sections, (2) observation route, (3) and scoring.
MAPS-SRTS 对 MAPS 和 MAPS 缩写工具进行了一些修改，用于评估学校附近和周围的社区环境，强调被认为与积极上下学行为最相关的特征。对 （1） 审计工具部分的结构和内容、（2） 观察路线、（3） 和评分进行了修改。

Structure and content 结构和内容

Several adaptations to the MAPS-SRTS tool related to the structure and content were made to ensure suitability for the school environment. The adaptation of the MAPS tool was initiated by a SRTS practitioner (co-author CS) to ensure that important and appropriate school-specific built environment features were included, and the adaptation was coordinated closely with an investigator from the MAPS tool (co-author JS). First, a section was added to assess micro-scale characteristics of the street segment or segments where children enter and exit the building (i.e., the street segment adjacent to the school, usually directly in front of the school property). This new section, called the “school access segment,” had the same items and subscales as the original segments section. A school access segment was defined as the section of street directly in front of the main school entrance between two intersections. The school access segment was intended to cover the entire front of the school building (including any places children can access the entrance). For schools that had more than one school access segment, such as if there was an intersection across from the entrance of the school, or if there were multiple entrances to the school on different segments, then all school access segments were audited. Driveways on the school access segment that were school entrances or exits were counted as crossings but did not divide the school access segment into two. Because the destination of the route for MAPS-SRTS was always the school, the destination, land use, and cul-de-sacs sections were not included in the MAPS-SRTS scoring. These items were removed to reduce the length of the instrument tool, and though land use has been shown to be associated with ACS in children [34], this is not a readily modifiable component of the school neighborhood environment, nor is it within the scope of SRTS projects, and this tool was designed for use among researchers, planners, and practitioners implementing SRTS projects.
对 MAPS-SRTS 工具的结构和内容进行了多次调整，以确保适合学校环境。MAPS 工具的改编是由 SRTS 从业者（合著者 CS）发起的，以确保包含重要和适当的学校特定建筑环境特征，并且与 MAPS 工具的研究人员（合著者 JS）密切协调调整。首先，增加了一个部分来评估儿童进出建筑物的一个或多个路段（即与学校相邻的路段，通常在学校财产的正前方）的微观特征。这个新部分称为“学校访问部分”，具有与原始部分部分相同的项目和分量表。学校通道路段被定义为两个十字路口之间学校主入口正前方的街道部分。学校通道部分旨在覆盖学校建筑的整个正面（包括儿童可以进入入口的任何地方）。对于具有多个学校访问段的学校，例如，如果学校入口对面有一个交叉路口，或者如果学校的不同段上有多个入口，则所有学校访问段都经过审核。学校通道段上作为学校入口或出口的车道被算作交叉口，但没有将学校通道段一分为二。由于 MAPS-SRTS 的路线目的地始终是学校，因此目的地、土地利用和死胡同路段不包括在 MAPS-SRTS 评分中。 删除这些项目是为了减少仪器工具的长度，尽管土地使用已被证明与儿童的 ACS 有关 [ 34]，但这不是学校社区环境的一个易于修改的组成部分，也不在 SRTS 项目的范围内，并且该工具是为研究人员设计的， 规划师和实施 SRTS 项目的从业者。

Because the environment around a school commonly has school-specific signage, several signage items were added to the school access segment section and segment section of the MAPS-SRTS tool to capture these elements. These items included: (1) the presence of school zone signage and (2) the presence of signage for a special speed zone during certain times of day (i.e. school drop-off and pick-up times).
由于学校周围的环境通常具有特定于学校的标牌，因此在 MAPS-SRTS 工具的学校访问段部分和段部分添加了多个标牌项目，以捕获这些元素。这些项目包括：（1） 存在学区标志，以及 （2） 在一天中的特定时间（即放学和接送时间）存在特殊速度区的标志。

Observation route 观察路线

For each school assessed using the MAPS-SRTS instrument, the observation route was first established (Fig. 1). The approach for route selection was based on ensuring measurement of segments and crossing that any kids that walk or cycle to school would have to use to access the school entrance. The observation route for each school began on the street where the school was located or the nearest street available, known as the school access segment, and the main school entrance was always the point of reference. The school entrance was confirmed by the study contact at each school. In addition to the school access segment, the route was determined using the “nearest-neighbor” method of spatial sampling [35]. This method used the following process for school observation route selection: (1) select all crossings that connect to the school access segment; (2) from each of those crossings, select all segments on one side of the street that connect to the crossing. Because it has been previously demonstrated that there is a high correlation between the features of both sides of the same street segment in micro-scale audit data, and to minimize data collection burden and make the tool practical for multiple audiences, including practitioners (non-researchers), it was decided to only measure one side of each street segment [27].This method captured the environment for those living within a very short distance (a few blocks) of school. While this sampling approach did not include all segments and crossings that a child would take to school, this route selection approach ensured that any route a child would take to school will include these components in any permutation or combination. This approach ensures that the final leg of the school journey was measured, and without an accessible final part of the trip, it does not matter how accessible the first parts of the route would be.
对于使用 MAPS-SRTS 仪器评估的每所学校，首先确定观察路线（图 1）。路线选择的方法基于确保测量任何步行或骑自行车上学的孩子都必须使用的路段和交叉路口才能进入学校入口。每所学校的观察路线从学校所在的街道或最近的可用街道（称为学校通道段）开始，学校的主要入口始终是参考点。学校入学由每所学校的研究联系人确认。除了学校访问部分外，该路线是使用空间采样的“最近邻”方法确定的 [ 35]。该方法使用以下过程进行学校观察路线选择：（1） 选择所有连接到学校访问段的交叉口;（2） 从每个十字路口中，选择街道一侧连接到十字路口的所有路段。因为之前已经证明，在微观尺度审计数据中，同一街段两侧的特征之间存在高度相关性，并且为了最大限度地减少数据收集负担并使该工具对包括从业者（非研究人员）在内的多个受众实用，因此决定只测量每个街段的一侧 [ 27]。这种方法捕捉了那些生活在学校很短（几个街区）内的人的环境。虽然这种抽样方法并未包括孩子上学的所有路段和交叉口，但这种路线选择方法确保孩子上学的任何路线都将以任何排列或组合包含这些组成部分。 这种方法确保了学校旅程的最后一段是经过测量的，如果没有旅行的最后一部分，路线的前半部分有多容易就无关紧要了。

Fig. 1 图 1
figure 1

Example school observation routes for use with the Micro-Scale Audit of Pedestrian Streetscapes for Safe Routes to School (MAPS-SRTS)
与安全上学路线的行人街景微尺度审计 （MAPS-SRTS） 一起使用的学校观察路线示例

Full size image 全尺寸图像

Scoring 得分

The MAPS-SRTS tool consisted of 90 items divided into three sections for data collection and scoring: (1) school access segments (33 items), (2) other segments near schools (30 items), and (3) crossings (27 items). The school access segment examines the micro-scale characteristics of the road where children enter and exit the school building. The other segment section of the instrument characterizes the roads surrounding and leading to the school, and the crossing section assesses characteristics of where pedestrians cross the road or school driveway. These items were used to compute multiple subscales for each section.
MAPS-SRTS 工具由 90 个项目组成，分为三个部分用于数据收集和评分：（1） 学校访问部分（33 个项目），（2） 学校附近的其他部分（30 个项目）和 （3） 十字路口（27 个项目）。学校通道部分研究了儿童进出学校建筑的道路的微观特征。该工具的另一部分描述了学校周围和通往学校的道路，而交叉部分则评估了行人穿过马路或学校车道的位置的特征。这些项目用于计算每个部分的多个分量表。

The subscale development for the MAPS-SRTS instrument relies on a hierarchical scoring system, as in the MAPS tool, where items collected for each section (school access segments, other segments near schools, and crossings) were summarized into subscales at several levels of aggregation and were rated as positive or negative valence scores, as shown in Fig. 2. The positive and negative scores for each subscale are combined into an overall score for each section. The total MAPS-SRTS score is an aggregate score of the overall school access segments, other segments near schools, and crossings scores, where a higher score indicates a more supportive micro-scale built environment for walking and bicycling to school. The initial MAPS-SRTS scoring schema involved 30 subscales, in addition to the total MAPS-SRTS score, including 11 school access subscales, 11 other segments near school subscales, and 8 crossing subscales.
MAPS-SRTS 工具的子量表开发依赖于分层评分系统，就像在 MAPS 工具中一样，其中为每个部分（学校访问部分、学校附近的其他部分和十字路口）收集的项目被总结为多个聚合级别的子量表，并被评为正或负效分数，如图 2 所示。每个子量表的正分和负分合并为每个部分的总分。MAPS-SRTS 总分是整个学校通道部分、学校附近的其他部分和十字路口分数的总分，其中分数越高表示步行和骑自行车上学的微型建筑环境更具支持性。除了 MAPS-SRTS 总分外，初始 MAPS-SRTS 评分方案还涉及 30 个分量表，包括 11 个学校访问分量表、11 个学校分量表附近的其他部分和 8 个交叉分量表。

Fig. 2 图 2
figure 2

Scoring schema of subscales and total score for the Micro-scale Aaudit of Pedestrian Streetscapes for Safe Routes to School (MAPS-SRTS) instrument
安全上学路线行人街景微尺度审计 （MAPS-SRTS） 工具的分量表评分方案和总分

Full size image 全尺寸图像

Data collection procedures
数据收集程序

Training 训练

Prior to data collection, each data collector underwent training both in the classroom and in the field. During 2 h of classroom training, data collectors watched a pre-recorded video by one of the lead investigators on the STREETS project. The classroom training included the following: (1) definition of the School Neighborhood Environment, (2) how to create and use printed maps to guide school audits, (3) a review of the contents of MAPS-SRTS Training Manual and Picture Guide, (4) expectations in the field, (5) how to use Qualtrics (Qualtrics, Provo, UT) instruments on iPad mini tablets (Apple Inc., Cupertino, CA, USA) for data collection, and practice on the tablets. During 3 h of field training, each data collector underwent hands-on training for each route type and practiced using both the tablet and hard-copy map/protocol. Each data collector was trained in the field by two of the lead data collectors across two school audits and had to reach consensus with trainers prior to going out into the field independently.
在数据收集之前，每个数据收集员都接受了课堂和现场培训。在 2 小时的课堂培训中，数据收集员观看了 STREETS 项目的一位首席调查员预先录制的视频。课堂培训包括以下内容：（1） 学校邻里环境的定义，（2） 如何创建和使用打印地图来指导学校审计，（3） MAPS-SRTS 培训手册和图片指南内容的回顾，（4） 现场期望，（5） 如何在 iPad mini 平板电脑上使用 Qualtrics（Qualtrics、Provo、UT）仪器（Apple Inc.， Cupertino， CA， USA） 进行数据收集，并在平板电脑上练习。在 3 小时的现场培训期间，每个数据收集员都接受了每种路线类型的实践培训，并使用平板电脑和硬拷贝地图/协议进行了练习。每个数据收集员都接受了两次学校审计中两名首席数据收集员的现场培训，并且在独立进入现场之前必须与培训师达成共识。

Data collection 数据采集

MAPS-SRTS baseline audits of 36 schools were completed between March 2019 and June 2021 by a total of 11 trained data collectors. During data collection, two raters audited each school. One person was responsible for entering data on the tablet, the other for managing the school neighborhood map and protocol. The two raters completed one or two school audits per day during the data collection period. Immediately following data collection, all data were uploaded to Qualtrics. The average time to complete audits was 77.8 min (SD = 29.5 min) per school, excluding travel time. To assess interrater reliability, 15% (n = 5) of schools were randomly selected and were independently assessed by two pairs of raters.
2019 年 3 月至 2021 年 6 月期间，共有 11 名训练有素的数据收集人员完成了对 36 所学校的 MAPS-SRTS 基线审计。在数据收集期间，两名评分员对每所学校进行审核。一个人负责在平板电脑上输入数据，另一个人负责管理学校社区地图和协议。在数据收集期间，两名评分员每天完成一到两次学校审计。数据收集后，所有数据立即上传到 Qualtrics。每所学校完成审计的平均时间为 77.8 分钟 （SD = 29.5 min），不包括旅行时间。为了评估评分者间的可靠性，随机选择了 15% （n = 5） 的学校，并由两对评分者独立评估。

Statistical analysis 统计分析

Analysis for each subscale within the originally proposed MAPS-SRTS instrument (n = 31 subscales) included descriptive and reliability analyses. The mean and standard deviation of scores for each subscale were calculated. To assess inter-rater reliability for each subscale, one-way random effects single-measure intraclass correlation coefficients (ICC) were used for ordinal and continuous scales, and an ICC of 0.60 or higher was deemed acceptable reliability [36]. For a subscale originally proposed in the MAPS-SRTS instrument to be included in the final scoring, decision rules were implemented based on reliability and theoretical relevance. Subscale reliability scores were considered acceptable if ICC values were classified as moderate or higher (ICC = 0.60 or higher). For subscales that originally included items or subscales with low reliability (ICC below 0.60), we excluded those items or subscales from aggregate scoring. In cases where an item had low variability in one section (e.g. school access segment section) and the same subscale was reliable in another section (e.g. segment section), the item or subscale was retained if the items did not reduce the aggregate section score’s reliability below acceptable standards. Items removed due to low reliability were removed from subscales in a stepwise process to assess the effects on the overall reliability of the subscale.
最初提议的 MAPS-SRTS 工具（n = 31 个分量表）中每个分量表的分析包括描述性和可靠性分析。计算每个分量表分数的平均值和标准差。为了评估每个分量表的评分者间信度，对顺序和连续量表使用单向随机效应单量组内相关系数 （ICC），ICC 为 0.60 或更高被认为是可接受的信度 [ 36]。对于最初在 MAPS-SRTS 工具中提议的要包含在最终评分中的子量表，根据可靠性和理论相关性实施了决策规则。如果 ICC 值被归类为中等或更高 （ICC = 0.60 或更高），则子量表可靠性评分被认为是可以接受的。对于最初包含低可靠性（ICC 低于 0.60）的项目或分量表的分量表，我们将这些项目或分量表排除在总分之外。如果一个项目在一个部分（例如学校访问部分）的可变性较低，而相同的分量表在另一个部分（例如部分）中是可靠的，则如果项目没有将总部分分数的可靠性降低到可接受的标准以下，则保留该项目或子量表。由于可靠性低而删除的项目以逐步过程从分量表中删除，以评估对子量表整体可靠性的影响。

Descriptive characteristics for each school were calculated using median and interquartile range. Geographic information systems (GIS) was used to construct measures of population density and street network connectivity within a 1-mile Euclidean buffer of each school address collected from 2020 U.S. Census Bureau five-year block group estimates and the Texas Department of Transportation Open Data Portal [32, 37]. Additionally, school characteristics from the 2018–2019 Texas Education Agency socio-demographics were used to describe the school sample, which included total enrollment and measures of the percentage of economically disadvantaged students (eligible for free or reduced lunch), racial and ethnic distribution as determined by the school district, percentage of students with limited English proficiency, and urbanicity [38].
使用中位数和四分位数范围计算每个学校的描述性特征。地理信息系统 （GIS） 用于构建每个学校地址 1 英里欧几里得缓冲区内的人口密度和街道网络连接的度量，这些测量是从 2020 年美国人口普查局五年区块组估计和德克萨斯州交通部开放数据门户 [ 32， 37] 收集的。此外，2018-2019 年德克萨斯州教育局社会人口统计数据中的学校特征用于描述学校样本，其中包括总入学人数和经济弱势学生百分比（有资格获得免费或减价午餐）的衡量标准、学区确定的种族和民族分布、英语水平有限的学生百分比和城市化 [ 38]。

Results 结果

The descriptive characteristics of the 36 schools in the sample are presented in Table 1. During the 2018–2019 school year, most of the students enrolled at the schools were Hispanic (median: 64%; IQR: 31%, 90% %), and most of the schools were classified as major urban schools (72%), which can be further explained by the population density (median: 647; IQR: 524, 694) and connectivity (median: 221; IQR: 169, 287) of streets within a 1-mile Euclidean buffer of each school.
表 1 列出了样本中 36 所学校的描述性特征。在 2018-2019 学年，这些学校的大多数入学学生是西班牙裔（中位数：64%;IQR：31%，90% %），大多数学校被归类为主要城市学校 （72%），这可以用人口密度进一步解释（中位数：647;IQR：524、694）和连通性（中位数：221;IQR：169、287）每所学校 1 英里欧几里得缓冲区内的街道。

Table 1 Descriptive characteristics of schools
表 1 学校的描述性特征
Full size table 全尺寸表格

Reliability results 可靠性结果

There were 30 subscales and the total MAPS-SRTS score that were tested for reliability in the initial MAPS-SRTS scoring framework using data from the 5 randomly selected schools from the sample. In initial reliability analyses, 22 out of 30 subscales had acceptable reliability, including five school access segment subscales, nine other segments near school subscales, seven crossing subscales, and the total MAPS-SRTS score. Eight out of the 30 subscales tested had ICC results that fell below the threshold of acceptable reliability in initial analyses, or the subscale lacked sufficient variability for reliability analyses. Five subscales from the school access segment section had reliability estimates that were below the threshold or lacked variability as denoted by an ICC of “n/a”: the positive buffer subscale (ICC = 0.27, 95% CI = -0.62, 0.89), the positive shade subscale (ICC = n/a), the positive bicycle infrastructure subscale (ICC = n/a), the overall positive school access segment subscale (ICC = 0.19, 95% CI = -0.67, 0.86), and the overall school access segment subscale (ICC = 0.57, 95% CI = -0.33, 0.94). Two subscales in the other segments near schools section were below acceptable reliability: the positive buffer subscale (ICC = -0.22, 95% CI = -0.84, 0.71) and the positive shade subscale (ICC = -0.13, 95% CI = -0.81, 0.76). The crossing section had one subscale with an initial reliability estimate below the acceptable threshold, which was the road width subscale (ICC = 0.25, 95% CI = -0.63, 0.88).
在初始 MAPS-SRTS 评分框架中，使用来自样本中随机选择的 5 所学校的数据，对 30 个分量表和 MAPS-SRTS 总分进行了可靠性测试。在初始可靠性分析中，30 个分量表中有 22 个具有可接受的可靠性，包括 5 个学校访问部分分量表、9 个学校分量表附近的其他部分、7 个交叉分量表和 MAPS-SRTS 总分。在测试的 30 个分量表中，有 8 个分量表的 ICC 结果低于初始分析中可接受的信度阈值，或者该分量表缺乏足够的信度分析变异性。学校访问部分的五个分量表的可靠性估计低于阈值或缺乏可变性，如 ICC 为“n/a”所表示的那样：正缓冲区分量表（ICC = 0.27,95% CI = -0.62,0.89），正阴影分量表（ICC = n/a），阳性自行车基础设施分量表（ICC = n/a），总体阳性学校访问部分分量表（ICC = 0.19， 95% CI = -0.67， 0.86） 和总体学校访问部分分量表 （ICC = 0.57， 95% CI = -0.33， 0.94）。学校附近部分的其他部分的两个分量表低于可接受的可靠性：阳性缓冲分量表 （ICC = -0.22， 95% CI = -0.84， 0.71） 和阳性阴影分量表 （ICC = -0.13， 95% CI = -0.81， 0.76）。交叉路段有一个分量表，其初始可靠性估计值低于可接受的阈值，即道路宽度分量表 （ICC = 0.25， 95% CI = -0.63， 0.88）。

Once initial reliability analyses were complete, inclusion of subscales in the final scoring was determined. Based on low reliability, there were four subscales removed from the MAPS-SRTS scoring schema: the positive buffer and positive shade subscales in the school access segment section and the positive buffer and positive shade subscales in the other segments near schools section. Once these subscales were removed, the overall school access segment and overall crossing subscales had acceptable reliability. The positive bicycle infrastructure subscale in the school access segment lacked sufficient variability for a reliability estimate. But the positive bicycle infrastructure subscale in the segment section, which includes the same items and scoring, was a reliable subscale (ICC = 0.89, 95% CI = 0.38, 0.99), and therefore this subscale was kept in the school access segment section. Additionally, the overall positive school access segment subscale had low reliability due to the lack of variability, but because the subscale consisting of the same items in the segment subscales showed acceptable reliability (ICC = 0.98, 95% CI = 0.90, 0.99), the subscale was retained in the school access segment section. There was one subscale (road width) that was revised due to initial low reliability results (ICC = 0.25, 95% CI = -0.63, 0.88). The initial road width subscale in the crossing section used two items (number of travel lanes and number of turn lanes) and trichotomized the sum of the two items. We recoded the subscale to be a dichotomous variable (≤ 4 lanes and > 4 total lanes) based on evidence that roads with more than 4 lanes increase the likelihood of pedestrian crashes [39]. With this revision, the subscale had higher reliability and was included in the negative crossing subscale (ICC = 0.58, 95% CI = -0.33, 0.95).
一旦初始可靠性分析完成，就确定了在最终评分中包括分量表。基于低可靠性，从 MAPS-SRTS 评分方案中删除了四个分量表：学校访问部分的正缓冲区和正阴影子量表以及学校附近其他部分的正缓冲区和正阴影子量表。一旦删除这些分量表，整个学校访问部分和整体交叉分量表具有可接受的可靠性。学校通道部分中的正自行车基础设施分量表缺乏足够的可变性来进行可靠性估计。但是，该分量表在部分（包括相同的项目和评分）是一个可靠的分量表（ICC = 0.89,95% CI = 0.38,0.99），因此该分量表保留在学校访问部分。此外，由于缺乏变异性，总体阳性学校访问部分分量表的可靠性较低，但由于由部分分量表中相同项目组成的分量表显示出可接受的可靠性（ICC = 0.98,95% CI = 0.90,0.99），该子量表保留在学校访问部分。由于初始结果可靠性低，有一个分量表（道路宽度）进行了修订 （ICC = 0.25,95% CI = -0.63,0.88）。交叉路段的初始道路宽度子量表使用两个项目（行车道数和转弯车道数），并将两个项目的总和进行三分法。我们将子量表重新编码为二分变量（≤ 4 车道，总共 >4 车道），因为有证据表明超过 4 车道的道路会增加行人碰撞的可能性 [ 39]。 通过此次修订，该分量表具有更高的可靠性，并被纳入负交叉分量表 （ICC = 0.58,95% CI = -0.33,0.95）。

After removal of four subscales, there were 26 final subscales that were included in the MAPS-SRTS instrument, in addition to the total MAPS-SRTS score, as shown in Table 2, which includes the final revised reliability measures for each subscale and the total score. These subscales were included in the final scoring schema, as shown in Fig. 2.
删除四个分量表后，除了 MAPS-SRTS 总分外，MAPS-SRTS 工具中还包括 26 个最终分量表，如表 2 所示，其中包括每个分量表的最终修订可靠性指标和总分。这些分量表包含在最终评分方案中，如图 2 所示。

Table 2 Micro-scale audit of pedestrian streetscapes– safe routes to school (MAPS-SRTS) subscale characteristics and final reliabilities
表 2 行人街景的微观尺度审计 – 安全上学路线 （MAPS-SRTS） 子量表特征和最终可靠性
Full size table 全尺寸表格

Discussion 讨论

This study developed the MAPS-SRTS audit tool for the assessment of micro-scale street-level features in school neighborhood environments and determined the reliability of the MAPS-SRTS tool within elementary school neighborhoods in central Texas. After revisions, results showed that out of the 30 original subscales, 26 subscales (86.7%) were retained in the final MAPS-SRTS tool. Once the four subscales with low reliability were removed from the MAPS-SRTS tool, the total MAPS-SRTS score had excellent reliability (ICC = 0.97). As a result, we suggest that the final version of the MAPS-SRTS tool that only retains subscales with high reliability and is presented within this study be utilized by researchers and practitioners aiming to assess how supportive the micro-scale environment surrounding schools is for active travel.
本研究开发了 MAPS-SRTS 审计工具，用于评估学校社区环境中的微尺度街道级特征，并确定了 MAPS-SRTS 工具在德克萨斯州中部小学社区内的可靠性。修订后，结果显示，在 30 个原始分量表中，26 个分量表 （86.7%） 保留在最终的 MAPS-SRTS 工具中。一旦从 MAPS-SRTS 工具中删除四个可靠性低的分量表，MAPS-SRTS 总分具有出色的可靠性 （ICC = 0.97）。因此，我们建议研究人员和从业者使用仅保留具有高可靠性的分量表并在本研究中介绍的 MAPS-SRTS 工具的最终版本，旨在评估学校周围的微观环境对主动旅行的支持程度。

The total reliability score of the MAPS-SRTS tool was comparable to two existing audit tools developed for school environments, which both had moderate to high inter-rater reliability [25, 27]. The ICC for the TCOPPE School Environmental Audit Tool and MAPS Global-SN were 0.602 and 0.97, respectively. While the TCOPPE Audit tool was developed within elementary school environments in Texas similar to the MAPS-SRTS tool, the items and scoring of the tool were not comparable to those used within the MAPS-SRTS tool [25]. In contrast, the MAPS Global-SN and the MAPS-SRTS were developed from the MAPS direct observation instruments [31, 40]. MAPS-SRTS and MAPS Global-SN both had high reliability scores for the positive streetscape subscale, the overall segment score, and the overall crossing score. However, the MAPS Global-SN was developed within secondary schools in New Zealand, and thus the modifications of the tool were made based on this context (e.g., round flashing orange lights, transportation facilities for adolescents) [27]. As the MAPS-SRTS tool was developed within an elementary school context in the U.S., both tools need to be adapted within more diverse geographic contexts.
MAPS-SRTS 工具的总可靠性得分与为学校环境开发的两种现有审计工具相当，它们都具有中到高的评估者间可靠性 [ 25， 27]。TCOPPE 学校环境审计工具和 MAPS Global-SN 的 ICC 分别为 0.602 和 0.97。虽然 TCOPPE 审计工具是在德克萨斯州的小学环境中开发的，类似于 MAPS-SRTS 工具，但该工具的项目和评分与 MAPS-SRTS 工具中使用的项目和评分没有可比性 [ 25]。相比之下，MAPS Global-SN 和 MAPS-SRTS 是从 MAPS 直接观测仪器发展而来的 [ 31， 40]。MAPS-SRTS 和 MAPS Global-SN 在积极街景分量表、总体路段得分和总体交叉得分方面均具有高可靠性得分。然而，MAPS Global-SN 是在新西兰的中学内开发的，因此该工具的修改是基于此背景进行的（例如，圆形闪烁的橙色灯、青少年的交通设施）[ 27]。由于 MAPS-SRTS 工具是在美国的小学环境中开发的，因此这两种工具都需要在更多样化的地理环境中进行调整。

Out of the 30 subscales within the MAPS-SRTS tool, we removed four subscales based on low reliability scores. For the school access segment and other segments near schools, the positive buffer subscale and shade subscale were removed completely for each section. The road width item was also revised because of low initial reliability, which was due to data collectors interpreting travel lanes versus turn lanes differently. The lower inter-rater reliability results for these subscales led to a recommendation that improved training protocols may be warranted to improve the reliability of these subscales. This training could include showing more visual image examples during the classroom training, practicing and quizzes with virtual street view examples, and ensuring high accuracy between test audit results of individual data collectors and lead data collectors during field training.
在 MAPS-SRTS 工具的 30 个分量表中，我们根据低可靠性分数删除了 4 个分量表。对于学校访问部分和学校附近的其他部分，每个部分的正缓冲区分量表和阴影分量表被完全删除。由于初始可靠性低，道路宽度项目也被修改，这是由于数据收集者对行驶车道和转弯车道的解释不同。这些分量表的评分者间信度结果较低，因此建议可能需要改进训练方案以提高这些分量表的可靠性。该培训可能包括在课堂培训期间展示更多的视觉图像示例，使用虚拟街景示例进行练习和测验，以及在现场培训期间确保单个数据收集者和主要数据收集者的测试审计结果之间的高度准确性。

There may be other opportunities to improve upon the measurement of the school neighborhood environment. Based on the findings, we have several recommendations for iterations of the MAPS-SRTS tool and within future assessment efforts. First, to improve the reliability of the MAPS-SRTS tool we present several considerations for the buffer and shade items. In the original MAPS-SRTS tool, a buffer was determined to be present if there was a sidewalk separated from a roadway by a parking lane, and this item was included within the positive subscale. Within the planning literature, there is an ongoing debate about the benefits or risks of on-street parking on the pedestrian environment, which depends on street-design and context [41, 42]. For example, there is evidence to suggest that on major streets, on-street parking is dangerous for pedestrian safety but on minor streets, on-street parking can slow traffic speeds and traffic volume [41, 43]. Transportation researchers have recommended the prohibition of on-street parking near schools, and thus a buffer created from on-street parking may need to be included as a negative subscale rather than a positive subscale in this context [43].
可能还有其他机会可以改进对学校社区环境的测量。根据调查结果，我们对 MAPS-SRTS 工具的迭代和未来的评估工作提出了几项建议。首先，为了提高 MAPS-SRTS 工具的可靠性，我们提出了缓冲区和阴影项目的几个注意事项。在最初的 MAPS-SRTS 工具中，如果存在由停车车道与道路隔开的人行道，则确定存在缓冲区，并且该项目包含在正子量表中。在规划文献中，关于路边停车对行人环境的好处或风险的争论一直存在，这取决于街道设计和环境 [ 41， 42]。例如，有证据表明，在主要街道上，路边停车对行人安全构成危险，但在次要街道上，路边停车会减慢交通速度和交通量 [ 41， 43]。交通研究人员建议禁止在学校附近停车，因此在这种情况下，可能需要将路边停车创建的缓冲区作为负分量表而不是正分量表 [ 43]。

Another construct for which measurement needs to be improved at the micro-scale is shade. The shade subscale, measured within the tool as the percentage of the length of the sidewalk/walkway covered by tree canopy or awnings, was also removed from this study due to a low reliability score. Auditors were asked to report a percentage of the sidewalk covered by shade (1–25%; 26–75%; 76–100%), which may have varied based on the time of day or season these data were collected. Future studies could improve the reliability of this measure by ensuring inter-rater assessments were completed during the same time period, and perhaps during times in which children actively commute to and from school. Audit-based measures of shade could be augmented with air temperature/relative humidity sensors, which can be used to develop thermal profiles to measure the environment children are walking and bicycling through [44]. As climate change is causing an overall rise in temperature and the intensity, duration, and frequency of heat waves [45], developing valid and reliable measures of environments that support safe physical activity of children in high temperatures is a pressing area for future research.
另一个需要在微观尺度上改进测量的结构是阴影。由于可靠性得分低，在工具中测量为树冠或遮阳篷覆盖的人行道/人行道长度的百分比的阴影分量表也从本研究中删除。审计员被要求报告人行道被阴影覆盖的百分比（1-25%;26-75%;76-100%），该百分比可能会根据收集这些数据的时间或季节而有所不同。未来的研究可以通过确保评分者间评估在同一时期完成，并且可能在儿童积极上下学期间完成，从而提高这一措施的可靠性。空气温度/相对湿度传感器可以增强基于审计的阴影测量，这些传感器可用于开发热曲线，以测量儿童行走和骑自行车的环境[44]。由于气候变化导致温度整体上升以及热浪的强度、持续时间和频率 [ 45]，开发有效且可靠的环境测量以支持儿童在高温下安全的身体活动是未来研究的紧迫领域。

A further recommendation when measuring the micro-scale characteristics of a school neighborhood is to consider the power dynamics and inclusion of lived experience when deciding whether to include negative neighborhood aesthetics as a measure in the audit tool. The MAPS-SRTS and other audit tools include an item that measures perceptions of signs of neglect or physical disorder in the neighborhood, such as graffiti, poorly maintained buildings, or abandoned buildings, that is used for a negative aesthetics subscale. While there is evidence of associations between neighborhood disorder and lower physical activity in children [46], it is also worth noting that associations of physical disorder with health outcomes may be misleading because they lack inclusion of relevant covariates such as socioeconomic status and collective efficacy [47]. According to the necessity- versus choice-based physical activity models framework, children from low and middle income countries or deprived settings may be operating out of necessity based active commuting because they have no other option, even in unsafe travel conditions [48]. Additionally, the use of this item has the potential for significant researcher prejudice and bias, as typically the academic-based researchers conducting audits are highly educated, which is often used as a proxy for socioeconomic status, so especially for studies in low-income areas, the researchers may have different lived experiences and norms/values than those living in the neighborhoods being audited. So, while the negative aesthetics subscale is a reliable measure in MAPS-SRTS, we recommend the removal of the negative aesthetics when the MAPS-SRTS tool is being conducted by people who are not a part of the community they are auditing, as community members’ familiarity with the community could cause them to be more or less sensitive to physical disorder or aesthetic features than researchers who do not reside in the community [49]. If this tool is being used in a community-engaged and driven research project, the negative aesthetics subscale can be included in consideration of the representation, inclusion, and lived experiences of community members and residents engaged in the efforts to collect the data, a key principle in community engagement [50].
在衡量学校社区的微观尺度特征时，进一步的建议是在决定是否将消极社区美学作为审计工具中的衡量标准时，考虑权力动态和生活经验的包容性。MAPS-SRTS 和其他审计工具包括一个项目，用于测量对附近被忽视或身体混乱迹象的看法，例如涂鸦、维护不善的建筑物或废弃的建筑物，用于负面美学分量表。虽然有证据表明邻里障碍与儿童身体活动量减少之间存在关联 [ 46]，但同样值得注意的是，身体障碍与健康结果的关联可能具有误导性，因为它们缺乏相关协变量，如社会经济地位和集体效能 [ 47]。根据基于必要性与选择的身体活动模型框架，来自中低收入国家或贫困地区的儿童可能出于必要性而积极通勤，因为他们别无选择，即使在不安全的旅行条件下也是如此 [ 48]。此外，使用此项目可能会对研究人员产生重大的偏见和偏见，因为通常进行审计的学术研究人员受过高等教育，这通常被用作社会经济地位的代表，因此特别是对于低收入地区的研究，研究人员可能具有不同的生活经历和规范/价值观与生活在被审计社区的人不同。 因此，虽然负面美学分量表是 MAPS-SRTS 中的可靠衡量标准，但我们建议当 MAPS-SRTS 工具由不属于他们所审计社区的人进行时，删除负面美学，因为社区成员对社区的熟悉程度可能导致他们对身体疾病或审美特征比不居住在社区的研究人员更或少敏感 [ 49]。如果该工具被用于社区参与和驱动的研究项目，则可以考虑到社区成员和居民参与数据收集工作的代表性、包容性和生活经历，包括负面美学分量表，这是社区参与的一个关键原则 [ 50]。

The MAPS-SRTS tool was developed to consider objective measures of the environment, which does not explain the experience, attitudes, and behaviors of those using and living in the environment. Future studies should seek to incorporate children’s and parents’ perspectives on the specific characteristics of the school travel environment, which could be informative for developing interventions and promoting active travel. Participatory mapping and qualitative GIS methods are two examples of participatory methods that could be used to engage children and parents [51, 52]. Wilson and colleagues (2019) used these methods to understand environmental barriers and facilitators of children’s active school travel, which included safety, material, and affective features [52].
MAPS-SRTS 工具的开发是为了考虑环境的客观测量，它不能解释使用和生活在环境中的人的经验、态度和行为。未来的研究应寻求纳入儿童和家长对学校旅行环境具体特征的看法，这可能为制定干预措施和促进积极旅行提供信息。参与式制图和定性 GIS 方法是可用于吸引儿童和家长的参与式方法的两个例子 [ 51， 52]。Wilson 及其同事 （2019） 使用这些方法来了解环境障碍和儿童积极学校旅行的促进因素，包括安全、物质和情感特征 [ 52]。

In addition to improvements in the MAPS-SRTS tool, there were several limitations specific to the design and methods of this study. The first was inclusion of a small sample (N = 36) of urban and suburban elementary school neighborhoods within one region of the U.S. Due to the low sample size of schools, there were few schools (n = 5) included within the reliability analysis, which may have resulted in lack of variability of some items for the school access segment. Future research should assess the MAPS-SRTS tools’ psychometric properties across a greater range of school districts, including suburban and rural geographic regions. Additionally, one promising avenue for future research could be to consider more time-efficient audit-based methods, such as virtually through Google Street View, which have been validated multiple times [53,54,55]. This could also allow for improving the external validity of the existing tool by allowing researchers to capture more diverse geographic contexts than one metropolitan area. For example, the application of the MAPS-SRTS tool to measure the micro-scale environment across rural schools could add to the limited evidence that has identified strategies to promote ACS in rural areas [56]. While data collection for the present study did not take place during peak periods of travel (i.e., drop off and pick-up from school), the items included within MAPS-SRTS consider the existing built environment, which would not be influenced by travel behaviors, though the supports for active travel can change during drop off and pick-up from school (e.g. crossing guards), so future research should consider differences in what is available in the built environment versus how and whether those supports are used. Lastly, we did not examine the validity of the MAPS-SRTS tool with active commuting to school or physical activity outcomes, which is an important next step for researchers in evaluation efforts of SRTS engineering projects. we are not assessing validity with this manuscript, and that is beyond scope of this paper. The MAPS instrument from which all items, scales, and sections were drawn has been extensively validated with this type of analysis in many population groups, including children [30, 57].
除了 MAPS-SRTS 工具的改进外，本研究的设计和方法还存在一些特定的限制。首先是纳入了美国一个地区内城市和郊区小学社区的小样本 （N = 36）。由于学校的样本量小，可靠性分析中包括的学校很少 （n = 5），这可能导致学校访问部分的某些项目缺乏可变性。未来的研究应该评估 MAPS-SRTS 工具在更大范围的学区（包括郊区和农村地理区域）的心理测量特性。此外，未来研究的一个有前途的途径可能是考虑更省时、基于审计的方法，例如通过虚拟的谷歌街景，这已经被多次验证了[53,54,55]。这还可以允许研究人员捕捉比一个大都市地区更多样化的地理环境，从而提高现有工具的外部有效性。例如，应用 MAPS-SRTS 工具来测量农村学校的微观环境可以增加已确定在农村地区推广 ACS 的策略的有限证据 [ 56]。虽然本研究的数据收集不是在旅行高峰期（即放学和接送）进行的，但 MAPS-SRTS 中包含的项目考虑了现有的建筑环境，这不会受到旅行行为的影响，尽管对主动旅行的支持在接送学校期间可能会发生变化（例如，过境警卫）， 因此，未来的研究应该考虑建筑环境中可用的资源与使用这些支撑物的方式和是否使用的差异。 最后，我们没有检查 MAPS-SRTS 工具与积极通勤上学或体育活动结果的有效性，这是研究人员评估 SRTS 工程项目工作的重要下一步。我们没有评估这份手稿的有效性，这超出了本文的范围。所有项目、量表和切片的 MAPS 工具已在许多人群（包括儿童）中通过这种类型的分析得到了广泛验证 [ 30， 57]。

The MAPS-SRTS was developed with practitioner usage in mind for implementing and evaluating Safe Routes to School interventions. Previous MAPS audits are not suitable for school-specific measurement of the streetscape or require specialized GIS skills to determine a buffer to audit [27, 31]. The MAPS-SRTS audit tool does not require specialized skills to determine the school travel neighborhood environment, and the collection of these data can be feasible (77.8 min on average per school) for researchers and practitioners. There is a need for reliable tools to use for local implementation of SRTS programs, highlighted by the fact that evaluation is one of the least-implemented components of the SRTS Six E’s (engagement, equity, engineering, encouragement, education, and evaluation), and the MAPS-SRTS can be used to both assess the need for specific interventions and document infrastructure changes around schools that are part of SRTS programs [58].
MAPS-SRTS 的开发考虑了从业者的使用，用于实施和评估安全上学路线干预措施。以前的 MAPS 审计不适合针对学校的街景测量，也不需要专门的 GIS 技能来确定审计缓冲区 [ 27， 31]。MAPS-SRTS 审计工具不需要专业技能来确定学校旅行社区环境，并且这些数据的收集对于研究人员和从业者来说是可行的（每所学校平均 77.8 分钟）。需要可靠的工具来在本地实施 SRTS 计划，这一事实突出表明，评估是 SRTS 六个 E（参与、公平、工程、鼓励、教育和评估）中实施最少的组成部分之一，MAPS-SRTS 可用于评估特定干预措施的需求，并记录作为 SRTS 计划一部分的学校周围的基础设施变化 [ 58]。

Conclusions 结论

The MAPS-SRTS audit tool is a reliable instrument for measuring the school travel environment that can be used for a variety of research and evaluation purposes. This tool’s potential importance, once validated, includes being used to evaluate SRTS and tactical urbanistic interventions and to inform policy or advocacy efforts by diagnosing the current condition of school access routes throughout an entire district or area of interest, and identify priority schools for investments in improvements. Additionally, this tool can be used to test the moderating effect of micro-scale factors near schools, on inter-personal/community-based interventions to motivate more children and families to actively commute to school. Finally, this tool can be used to compare and contrast the level of contribution of specific micro-scale factors on active commuting to school behaviors. While there is still further research needed to validate the tool with child ACS behavior and improve the quality of assessment for several constructs, this tool has the potential for use by researchers and practitioners to document and identify areas for intervention around schools to improve safety and active travel for children.
MAPS-SRTS 审计工具是测量学校旅行环境的可靠工具，可用于各种研究和评估目的。该工具的潜在重要性，一旦得到验证，包括用于评估 SRTS 和战术城市干预，并通过诊断整个地区或感兴趣地区的学校入学路线的现状来为政策或宣传工作提供信息，并确定优先学校进行投资改进。此外，该工具可用于测试学校附近微尺度因素对人际/基于社区的干预措施的调节作用，以激励更多儿童和家庭积极通勤上学。最后，该工具可用于比较和对比特定微尺度因素对积极通勤到学校行为的贡献水平。虽然仍需要进一步的研究来验证该工具与儿童 ACS 行为并提高多种结构的评估质量，但研究人员和从业者有可能使用该工具来记录和确定学校周围的干预区域，以改善儿童的安全和积极出行。