Methods for Evaluating the Physical and Social Environment of Urban Neighborhoods

描述(申请人提供)：这项研究的目的是设计和评估一种全面、系统的方法，通过直接观察社区特征来描述城市社区的特征。公共卫生研究人员和从业者已开始将重点重新放在作为健康的重要决定因素的个人生活环境上。社区特征通常通过参与者自我报告、观察(通常通过视频)或二级数据库(如美国人口普查)来衡量。这些方法的主要局限性包括无法监测随时间推移的变化、成本和大量的人力资源。这限制了他们在不止几项研究中的纳入，并导致在研究期间对社区进行单点评估。然而，在一项多年的研究中，城市邻里环境可能会因为季节性、士绅化、移民等因素而发生显著变化。我们建议评估和修订研究团队已经开发的简明量表，旨在通过直接观察来测量城市社区的社会、物理和资源环境，并检查该量表的信度和效度。为了实现这些目标，我们将在一项更大的多层次家长研究中开展工作，旨在研究纽约市36个社区的城市环境特征与精神健康、药物使用和艾滋病毒风险之间的联系，这些社区已经被人种学定义。我们将首先在纽约市的富裕和/或以白人为主的社区中定义另外15个社区，因为我们目前的抽样框架被经济贫困和少数族裔居高不下的社区所代表。接下来，我们将使用我们的邻里评价表和邻里地图来衡量城市环境。51个目标社区中的每个社区将由两名独立观察员每年两次评级，为期2年，提供4次迭代，用于数据收集、分析和量表修订。信度将通过评分员间和测试重测信度和内部一致性的测量进行评估。有问题的项目将在下一次迭代中被删除或返工。该量表还将针对每种类型的效度进行评估。面子效度将通过研究调查人员、顾问和工作人员的审查进行评估。内容效度将通过培训前和培训后的内部一致性和项目值评估来评估。因素分析将被用来探索因素的存在和性质。标准相关效度将通过录像进行评估，并将区块的一个小样本评定为量表最终迭代的“黄金标准”。将执行外部结构有效性检查，将每个社区的量表结果与美国人口普查的关键指标(例如，居住隔离、收入不平等)进行比较。在第二年，我们还将在底特律随机选择的50个社区管理该量表，以评估该量表在另一个城市环境中的表现，并评估其普适性。公共卫生相关性：公共卫生研究人员和从业者已开始重新关注作为健康的重要决定因素的个体生活的背景。社区特征通常通过参与者进行衡量 自我报告、观察(通常通过视频)或二级数据库(如美国人口普查)监控随时间变化的能力有限，成本高昂，需要大量人力资源。这项研究的目的是设计和评估一种全面、系统的方法，通过直接观察社区特征来描述城市社区的特征，以解决这些局限性。

DESCRIPTION (provided by applicant): The purpose of this study is to design and evaluate a comprehensive, systematic approach to characterizing urban neighborhoods through direct observation of neighborhood characteristics. Public health researchers and practitioners have begun to refocus on the context in which an individual lives as an important determinant of health. Neighborhood characteristics are typically measured through participant self-report, observation (often via video), or secondary databases like the US Census. Major limitations of these approaches include the inability to monitor changes over time, costs and significant staff resources. This has limited their inclusion in more than a few studies and resulted in neighborhoods being evaluated at a single point during a study. Yet, during a multi-year study, the urban neighborhood context may significantly change as a result of seasonality, gentrification, immigration, etc. We propose to evaluate and revise a brief scale, already developed by the research team, aimed at measuring the social, physical and resource environment of urban neighborhoods through direct observation and examine the reliability and validity of the scale. To achieve these aims, we will be working within a larger multi-level parent study aimed at examining the associations between features of the urban environment and mental health, drug use, and HIV risk in 36 NYC neighborhoods that have already been ethnographically defined. We will first define an additional 15 neighborhoods in affluent and/or predominantly white neighborhoods in New York City, as our current sampling frame is overrepresented by neighborhoods with a high prevalence of economic deprivation and minorities. Next, we will use our neighborhood evaluation scale and neighborhood mapping to measure the urban environment. Each of the 51 target neighborhoods will be rated by two independent observers twice a year for 2 years, providing 4 iterations for data collection, analysis and scale revision. Reliability will be assessed via measures of inter-rater and test retest reliability and internal consistency. Problematic items will be eliminated or reworked in the next iteration. The scale will also be assessed for each type of validity. Face validity will be assessed through review by study investigators, consultants and staff. Content validity will be evaluated through internal consistency and evaluation of item values pre- and post-training. Factor analysis will be used to explore the presence and nature of factors. Criterion-related validity will be assessed by video recording and rating a sub-sample of blocks as a "gold standard" for the final iteration of the scale. External construct validity checks comparing the scale results for each neighborhood to key indicators from the US Census (e.g., residential segregation, income inequality) will be performed. In year 2, we will also administer the scale in 50 randomly selected Detroit neighborhoods to evaluate the scale's performance in another urban setting and assess the generalizability. PUBLIC HEALTH RELEVANCE: Public health researchers and practitioners have begun to refocus on the context in which as individual lives as an important determinant of health. Neighborhood characteristics are typically measured through participant self-report, observation (often via video), or secondary databases like the US Census which are limited in their ability to monitor changes over time, are expensive and require significant staffing resources. The purpose of this study is to design and evaluate a comprehensive, systematic approach to characterizing urban neighborhoods through direct observation of neighborhood characteristics that will address these limitations.