Introduction

The ways in which neighborhood characteristics influence health outcomes, daily functioning, and quality of life have drawn attention from fields of public health research and policy. Neighborhood characteristics have been associated with a wide range of health outcomes, including behavioral health (Hosokawa and Katsura 2020), sleep quality (Johnson et al. 2016; Troxel et al. 2020), preterm births (Giurgescu et al. 2017), obesity (Tamayo et al. 2016; Hoenink et al. 2019), tobacco dependence (Reitzel et al. 2012), and limited life spans (General OotS 2021). Given these associations, it is unsurprising that the US Department of Health and Human Services included “neighborhood and built environment” as one of the five defined domains of social determinants of health (“Social Determinants of Health” n.d.), underscoring the role of neighborhoods in health outcomes.

Tools that capture residential measurements are important for engaging residents and stakeholders while assessing the effectiveness of community development initiatives. The history of community development and renewal in the United States can be traced back to reform movements during the Progressive Era (Hoffman 2012) and continues into today. Initially, such efforts failed to include community input, but that has slowly changed. Collaborative partnerships with community groups create multifaceted teams guided by both bottom-up and top-down approaches. These partnerships have taken place in urban, suburban, and rural communities, in the United States and abroad (Savage et al. 2018; Freudenberg and Golub 1987; Liss-Levinson et al. 2020). A review of the strategies utilized by these partnerships yields mixed results in their ability to improve population health outcomes, community behaviors, public policy, or community environments (Roussos and Fawcett 2000). Despite supportive evidence of community change prompted by these partnerships, the degree of change has been called into question due to a lack of clear standards for measuring neighborhood quality of life or change (Roussos and Fawcett 2000).

Efforts have been made to identify and define neighborhood measurements while differentiating temporary from permanent features (Ndjila et al. 2019) to better determine how neighborhoods may affect resident health and the effectiveness of interventions aimed at improving these neighborhoods. Neighborhood measurement methods include in-person evaluation, administrative data (e.g., police records), census data, and review of video recordings (Sampson and Raudenbush 2004). Other examples include the presence of alcohol and tobacco stores, commercial building security, crime data, and neighborhood walk scores (Sampson and Raudenbush 2004; Diez Roux et al. 2017). However, measurements such as resident perceptions of neighborhood characteristics hold predictive power that administrative or built environment measurements cannot capture. For example, perceived neighborhood characteristics have been reported as more accurately predicting physical activity than built environment assessments (Orstad et al. 2017).

As the relationship between neighborhood characteristics and health becomes clearer, it is increasingly important that collaborative partnerships and healthcare organizations are equipped with standardized, comprehensive tools for assessing changes in community development. This is especially true when large investments are made from multiple sectors with expectations for demonstrated improvement.

Resident-reported measures of neighborhood characteristics come from two primary sources. First, academic publications include a variety of surveys employed, especially from criminology and sociology literature. Naturally, these publications include an assessment of numerous characteristics of neighborhoods that are associated with violence and crime. Separate from the published literature are evaluation tools designed for neighborhoods and neighborhood interventions by third party organizations and governmental groups engaged in community development. These latter sources have produced neighborhood measurement tools to assist community development programs and published them in what is called “gray literature” (19-27 Stiefel et al. 2020; “User’s Guide for the Fragile Families and Child Wellbeing Study Public Data” 2018, Mulcahy et al. 2020; Mulcahy 2017; “The Healthy Housing Outcomes Survey” 2019; “National Survey of Children’s Health Questionnaire” 2020; “CHIS Questionnaires” 2021; “Secondary Student Community Health In-School and Hybrid Only Module” 2020; “2021 State and Local Youth Risk Behavior Survey” 2021). This review aimed to assess the state of both academic and gray literature, while identifying resident-reported measures of neighborhoods currently being utilized and reported in the domestic and international literature to consider domains of coverage, populations included in the assessments, and characterizations of neighborhoods.

Methods

Search strategy and criteria selection

A scoping literature review was conducted on resident self-reported measurements of neighborhoods to identify commonly assessed neighborhood characteristics and the items chosen to represent them in the academic and gray literature, the latter of which is defined as publications outside of traditional academic scholarships, including reports, keynote addresses, briefs, and similar. A PubMed search of the Major MeSH term “Self-Report” AND “neighborhood disorder” yielded 28 results in the academic literature. A second PubMed search was conducted with the Boolean search terms “neighborhood disorder,” “neighborhood safety,” “neighborhood poverty,” “neighborhood cohesion,” and “self-report” and yielded 408 results. Filters applied in both searches isolated full-text English language articles published between 2010 and 2021. The results of these two searches underwent a title and abstract screening, in which any paper that indicated having self-reported neighborhood conditions as a variable was included for a full manuscript review. Throughout the screening, reference sections of the articles became a source for finding additional publications to increase the breadth and depth of the findings. Once the review reached a point at which the researchers observed that trends in neighborhood measures were apparent, and questionnaire items ceased to yield new information or additional insights (suggesting that a point of saturation had been reached), the review was concluded. At the conclusion of the screening, 27 published articles were included in the review (Table 1).

Table 1 Published articles including resident self-reported surveys
Full size table

An advanced search using Google Scholar was conducted to identify gray literature sources for neighborhood measurements. This search was conducted using the Boolean terms “self-report,” “self-assessed,” “self-rated,” “self-estimated,” to search for papers that contained “neighborhood” in relation to health, safety, community, poverty, or disorder. Gray literature meeting inclusion criteria comprised surveys published in English since 2010 and those items evaluating resident self-reported measurement of one or more neighborhood characteristics through a primary source. While reviewing gray literature data sets, surveys, and websites, additional resources surfaced that were then evaluated for inclusion. The review of gray literature sources continued until the neighborhood domains assessed and their qualifying items became redundant. At the conclusion of the gray literature review, surveys developed by eight different organizations were included (Table 2).

Table 2 Gray literature organization self-reported surveys
Full size table

Data extraction and organization

Information regarding self-reported neighborhood measures in published studies was extracted and compiled to identify trends in measured domains and items (Table 1). These items included the country and setting in which the study took place; method of self-reported data collection; characteristics these measurements assessed; and the validated scales used or modified in the resident survey. Although the majority of surveys were implemented in respondent residential areas, Kim et al. (2019) and Giurgescu et al. (2017) administered surveys at urban hospitals; therefore, gathering information potentially representative of many different neighborhoods simultaneously. These studies were categorized as urban, despite the possibility that a portion of the participants resided in rural and suburban areas.

Self-administered methods of data collection included online or paper surveys completed by respondents. Paper surveys were mailed out or dispersed at local centers. Interviews were conducted in-person, over the phone, or online. During review, if a survey item did not name the domain it was intended to capture, the item was assigned to a descriptive term commonly seen in the literature. After reviewing survey items within each category individually, recategorization and relabeling occurred when necessary to allow for comparison across all collected literature.

Details regarding the gray literature survey respondents and questionnaires were extracted and organized for comparison purposes (Table 2). The domain of neighborhood characteristics assessed in each gray literature source was recorded. Nomenclature from the source itself was reported, when possible, to define the measured neighborhood qualities. When necessary, the authors applied appropriate terminology used across the literature to review self-report questionnaire items that did not explicitly identify their measurement.

Categorization methodology and rationale

After extracting all items and questions available in the selected sources, categories of the most common domains measured were developed from terminology available in the literature (Table 3). These domains were further stratified into subcategories. For example, items comprising the domain “physical conditions” were organized and counted as either “physical order” or “physical disorder.” If items and survey questions described the same characteristic but under a different name, they were organized into the same subcategory and domain pairing. If a literature source provided questions without explicitly stating the aspect of neighborhoods they intended to measure, appropriate descriptive terms seen in other literature were assigned to them in the “characteristics assessed” column.

Table 3 Neighborhood characteristic domains
Full size table

The several sources that noted what domains were measured, but did not directly provide all survey items, were still added to the “instances” count. Depending on the survey phrasing, the same item could be arguably used to support different neighborhood domains. For example, items concerning violence could be categorized as reflecting violence if asking about its presence as a neighborhood problem (Troxel et al. 2020), safety if asking about its absence (Palumbo et al. 2019), crime if inquiring after the frequency of criminal occurrences (Tamayo et al. 2016), fear of crime if reporting residential fear of being attacked, raped, or robbed (Giurgescu et al. 2017; Prado et al. 2017; Loh et al. 2019), or more broadly social disorder (Ndjila et al. 2019). Each item was evaluated to determine which aspects were most specifically addressed to ensure proper item placement. For concision, redundant or differently phrased items were grouped together or omitted from tables. For example, “attractive sites” and “interesting things to look at” both can be combined into “interesting and attractive things to look at” to qualify the term “neighborhood aesthetics. Overall, a comparison between academic and gray sources indicated that gray literature includes more survey items for diverse populations, such as elderly and disabled communities, and more positive attributes of neighborhood conditions.

Results

Following the search and screening process, 28 papers and eight gray literature sources were included in this review. Survey items from 22 published papers and one gray literature source were derived from previously reported or validated scales (Appendix Table 4). Surveys reported in the collected literature could be grouped into one of the following five domains: physical conditions, social conditions, community resources, social agency, and neighborhood dynamics. Domain items suggesting laudatory or positive connotations were placed in a separate subcategory from items capturing a similar neighborhood characteristic with negative connotations.

Of the 29 instances of physical condition measures, 18 measured aspects of physical disorder, while 11 measured aspects of physical order. Social conditions were evaluated by the number of studies asking questions regarding safety (20), social cohesion (16), social support (16), social trust (13), crime (13), violence (12), social disorder (12), engagement with community activities (6), fear (5), social disconnectedness (4), and social order (4). There were 21 instances of community resources measurements that were observed and broken down further into subcategories of amenities (12) and community services (10) available in the neighborhood. Social agency was measured most often at an organizational level as civic engagement (6) or more interpersonally as social control (5). Neighborhood dynamics were assessed by measures of community stability (7) and community development (3). Examples of items used to qualify subcategories are listed in Table 3 to demonstrate the range observed in the literature.

Differences between literature published in the United States and internationally were evident in both the populations of focus and the framing of neighborhood characteristics. Additionally, gray literature sources and academic literature differed in terms of target populations and content of survey tools. These differences suggest implications for practice and require further consideration to improve the measurement of neighborhoods.

Discussion

This study examined articles and gray literature that used resident-reported measures to understand neighborhood characteristics. Although international self-reports’ survey items distinguished them from domestic ones, these differences may be explained by demographic, economic, and cultural differences. To date, this appears to be the first review of resident-reported neighborhood measures inclusive of surveys from the gray literature.

Characteristics of neighborhood self-report toolkits in international literature

Academic studies conducted outside of the United States were closer to gray literature publications in two ways. First, they often included more information on diverse age groups, such as the elderly. Second, they often assessed neighborhood characteristics with a more positive connotation. Investigating the neighborhood friendliness suitability for the elderly sets international publications apart from American articles. Two international studies conducted in low- and middle-income countries measured neighborhood friendliness toward elderly and disabled communities (Shagdarsuren et al. 2017) or focused on adults over the age of 65 (Hsueh et al. 2016). Of all the US sources, Success Measures, a source from the gray literature, was the only one to measure neighborhood characteristics specifically for the elderly (Mulcahy et al. 2020; Mulcahy 2017). Older adults may be the most affected by disparate neighborhood characteristics as their mobility declines and dependence on their social network increases. Negative neighborhood characteristics have been associated with hastened cognitive decline and functional limitations (“Today’s Research on Aging” 2017).

Survey items that were most often characterized in negative terms in the US literature were often presented positively in the international literature. For example, Hosokawa and Katsura (2020) developed Likert scale survey items regarding physical conditions and safety to reflect positive aspects of the neighborhood. This is in contrast to many US studies that orient questions to be about the presence of disorder or crime. Shagdarsuren et al. presented positive statements of physical order and safety featuring both manmade and environmental characteristics, including the quality and cleanliness of the soil, air, river water, drinking water, and nature (Shagdarsuren et al. 2017).

Characteristics of neighborhood self-report toolkits in gray and academic literature

Patterns in survey content, format, and target population became apparent when analyzing survey items reported in gray and peer-reviewed literature. The peer-reviewed literature placed a greater emphasis on certain measurements, most notably physical disorder (14/18), physical order (9/11), social disorder (10/12), fear of crime (4/4), crime (11/13), and violence (8/12). Meanwhile, gray literature placed a greater emphasis on civic engagement (4/6) and community stability (5/7). While the majority of published surveys contained categorical scales, some gray literature included free response portions to better solicit resident perspectives. Additionally, gray literature sources were more child- and family-oriented in their target audience and survey content. Some surveys captured child neighborhood perceptions with direct survey administration (“CHIS Questionnaires” 2021; “Secondary Student Community Health In-School” 2020; “2021 State and Local Youth Risk Behavior Survey” 2021)), while others targeted parents to respond on behalf of their child (“National Survey of Children’s Health Questionnaire” 2020; “CHIS Questionnaires” 2021).

Gray literature measured similar neighborhood domains to those in the academic literature, contributing to a greater number of overall items measured. Although only two gray literature sources contributed to the six instances of measurement of neighborhood engagement with community activities, they accounted for 13 of the 18 measurement items (Mulcahy et al. 2020;“National Survey of Children’s Health Questionnaire” 2020). Four of the six sources measuring civic engagement were from gray literature (Stiefel et al. 2019; Mulcahy et al. 2020; Mulcahy 2017; “CHIS Questionnaires” 2021; “Secondary Student Community Health In-School” 2020). Notably, only one gray literature collection (Mulcahy et al. 2020; Mulcahy 2017) inquired about disability accessibility features and another about eco-friendly outdoor features (Mulcahy et al. 2020).

Terminological variability in residential self-reports

Although both academic and gray literature include similar aspects of resident reports on neighborhoods, there is little consistency in the language used to describe these aspects across the literature. Even when there is agreement in the terminology for the domain and subcategory being measured, the illustrating items vary greatly. In cases where the items substantively agree, varying specificity or cultural differences led to a breadth of survey items. For example, although some studies listed “trash” to support physical disorder (Hosokawa and Katsura 2020; Johnson et al. 2016; Troxel et al. 2020; Palumbo et al. 2019), others listed “litter,” (Johnson et al. 2016; ;“National Survey of Children’s Health Questionnaire” 2020; Matthews et al. 2019) “garbage,” (“User’s Guide for the Fragile Families and Child Wellbeing” 2018) “rubbish” (Matthews et al. 2019) or specific forms of trash such as “broken bottles” in addition to “trash” or in place of it.

Our results are similar to those previously reported by Ndjila et al. who found linguistic variability and diversity in the peer-reviewed literature’s survey items supporting synonyms of physical and social conditions (Ndjila et al. 2019). Not only are many synonyms used, but the same items are interpreted as evidence for different neighborhood characteristics. While Palumbo et al. (2019) categorized “using/selling drugs” and “drug dealers have the most money” as physical disorder, other studies cited the use and selling of drugs as an example of social disorder (Giurgescu et al. 2017; Pruitt et al. 2012; Hastings and Snowden 2019) or more generally as a neighborhood problem. (“Secondary Student Community Health In-School and Hybrid Only Module” 2020). It is of course unsurprising that a qualitative domain such as neighborhood perceptions, which has arisen organically over time, would lack conceptual clarity. Bringing clarity to the terms of this reporting is an important task ahead.

Limitations

This review has several limitations. As discussed previously, the dissimilarity of items and language in literature created difficulty in evaluating and organizing characteristics and items (Table 3). Nonetheless, though this is a limitation, it might also be considered a finding given this study’s parameters and aims. Although a consistent strategy was applied in sorting items, it relied heavily on interpretation. Many measures used Likert scales, which may lead to an underestimation of attention paid to positive or negative aspects of neighborhoods. International studies translated either questionnaires or study findings, which may imprecisely convey the characteristics that the researchers in those settings attempted to measure.

Conclusions & recommendations

Items in resident self-report surveys were organized into the following domains: physical conditions, social conditions, community resources, social agency, and neighborhood dynamics. Within these domains, there is great variability in terminology and items used to evaluate neighborhoods. This review highlights the need for future research to develop a systematic approach for characterizing and categorizing resident neighborhood perceptions. Few studies define geographical boundaries or inquired as to how long respondents had resided in their neighborhoods, limiting the strength and reliability of gathered information to inform community development practices.

There remains a need to generate items that intentionally include currently overlooked and vulnerable residents in communities, such as older adults and those with disabilities. In addition, clear definitions, standardized scales, and the consolidation of items representing the domains presented in this review would lead to better appraisal of community development efforts, more generalizable results, and easier comparisons of related work.