Beyond Activity Space: Detecting Communities in Ecological Networks

Place-Based Developmental Research: Conceptual and Methodological Advances in Studying Youth Development in Context

Scientists have, for some time, recognized that development unfolds in numerous settings, including families, schools, neighborhoods, and organized and unorganized activity settings. Since the turn of the 20th century, the body of mainstream neighborhood effects scholarship draws heavily from the early 20th century Chicago School of Sociology frameworks and have been situating development in neighborhood contexts and working to identify the structures and processes via which neighborhoods matter for a range of developmental outcomes, especially achievement, behavioral and emotional problems, and sexual activity. From this body of work, two new areas of developmental scholarship are emerging. Both areas are promising for advancing an understanding of child development in context. First, cultural-developmental neighborhood researchers are advancing neighborhood effects research that explicitly recognizes the ways that racial, ethnic, cultural, and immigrant social positions matter for neighborhood environments and for youths' developmental demands, affordances, experiences, and competencies. This body of work substantially expands the range of developmental outcomes examined in neighborhood effects scholarship to recognize normative physical, emotional, cognitive, behavioral, social, and cultural competencies that have largely been overlooked in neighborhood effects scholarship that espoused a more color-blind developmental approach. Second, activity space neighborhood researchers are recognizing that residential neighborhoods have important implications for broader activity spaces-or the set of locations and settings to which youth are regularly exposed, including, for example, schools, work, organized activities, and hang-outs. They are using newer technologies and geographic frameworks to assess exposure to residential neighborhood and extra-neighborhood environments. These perspectives recognize that time (i.e., from microtime to mesotime) and place are critically bound and that exposures can be operationalized at numerous levels of the ecological system (i.e., from microsystems to macrosystems). These frameworks address important limitations of prior development in context scholarship by addressing selection and exposure. Addressing selection involves recognizing that families have some degree of choice when selecting into settings and variables that predict families' choices (e.g., income) also predict development. Considering exposure involves recognizing that different participants or residents experience different amounts of shared and nonshared exposures, resulting in both under-and over-estimation of contextual effects. Activity space scholars incorporate exposure to the residential neighborhood environments, but also to other locations and settings to which youth are regularly exposed, like schools, after-school settings, work, and hang-outs. Unfortunately, the cultural-development and activity space streams, which have both emerged from early 20th century work on neighborhood effects on development, have been advancing largely independently. Thus, the overarching aim of this monograph is to integrate scholarship on residential neighborhoods, cultural development, and activity spaces to advance a framework that can support a better understanding of development in context for diverse groups. In Chapters I and II we present the historical context of the three streams of theoretical, conceptual, and methodological research. We also advance a comprehensive cultural-developmental activity space framework for studying development in context among children, youth, and families that are ethnically, racially, and culturally heterogeneous. This framework actively recognized diversity in ethnic, racial, immigrant, and socioeconomic social positions. In Chapters III-V we advance specific features of the framework, focusing on: (1) the different levels of nested and nonnested ecological systems that can be captured and operationalized with activity space methods, (2) the different dimensions of time and exposures or experiences that can be captured and operationalized by activity space methods, and (3) the importance of settings structures and social processes for identifying underlying mechanisms of contextual effects on development. Structures are setting features related to the composition and spatial arrangement of people and institutions (e.g., socioeconomic disadvantage, ethnic/racial compositions). Social processes represent the collective social dynamics that take place in settings, like social interactions, group activities, experiences with local institutions, mechanisms of social control, or shared beliefs. In Chapter VI, we highlight a range of methodological and empirical exemplars from the United States that are informed by our comprehensive cultural-developmental activity space framework. These exemplars feature both quantitative and qualitative methods, including method mixing. These exemplars feature both quantitative and qualitative methods, including method mixing. The exemplars also highlight the application of the framework across four different samples from populations that vary in terms of race, ethnicity, gender, age, socioeconomic status (SES), geographic region, and urbanicity. They capture activity space characteristics and features in a variety of ways, in addition to incorporating family shared and nonshared activity space exposures. Finally, in Chapter VII we summarize the contributions of the framework for advancing a more comprehensive science of development in context, one that better realizes major developmental theories emphasizing persons, processes, contexts, and time. Additionally, we offer a place-based, culturally informed developmental research agenda to meet the needs of an increasingly diverse population.

Identifying latent activity behaviors and lifestyles using mobility data to describe urban dynamics

Urbanization and its problems require an in-depth and comprehensive understanding of urban dynamics, especially the complex and diversified lifestyles in modern cities. Digitally acquired data can accurately capture complex human activity, but it lacks the interpretability of demographic data. In this paper, we study a privacy-enhanced dataset of the mobility visitation patterns of 1.2 million people to 1.1 million places in 11 metro areas in the U.S. to detect the latent mobility behaviors and lifestyles in the largest American cities. Despite the considerable complexity of mobility visitations, we found that lifestyles can be automatically decomposed into only 12 latent interpretable activity behaviors on how people combine shopping, eating, working, or using their free time. Rather than describing individuals with a single lifestyle, we find that city dwellers' behavior is a mixture of those behaviors. Those detected latent activity behaviors are equally present across cities and cannot be fully explained by main demographic features. Finally, we find those latent behaviors are associated with dynamics like experienced income segregation, transportation, or healthy behaviors in cities, even after controlling for demographic features. Our results signal the importance of complementing traditional census data with activity behaviors to understand urban dynamics.

Supplementary Information

The online version contains supplementary material available at 10.1140/epjds/s13688-023-00390-w.

Geographic Isolation, Compelled Mobility, and Everyday Exposure to Neighborhood Racial Composition Among Urban Youth. AJS; AMERICAN JOURNAL OF SOCIOLOGY 2022;128:914-961. [PMID: 38213504 PMCID: PMC10783811 DOI: 10.1086/721666]

Foundational urban social theories view heterogeneity of exposure to spatial and social contexts as essential aspects of the urban experience. In contrast, contemporary neighborhood research emphasizes the isolation of city dwellers - particularly residents of racially segregated neighborhoods. Using geospatial data on a sample of youth from the 2014-16 Columbus, OH-based Adolescent Health and Development in Context study, we explore the extent to which the neighborhood locations of everyday activities vary with respect to residential racial composition. In the context of segregated US metro areas, the geographic isolation approach expects home census tract racial composition to powerfully shape the racial composition of activity location neighborhoods. In this view, Black youth residing in high proportion Black neighborhoods are expected to spend the vast majority of their time exposed to similarly Black-concentrated neighborhoods. Consistent with an alternative compelled mobility approach, we find that Black youth residing in high proportion Black neighborhoods exhibit among the highest levels of heterogeneity in the racial composition of neighborhoods encountered. Moreover, Black youth residing in high proportion Black neighborhoods are expected to spend 39% of their non-home time (roughly 2.5 hours a day) in low proportion Black neighborhoods compared to 23% (1.5 hours) in high proportion Black neighborhoods. Exposures to low proportion Black neighborhoods among these youth are largely driven by organizationally-based resource seeking. These findings call into question the assumption that residence in Black segregated neighborhoods leads to homogeneously Black segregated neighborhood exposures and encourage theoretical development and data collection strategies that acknowledge the potential for significant heterogeneity in the everyday neighborhood experiences of urban youth.

Leveraging Network Representation Learning and Community Detection for Analyzing the Activity Profiles of Adolescents

Human mobility analysis plays a crucial role in urban analysis, city planning, epidemic modeling, and even understanding neighborhood effects on individuals' health. Often, these studies model human mobility in the form of co-location networks. We have recently seen the tremendous success of network representation learning models on several machine learning tasks on graphs. To the best of our knowledge, limited attention has been paid to identifying communities using network representation learning methods specifically for co-location networks. We attempt to address this problem and study user mobility behavior through the communities identified with latent node representations. Specifically, we select several diverse network representation learning models to identify communities from a real-world co-location network. We include both general-purpose representation models that make no assumptions on network modality as well as approaches designed specifically for human mobility analysis. We evaluate these different methods on data collected in the Adolescent Health and Development in Context (AHDC) study. Our experimental analysis reveals that a recently proposed method (LocationTrails) offers a competitive advantage over other methods with respect to its ability to represent and reflect community assignment that is consistent with extant findings regarding neighborhood racial and socio-economic differences in mobility patterns. We also compare the learned activity profiles of individuals by factoring in their residential neighborhoods. Our analysis reveals a significant contrast in the activity profiles of individuals residing in white-dominated vs. black-dominated neighborhoods and advantaged vs. disadvantaged neighborhoods in a major metropolitan city of United States. We provide a clear rationale for this contrastive pattern through insights from the sociological literature.

Neighborhoods, Activity Spaces, and the Span of Adolescent Exposures

Since the inception of urban sociology, the "neighborhood" has served as the dominant context thought to capture developmentally significant youth experiences beyond the home. Yet no large-scale study has examined patterns of exposure to the most commonly used operationalization of neighborhood - the census tract - among urban youth. Using smartphone GPS data from the Adolescent Health and Development in Context study (N=1405), we estimate the amount of time youth spend in residential neighborhoods and consider explanations for variation in neighborhood exposure. On average, youth (ages 11 to 17) spend 5.7% of their waking time in their neighborhood but not at home, 60% at home, and 34.3% outside their neighborhood. Multilevel negative binomial regression models indicate that residence in economically disadvantaged neighborhoods is associated with less time in neighborhood. Higher levels of local violence and the absence of a neighborhood school the youth is eligible to attend are negatively associated with time in neighborhood and mediate the concentrated disadvantage effect. Fractional multinomial logit models indicate that higher violence is linked with increased time at home while school absence is associated with increased outside-neighborhood time. Theoretical development and empirical research on neighborhood effects should incorporate findings on the extent and nature of neighborhood and broader activity space exposures among urban youth.