ABSTRACT
Bipartite projections (e.g., event co-attendance) are often used to measure unipartite networks of interest (e.g., social interaction). Backbone extraction models can be useful for reducing the noise inherent in bipartite projections. However, these models typically assume that the bipartite edges (e.g., who attended which event) are unconstrained, which may not be true in practice (e.g., a person cannot attend an event held prior to their birth). We illustrate the importance of correctly modeling such edge constraints when extracting backbones, using both synthetic data that varies the number and type of constraints, and empirical data on children's play groups. We find that failing to impose relevant constraints when the data contain constrained edges can result in the extraction of an inaccurate backbone. Therefore, we recommend that when bipartite data contain constrained edges, backbones be extracted using a model such as the Stochastic Degree Sequence Model with Edge Constraints (SDSM-EC).
Subject(s)
Social Interaction , Humans , Models, Theoretical , ChildABSTRACT
Childfree adults neither have nor want children, but estimates of their prevalence vary widely, leading to ambiguity about how common this family status actually is. The goal of this study is to examine the effects of sample composition, time, and question wording on estimates of the prevalence of childfree adults. We pool 83 nationally representative estimates of the prevalence of childfree adults in Japan since 2000 using meta-regression to identify the influence of sex, marital status, year, and survey question. Prevalence estimates are higher when computed from samples of women than men, from samples of singles than married people, from samples collected more recently, and from surveys asking questions about expectations than wants. Most of the variation in estimates of the prevalence of childfree adults can be attributed to differences in sample composition, time, and question wording. Taking these factors into account, we estimate that over 2.5 million Japanese adults age 18-50, or 5.64% of this population, were childfree in 2020.
Subject(s)
Prevalence , Adult , Male , Child , Humans , Female , Adolescent , Young Adult , Middle Aged , Japan/epidemiology , Marital Status , Surveys and QuestionnairesABSTRACT
Distinctiveness centrality, which was proposed in 2020 to identify nodes that are connected to poorly-connected neighbors, is simply a minor variation on two existing centrality measures: beta centrality proposed in 1987, and gamma centrality proposed in 2011. In toy, empirical, and generated networks, I show that these three centrality measures yield identical node rankings under nearly all circumstances. Researchers seeking to identify nodes that are connected to poorly-connected others should not use distinctiveness centrality, and instead should use either beta or gamma centrality because they are more widely-known in the literature, are more flexible, and are computationally simpler. Additionally, researchers should be cautious when proposing new centrality measures, taking care to avoid duplicating measures that already exist.
ABSTRACT
Childfree adults are the most common type of non-parent in the United States and are distinguished by their lack of desire to have children. Although there are many reasons one may choose not to have children, recent restrictions on reproductive health care may also contribute to this decision. For example, the United States Supreme Court's decision in Dobbs v. Jackson eliminated a long-standing constitutional protection for abortion access, which reduced patients' medical autonomy and increased the risks of pregnancy and childbirth, and therefore may have led adults to decide not to have children. In this study, we use representative data on Michigan adults immediately before and after the Dobbs decision to examine changes in the prevalence of childfree adults in this population. We find that 21% of Michigan adults were childfree before the Dobbs decision, but this number rose to nearly 26% after the decision. Controlling for demographic characteristics, a Michigan adult was 32.8% more likely to be childfree after the Dobbs decision than before. We conclude that when access to safe reproductive health care is uncertain or unavailable, adults that do not already have children may decide that they do not want children.
Subject(s)
Abortion, Induced , Female , Pregnancy , Adult , Child , Humans , Michigan , Prevalence , Cluster Analysis , Delivery, ObstetricABSTRACT
OBJECTIVE: The overall goal of this work is to produce a set of recommendations (SoNHR-Social Networks in Health Research) that will improve the reporting and dissemination of social network concepts, methods, data, and analytic results within health sciences research. METHODS: This study used a modified-Delphi approach for recommendation development consistent with best practices suggested by the EQUATOR health sciences reporting guidelines network. An initial set of 28 reporting recommendations was developed by the author team. A group of 67 (of 147 surveyed) experienced network and health scientists participated in an online feedback survey. They rated the clarity and importance of the individual recommendations, and provided qualitative feedback on the coverage, usability, and dissemination opportunities of the full set of recommendations. After examining the feedback, a final set of 18 recommendations was produced. RESULTS: The final SoNHR reporting guidelines are comprised of 18 recommendations organized within five domains: conceptualization (how study research questions are linked to network conceptions or theories), operationalization (how network science portions of the study are defined and operationalized), data collection & management (how network data are collected and managed), analyses & results (how network results are analyzed, visualized, and reported), and ethics & equity (how network-specific human subjects, equity, and social justice concerns are reported). We also present a set of exemplar published network studies which can be helpful for seeing how to apply the SoNHR recommendations in research papers. Finally, we discuss how different audiences can use these reporting guidelines. CONCLUSIONS: These are the first set of formal reporting recommendations of network methods in the health sciences. Consistent with EQUATOR goals, these network reporting recommendations may in time improve the quality, consistency, and replicability of network science across a wide variety of important health research areas.
Subject(s)
Research Design , Social Networking , Humans , Guidelines as TopicABSTRACT
Childfree individuals, who are also described as 'childless by choice' or 'voluntarily childless', have decided they do not want biological or adopted children. This is an important population to understand because its members have unique reproductive health and end-of-life needs, and they encounter challenges managing work-life balance and with stereotypes. Prior estimates of childfree adults' prevalence in the United States, their age of decision, and interpersonal warmth judgements have varied widely over time and by study design. To clarify these characteristics of the contemporary childfree population, we conduct a pre-registered direct replication of a recent population-representative study. All estimates concerning childfree adults replicate, boosting confidence in earlier conclusions that childfree people are numerous and decide early in life, and that parents exhibit strong in-group favoritism while childfree adults do not.
Subject(s)
Child, Adopted , Judgment , Child , Humans , Adult , United States , Prevalence , ParentsABSTRACT
Urbanists have long been interested in understanding what makes people satisfied with their neighborhoods. However, relatively little is known about how residents' personality traits may affect their neighborhood satisfaction. In this paper, we explore the direct and indirect associations of personality traits with neighborhood satisfaction in a representative sample of adults in Michigan (USA). We find that each of the personality traits in the five factor model are associated with neighborhood satisfaction in the same way that they are known to be associated with subjective well-being. However, we fail to observe evidence that personality traits moderate the association between perceptions of neighborhoods and neighborhood satisfaction, or that personality's association with neighborhood satisfaction is mediated by neighborhood perceptions. We conclude that there is potential for drawing on theoretical and empirical developments in positive psychology for understanding neighborhood satisfaction, but observe that the underlying mechanisms for the association between personality and neighborhood satisfaction remain unknown.
Subject(s)
Personality , Residence Characteristics , Adult , Humans , Personal Satisfaction , Personality Disorders , MichiganABSTRACT
The analysis of psychological networks has become common in multiple subfields including clinical, social, and personality psychology, where the focus is often on identifying highly central nodes that represent symptoms, beliefs, or traits. However, the boundaries of these networks are often ambiguous and relevant nodes are often missing from the network. In this article, we use a series of simulations to show that even under typical conditions of missingness, the centrality of nodes in an empirical psychological network are poorly correlated or uncorrelated with their centrality in a hypothetical "true" psychological network, and thus are invalid. We illustrate the implications of this lack of validity using an empirical example drawn from a recent study of political belief system networks, demonstrating that the original study would have drawn incorrect conclusions about American's most central political beliefs. We conclude by recommending that centrality measures should be computed and interpreted only in psychological networks that include (nearly) all the nodes inside a theoretically meaningful boundary. (PsycInfo Database Record (c) 2023 APA, all rights reserved).
ABSTRACT
Both urban planners and urban scholars have been keenly interested in identifying the characteristics associated with neighborhood satisfaction. One robust but surprising pattern is that the presence or number of children in a household has no effect on neighborhood satisfaction. To clarify this pattern, we measured the neighborhood satisfaction of a representative sample of 1,000 Michigan adults, whom we divided into six distinct reproductive statuses: co-parents, single-parents, empty nesters, not-yet-parents, childless individuals, and childfree individuals. We found that a simple parent vs. non-parent dichotomy hides significant heterogeneity among these groups. Specifically, we found that single parents and childfree individuals experience significantly less neighborhood satisfaction than other groups. We conclude by reflecting on the methodological and practical implications of differences in neighborhood satisfaction when more nuanced reproductive statuses are considered.
Subject(s)
Personal Satisfaction , Residence Characteristics , Adult , Child , Family Characteristics , Humans , Parents , Single ParentABSTRACT
Childfree adults do not want to have children, making them distinct from parents and other adults without children. However, they are difficult to study because they cannot be identified using conventional data on fertility. We use data from a representative sample in the United States to examine the prevalence, age of decision, and interpersonal warmth judgments by and about childfree adults. Our prevalence estimates suggest that childfree adults are quite common, comprising over one-fifth (21.64%) of the population. Our analysis of age-to-decision suggests that most childfree adults reported that they decided they did not want children early in life. Finally, our analysis of interpersonal warmth suggests asymmetric affective polarization among parents and childfree adults driven primarily by parent's ingroup favoritism. We discuss the implications of these findings for our understanding of childfree adults and for future research on this historically overlooked segment of the population.
Subject(s)
Fertility , Judgment , Adult , Child , Humans , Parents , Prevalence , United StatesABSTRACT
Networks are useful for representing phenomena in a broad range of domains. Although their ability to represent complexity can be a virtue, it is sometimes useful to focus on a simplified network that contains only the most important edges: the backbone. This paper introduces and demonstrates a substantially expanded version of the backbone package for R, which now provides methods for extracting backbones from weighted networks, weighted bipartite projections, and unweighted networks. For each type of network, fully replicable code is presented first for small toy examples, then for complete empirical examples using transportation, political, and social networks. The paper also demonstrates the implications of several issues of statistical inference that arise in backbone extraction. It concludes by briefly reviewing existing applications of backbone extraction using the backbone package, and future directions for research on network backbone extraction.
ABSTRACT
Understanding educators' networks can inform the field of school psychology by offering insight into how the structure of social relationships supports the implementation of school-based programs. However, the difficulties of collecting and modeling network data remain barriers to using network methods in school psychology. To overcome these barriers, we provide a step-by-step tutorial for collecting, modeling, and visualizing network data from educators. We draw on an example from a study designed to understand advice networks among middle and high school educators involved in implementing a system-level intervention to prevent school dropout. (PsycInfo Database Record (c) 2023 APA, all rights reserved).
Subject(s)
Interpersonal Relations , Psychology, Educational , Humans , SchoolsABSTRACT
Racial disparities have been observed in the impacts of COVID-19 in the USA. In the present paper, we used a representative sample of adults in Michigan to examine differences in COVID-19 impacts on Blacks and Whites in four domains: direct, perceived, political, and behavioral. We found that in the initial wave of the outbreak in May 2020, Blacks experienced more severe direct impacts: they were more likely to be diagnosed or know someone who was diagnosed, and more likely to lose their job compared to Whites. In addition, Blacks differed significantly from Whites in their assessment of COVID-19's threat to public health and the economy, the adequacy of government responses to COVID-19, and the appropriateness of behavioral changes to mitigate COVID-19's spread. Although in many cases these views of COVID-19 were also associated with political ideology, this association was significantly stronger for Whites than Blacks. Continued investigation of racial disparities in COVID-19's impact is necessary; however, these preliminary findings of a race-by-ideology interaction are important because they suggest some racial disparities are restricted to conservatives, while more liberal Whites and Blacks exhibit few differences.
Subject(s)
COVID-19 , Adult , Health Status Disparities , Humans , Michigan/epidemiology , SARS-CoV-2 , White PeopleABSTRACT
Projections of bipartite or two-mode networks capture co-occurrences, and are used in diverse fields (e.g., ecology, economics, bibliometrics, politics) to represent unipartite networks. A key challenge in analyzing such networks is determining whether an observed number of co-occurrences between two nodes is significant, and therefore whether an edge exists between them. One approach, the fixed degree sequence model (FDSM), evaluates the significance of an edge's weight by comparison to a null model in which the degree sequences of the original bipartite network are fixed. Although the FDSM is an intuitive null model, it is computationally expensive because it requires Monte Carlo simulation to estimate each edge's p value, and therefore is impractical for large projections. In this paper, we explore four potential alternatives to FDSM: fixed fill model, fixed row model, fixed column model, and stochastic degree sequence model (SDSM). We compare these models to FDSM in terms of accuracy, speed, statistical power, similarity, and ability to recover known communities. We find that the computationally-fast SDSM offers a statistically conservative but close approximation of the computationally-impractical FDSM under a wide range of conditions, and that it correctly recovers a known community structure even when the signal is weak. Therefore, although each backbone model may have particular applications, we recommend SDSM for extracting the backbone of bipartite projections when FDSM is impractical.
ABSTRACT
In network science, identifying optimal partitions of a signed network into internally cohesive and mutually divisive clusters based on generalized balance theory is computationally challenging. We reformulate and generalize two binary linear programming models that tackle this challenge, demonstrating their practicality by applying them to partition signed networks of collaboration and opposition in the US House of Representatives. These models guarantee a globally optimal network partition and can be practically applied to signed networks containing up to 30,000 edges. In the US House context, we find that a three-cluster partition is better than a conventional two-cluster partition, where the otherwise hidden third coalition is composed of highly effective legislators who are ideologically aligned with the majority party.
ABSTRACT
Childfree individuals choose not to have children, which makes them a distinctive group from parents who have had children, not-yet-parents who plan to have children, and childless indivduals who would have liked to have children. Most research on parental status and psychosocial characteristics has not effectively distinguished childfree individuals from other non-parents or has relied on non-representative samples. In this study, we use a representative sample of 981 Michigan adults to estimate the prevalence of childfree individuals, to examine how childfree individuals differ from parents and other types of non-parents in life satisfaction, political ideology, and personality, and to examine whether childfree individuals are viewed as an outgroup. We find that over a quarter of Michigan adults identified as childfree. After controlling for demographic characteristics, we find no differences in life satisfaction and limited differences in personality traits between childfree individuals and parents, not-yet-parents, or childless individuals. However, childfree individuals were more liberal than parents, and those who have or want(ed) children felt substantially less warm toward childfree individuals than childfree individuals felt toward each other. Given the prevalence of childfree individuals, the risks of their outgroup status, and their potential role in politics as a uniquely liberal group, it is important for demographic research to distinguish the childfree from others and to better understand these individuals.
Subject(s)
Personality/physiology , Data Collection , Family Planning Services , Female , Humans , Male , Michigan/epidemiology , Parents , Prevalence , Surveys and QuestionnairesABSTRACT
Bipartite projections are used in a wide range of network contexts including politics (bill co-sponsorship), genetics (gene co-expression), economics (executive board co-membership), and innovation (patent co-authorship). However, because bipartite projections are always weighted graphs, which are inherently challenging to analyze and visualize, it is often useful to examine the 'backbone,' an unweighted subgraph containing only the most significant edges. In this paper, we introduce the R package backbone for extracting the backbone of weighted bipartite projections, and use bill sponsorship data from the 114th session of the United States Senate to demonstrate its functionality.
Subject(s)
User-Computer Interface , Algorithms , Politics , United StatesABSTRACT
This study draws on two communities theory to address two major research questions related conceptions of research in educational practice and policy. First, how do educators conceptualize research? Second, to what extent do educators' conceptions of research align with recent U.S. federal educational policies? We conducted 90 semi-structured interviews with educators in the United States, asking them what comes to mind when they think of research. We used open, axial, and selective coding to characterize educators' conceptions of research. We also compared educators' conceptions of research to two U.S. federal educational policies that define scientifically based research and evidence-based interventions. Findings indicate that educators and policies defined research in similar ways, but each included some unique characteristics. Implications from the study include the need for increased communication between federal policy-makers and educators and improved reporting by researchers to better attend to the needs of educators and policymakers.
ABSTRACT
Despite growing interest in data visualization and graphically aided reporting, the evaluation literature could benefit from additional guidance on systematically integrating visual communication design and marketing into comprehensive communication strategies to improve data dissemination. This article describes the role of targeted communication strategies-based on visual communications, design, and marketing theory-in producing more effective reports. In evaluation practice, well-synthesized and translated reports often require the integration of data from multiple sources, methods, and/or time points to communicate complex findings in ways that elicit productive responses. Visual communication strategies, such as project branding or designing actionable tools with marketing principles in mind, can be applied to optimize effective reporting of complex evaluation findings. This article references a longitudinal, mixed-method evaluation of public school administrators in Michigan to illustrate the application of a systematic communication design framework to produce several graphically aided project materials and subsequent findings reports.
ABSTRACT
PURPOSE: We apply diffusion of innovations theory to examine two key research questions designed to inform efforts to improve the research-practice gap in education: (1) Are there distinct types of educators that differ in their prioritization of the compatibility, observability, complexity, relative advantage, and trialability of research? and (2) Are educators' roles or context associated with their categorization in this typology? RESEARCH METHODS: Using semi-structured interview data in two Michigan counties from intermediate school district staff (N=24), district central office staff (N=18), principals (N=22), and school building staff (N=23), we first used directed content analysis to code for mentions of compatibility, observability, complexity, relative advantage, and trialability. Next, using the coded data, we conducted a hierarchical agglomerative cluster analysis and follow-up cross-tabulations to assess whether cluster memberships were associated with educators' roles or county context. FINDINGS: Educators in our sample could be categorized in one of five clusters distinguished primarily by different patterns of prioritization of the compatibility, observability, and complexity of research. Membership in these clusters did not vary by role but did vary by county, suggesting the importance of context for educators' perceptions of research. IMPLICATIONS FOR RESEARCH AND PRACTICE: These findings suggest that narrowing the research-practice gap in education will require attending to multiple audiences of educators with distinct priorities that guide their perceptions and use of educational research and evidence-based practices.