Socius最新文献

Many infectious diseases such as coronavirus disease 2019 spread through preexisting social networks. Although network models consider the implications of micro-level interaction patterns for disease transmission, epidemiologists and social scientists know little about the meso-structure of disease transmission. Meso-structure refers to the pattern of disease spread at a higher level of aggregation, that is, among infection clusters corresponding to organizations, locales, and events. The authors visualizes this meso-structure using publicly available contact tracing data from Singapore. Visualization shows that one highly central infection cluster appears to have generated on the order of seven or eight infection chains, amounting to 60 percent of nonimported cases during the period considered. However, no other cluster generated more than two infection chains. This heterogeneity suggests that network meso-structure is highly consequential for epidemic dynamics.

The coronavirus disease 2019 pandemic has caused unprecedented disruptions to education in the United States, with a large proportion of schooling moving to online formats, which has the potential to exacerbate existing racial/ethnic and socioeconomic disparities in learning. The authors visualize access to online learning technologies using data from the Household Pulse Survey from the early fall 2020 school period (August 19 to October 26). The authors find that 10.1 percent of children participating in online learning nationally did not have adequate access to the Internet and a computer. Rates of inadequate access varied nearly 20-fold across the gradient of parental race/ethnicity and education, from 1.9 percent for children of Asian parents with graduate degrees to 35.5 percent among children of Black parents with less than a high school education. These findings indicate alarming gaps in potential learning among U.S. children. Renewed investments in equitable access to distance-learning resources will be necessary to prevent widening racial/ethnic and class learning disparities.

Previous studies have established that firm size is associated with a wage premium, but the wage premium has declined in recent decades. The authors examine the risk for unemployment by firm size during the initial outbreak of coronavirus disease 2019 in the United States. Using both yearly and state-month variation, the authors find greater excess unemployment among workers in small enterprises than among those in larger firms. The gaps cannot be entirely attributed to the sorting of workers or to industrial context. The firm size advantage is most pronounced in sectors with high remotability but reverses in the sectors most affected by the pandemic. Overall, these findings suggest that firm size is linked to greater job security and that the pandemic may have accelerated prior trends regarding product and labor market concentration. They also point out that the initial policy responses did not provide sufficient protection for workers in small and medium-sized businesses.

A large body of sociological research has shown that racial minorities and women experience significant disadvantages in the labor market. In this visualization, the author presents evidence from the Current Population Survey examining the effects of the coronavirus disease 2019 crisis on racial and gender inequalities in employment in the United States among prime-age workers. The author shows that the white-nonwhite gap in employment increased significantly during the post-outbreak period. Results from individual fixed-effects regression models show a strong white male advantage in the likelihood of being laid off for post-outbreak months compared with women, black men, Hispanic men, and Asian men.

Exercising is crucial to keeping up physical and mental health during the coronavirus disease 2019 (COVID-19) pandemic. In this visualization, the authors consider how existing social inequalities may create unequal physical exercise patterns during COVID-19 in the United States. Analyzing data from a nationally representative Internet panel of the University of Southern California Center for Economic and Social Research Understanding Coronavirus in America project (March to December), the authors find that although all Americans have become physically more active since the outbreak, the pandemic has also exacerbated the inequality in physical exercise. Specifically, the authors show that the gaps in physical exercise have widened substantially between men and women, whites and nonwhites, the rich and the poor, and the educated and the less educated. Policy interventions addressing the widening inequality in physical activity can help minimize the disproportionate mental health impact of the pandemic on disadvantaged populations.

COVID-19 has had dramatic impacts on economic outcomes across the United States, yet most research on the pandemic's labor-market impacts has had a national or urban focus. We overcome this limitation using data from the U.S. Current Population Survey's COVID-19 supplement to study pandemic-related labor-force outcomes in rural and urban areas from May 2020 through February 2021. We find the pandemic has generally had more severe labor-force impacts on urban adults than their rural counterparts. Urban adults were more often unable to work, go unpaid for missed hours, and be unable to look for work due to COVID-19. However, rural workers were less likely to work remotely than urban workers. These differences persist even when adjusting for adults' socioeconomic characteristics and state-level factors. Our results suggest rural-urban differences in the nature of work during the pandemic cannot be explained by well-known demographic and political differences between rural and urban America.

Adolescents with high educational expectations are more likely to enroll in college. Although most adolescents today report high educational expectations, there remains important racial/ethnic heterogeneity in college enrollment patterns. In particular, at every level of socioeconomic status, minority youth have higher educational expectations than their white peers yet enroll in college at lower rates. The rapidly increasing size and college enrollment of the Hispanic population motivate renewed examination of the expectation-enrollment relationship. Using data from the Education Longitudinal Study (ELS) and the High School Longitudinal Study (HSLS), the author examines whether the relationship between adolescent educational expectations and enrollment in a four-year college within two years of high school graduation differs by race/ethnicity and whether this relationship changed over time. The author finds that the expectation-enrollment relationship is positive for all students but is smaller for black and Hispanic students in the ELS cohort. However, by the HSLS cohort, the gaps have largely closed.

We analyze the expression of sexualities in the contemporary United States using data about disclosure on social media. Through the Facebook advertising platform, we collect aggregate counts encompassing 200 million Facebook users, 28% of whom disclose sexuality-related information. Stratifying by age, gender, and relationship status, we show how these attributes structure the propensity to disclose different sexual identities. We find a large generational difference; younger social media users share their sexualities at high rates, while for older cohorts marital status substitutes for sexual identity. Consistent with gendered expectations, women more often express a bisexual interest in men and women; men are more explicit about their heterosexuality. We interpret these variations in sexuality disclosure on social media to reflect the salience of sexual identity, intersected at times with availability. Our study contributes to the sociology of sexuality with a quantitative analysis, using novel digital data, of how sexuality is signaled socially.

Sociologists have argued that there is value in incorporating computational tools into qualitative research, including using machine learning to code qualitative data. Yet standard computational approaches do not neatly align with traditional qualitative practices. The authors introduce a hybrid human-machine learning approach (HHMLA) that combines a contemporary iterative approach to qualitative coding with advanced word embedding models that allow contextual interpretation beyond what can be reliably accomplished with conventional computational approaches. The results, drawn from an analysis of 87 human-coded ethnographic interview transcripts, demonstrate that HHMLA can code data sets at a fraction of the effort of human-only strategies, saving hundreds of hours labor in even modestly sized qualitative studies, while improving coding reliability. The authors conclude that HHMLA may provide a promising model for coding data sets where human-only coding would be logistically prohibitive but conventional computational approaches would be inadequate given qualitative foci.