Amid escalating global affective polarization, this study examines the ramifications of within-couple political dissimilarities for marital relationships using representative dyadic data from Hong Kong (N = 1003 married couples)—the first such examination in a non-democratic, rapidly polarizing context. Drawing on value incongruence theory and the affective polarization literature, the study proposes three sets of hypotheses regarding the roles of spousal political identity and value dissimilarities in the frequency, scope, and intensity of marital conflict, as well as the moderating role of affective polarization. Regression analyses show that political identity dissimilarity is a robust and independent predictor of more frequent marital conflicts over political matters and relationship issues. Notably, identity dissimilarity is associated with a higher likelihood of spousal psychological aggression even after controlling for value dissimilarities and conflict frequencies, indicating that identity-based antagonism is not reducible to simply more frequent and broader disagreement. Affective polarization strengthens the associations between political identity dissimilarity and conflicts over non-political domains. Consequently, politically misaligned couples experience both more frequent conflicts across multiple domains and more intense conflicts, manifested as psychological aggression. Value dissimilarities, by contrast, exhibit weaker and less consistent links to conflict frequency and intensity. These results demonstrate that identity-based antagonism—amplified by the moralization of political orientations—drives marital strain beyond substantive policy or value disagreements.
The COVID pandemic expanded work-from-home (WFH) access beyond highly educated groups, potentially reshaping gendered divisions of care. Drawing on doing gender, absolute resources and the time-flexibility framework, this study examines how WFH affects care-related travel in German cities, focusing on gender and educational inequalities. Care-related travel is a significant yet frequently disregarded component of unpaid labor, thus making mobility data crucial for capturing gender differences when WFH. Using representative mobility data from the Mobility in Germany (2017, n=14,421) and Mobility under the Influence of the Coronavirus Pandemic (2022/2023, n= 695) surveys, I examine changes in WFH likelihood and daily care-related travel before and after COVID. OLS regressions estimate differences in care-related trips, while logistic regressions assess WFH probabilities by education and gender. The findings reveal distinct educational patterns in gender disparities. Before COVID-19, non-university-educated men were more likely to WFH than women, but this gender gap disappeared after the pandemic. WFH is associated with higher care-related mobility among non-university-educated women before and after the pandemic, while this association is weaker or absent among university-educated women. These findings suggest that WFH may reconfigure gendered care responsibilities in ways that differ by educational attainment, with care-related travel burdens increasing most clearly among women without a university degree.
This study examines associations between vertical and horizontal educational stratification and the risk of reduction in subjective socioeconomic status between young adulthood and early midlife. A consistent predictor of well-being in psychology research, status loss is an understudied aspect of US economic stratification in sociology. The analysis uses data from the National Longitudinal Study of Adolescent to Adult Health to model the associations between vertical—across education-level—and horizontal—among college graduates—education measures with subjective status loss in early adulthood. Adjusted estimates reveal that both college degree completion and higher college selectivity are associated with reduced risk of status loss. Current household income mediates a substantial portion of the associations between college degree completion and college selectivity with the outcome. Thus, there is both stratification by college completion and among college graduates in the risk of status loss. Future research should explore dynamic effects of life events—such as divorce—and changes in economic and social circumstances—such as job loss—on status loss using contemporary panel data approaches.
Educational choices are heavily stratified by gender. In this study we examine whether an important structural factor in the family of origin, sex composition of the sibling group, affects the emergence of the gender-typed preferences that may drive these choices. Using Swedish population register data, we use multinomial logistic regression models to examine educational choices at upper-secondary school (N = 424,788) and university (N = 133,222) for first-born Swedish men and women in two-child sibling groups, with analyses based on cohorts born 1980–2003. We further examine whether there are any differences in educational choices by parental educational level. We focus on first-born children in two-child families, and we present supplementary analyses for second-born children. We find that men in two-child sibling groups with brothers are more likely to apply for female-dominated upper-secondary programs than men with sisters, while men with sisters are significantly more likely to follow some male-dominated tracks than men with brothers. These differences are concentrated amongst men raised by parents with lower education. However, we do not find that sex composition is associated with university subject choices for men, and we do not find any differences in educational choices for women by the sex composition of the sibling group. These results suggest some support for the predictions that a mixed sex sibling group can lead to a more gender-typed upbringing for men.
Racialized population displacement is one of the mechanisms through which a disaster may entrench existing social structures and processes. The present study uses Hurricane Katrina's residential mobility and outmigration impacts on Black and White adults as a case study to explore this question. Drawing on a population-representative survey of pre-Katrina Orleans Parish (City of New Orleans) residents four to five years after Katrina, and on 1985-1990 and 1995-2000 decennial Census microdata, we first compared observed 2005-2009/10 mobility to mobility patterns of all Black and White Orleans Parish adults estimated in the absence of Katrina. Second, we used the survey data to counterfactually apply working-age White mobility rates to working-age Black Orleans Parish residents with equivalent housing damage and socio-economic characteristics, and vice versa. We estimated that 53,000, or almost one in four, Black Orleans Parish adult residents were displaced long term from the New Orleans metropolitan area, whereas at the population level White Orleans Parish residents suffered no displacement. Racial disparities in residential moves from Orleans Parish to suburban New Orleans parishes were crucial in explaining racial disparities in displacement from the metropolitan area. Had White working-age Orleans Parish residents suffered the same housing damage as Black residents and had equivalent socio-economic characteristics, 29% would have moved to suburban New Orleans parishes, versus an observed 4% of Black Orleans Parish residents who made such moves, indicating strong entrenchment of pre-Katrina residential racial stratification and segregation processes.
The constrict thesis hypothesises that people ‘hunker down’ or withdraw from social and civic life in the face of migrant and ethnic diversity, degrading the collective social capital and cohesion of diverse communities. The large body of research since has focused on the second part of this thesis: the relationship between diversity and social capital and cohesion and the moderating role of the wider urban ecology, neglecting the first part: whether people are engaged and socially connected within diverse spaces or whether they withdraw and hunker down. In this study, we draw on longitudinal survey data from Australia to quantitatively examine both aspects, analysing and projecting trajectories of pro- and anti-social behaviours and trust in linguistically diverse areas. The results suggest that pro-social behaviours become more common the longer people have lived in diverse areas, while hostility between neighbours declines. This finding is argued to be important for understanding processes of intergroup contact and the development of intercultural harmony. Indeed, the results suggest that while an element of hunkering down in diverse spaces appears at play, this can with time and enabling conditions give way to greater sociability and cohesion in diverse spaces.
Prior research shows that cohabiting individuals are more likely than married individuals to form interracial unions. This pattern is frequently attributed to the perception of cohabitation as a lower-commitment union, which may lead single men and women to be less selective in choosing cohabiting partners than in choosing marital spouses. Yet, the meaning of cohabitation varies across groups: for some, it functions as a marriage-like union, while for others, it serves as a form of engagement, trial marriage, or an alternative to dating. This paper examines whether the gap in mate selection behavior by cohabitation status varies according to the role that cohabitation plays in the family life of different racial/ethnic, nativity, and educational groups. Using data from the 2008–2024 American Community Survey, we estimate log-linear models examining how the odds of forming interracial unions vary by cohabitation status as well as the partner's race/ethnicity, education, nativity status, and gender. Results confirm that cohabitation is more likely than marriage to involve partners of a different racial/ethnic background. The size of this gap varies by partner's race/ethnicity, nativity status, education, and gender. Specifically, among women, this gap is smallest among Hispanics and largest among Blacks. Among men, this gap is smallest among Hispanics and largest among Asian Americans. Overall, our findings reveal that mate selection behavior varies not only by union type (cohabitation or marriage) but also by partner's race/ethnicity, nativity status, education, and gender.
This study explores the impact of subsidized childcare on maternal employment, focusing on the heterogeneous effects of Germany's Childcare Funding Act (2013) on native and first- and second-generation migrant mothers. Despite lower childcare use and higher unemployment rates among migrant women, research on this topic has been limited. Using the SOEP dataset from 2010 to 2019 and a triple difference approach, we find that subsidized childcare significantly increases maternal employment, particularly for mothers with children aged 2, with lasting effects. However, migrant women's policy responses remain lower than that of natives and no significant differences are observed between the first- and second-generation mothers. Our findings, hence, provide insights for policymaking which targets labour market integration of migrant women.
Social class origin is closely linked to individuals' labor market outcomes. This study examines the class ceiling—the earnings disadvantage that those from lower-class origins experience within higher-class occupations. We propose that job task segregation, i.e., the non-random selection into different task profiles within granular occupations, contributes to the class ceiling. We study these dynamics in Germany, characterized by rigid social structures and low social mobility. Using high-quality survey data with individual measures of job tasks (National Educational Panel Study Adult Cohort), we estimate Poisson pseudo-maximum likelihood (PPML) regression with a log link. Class origin effects on earnings are most pronounced in higher salariat positions, where individuals from privileged origins earn substantially more than their lower-origin counterparts, even within the same occupations. Job tasks differentials by social origin are visible; those from an advantaged background perform more analytical and creative job tasks, even within the same occupations. These differentials explain a small part of the earnings gap by class origin. The findings contribute to understanding the intergenerational transmission of advantage, particularly how class origin shapes outcomes within occupations.
This paper investigates the relation between Americans’ cultural models of religion in public life, and views of religious groups in American society. We explicate how three dominant cultural models, Christian nationalism, civil religion, and liberal secularism, influence attitudes toward atheists, Buddhists, Muslims, Jews, LDS (Mormons), and Conservative Christians. Do survey respondents see each religious group as consistent with their vision of the ideal American society? Using the national survey data available in the American Mosaic Study of 2014, we find that Christian nationalism and civil religion operate in distinct ways to shape views of religious and nonreligious outgroups. We find evidence that Christian nationalists oppose Muslims, atheists, and other religious groups in public life, with the exception of conservative Christians, Jews and LDS. Civil religionists are more open to a pluralistic public square and thus have more positive views of most other religions. Liberal secularists favor atheists but are generally opposed to religious groups in their vision of American society, especially conservative Christians and Mormons.
This study investigates the long-term, gender-specific impacts of parental migration during childhood on educational attainment in adulthood. While migration can bring economic benefits through remittances, it also disrupts family caregiving arrangements, triggers emotional distress, and changes values and preferences. Using data from the China Family Panel Study, this article examines how paternal, maternal, and dual-parent outmigration before age 12 influences children's year of completed schooling by adulthood. The findings show that the father's outmigration is associated with an increase in educational attainment, particularly when the father leaves for shorter durations and in early childhood. Conversely, the mother's outmigration is associated with a reduction in years of education, with stronger impacts when the migration lasts longer (over 1 year) and occurs in later childhood (ages 4–12). The impact of parental outmigration also varies by the child's gender and rural-urban context.
Sociologists argue that cultural tastes shape perceptions of individuals’ status and personal qualities. Yet it remains unclear whether highbrow and omnivorous tastes create different perceptions. To address this gap, we designed two vignette survey experiments to identify the effect of highbrow (vs. lowbrow) and omnivorous (vs. univorous) tastes on a wide range of perceptions. Results for Denmark, replicated in two experiments, show that while highbrow (vs. lowbrow) tastes elicit favorable perceptions of status, labor market success, and sophistication (i.e., being high-class, high-earning, and cultured), omnivorous (vs. univorous) tastes mainly elicit favorable perceptions of sociability (i.e., being socially oriented and warm). Supplementary analyses of qualitative text data corroborate these findings, including that omnivorous tastes elicit favorable perceptions of sociability. Finally, highbrow and omnivorous tastes create clearer perceptions of status, measured by lower variance in ratings of status. We end by addressing the implications of our results for research on cultural taste, labor markets, elites, and intersectional inequality.
Although maternal employment has long been linked to children's later labor market outcomes, most research treats employment as a static status rather than an evolving process across childhood. This study reassesses the relationship between maternal employment and adult children's labor market participation by adopting a trajectory-based perspective that captures the dynamic nature of mothers' work patterns, with particular attention paid to timing and intensity. Drawing on mother-child dyadic longitudinal data from the National Longitudinal Survey of Youth (1979) (NLSY79) and the NLSY79 Child and Young Adult (n = 2,186), this study applies sequence analysis to classify distinct patterns of maternal employment trajectories across children's childhood and random-effects multinomial logit models with inverse probability of treatment weighting to assess how these trajectories predict adult children's labor market participation. Five trajectories emerge: continuously full-time, early return, late return, part-time, and not working. Relative to continuously full-time employment, a not-working trajectory is associated with a substantially higher likelihood of non-employment in adulthood for both sons and daughters, though the association for sons weakens after accounting for pre-birth family background. Parttime and late-return trajectories predict higher odds of reduced employment only among daughters. These findings underscore the gendered nature of intergenerational transmission in labor market participation.
Educational attainment and achievement are key areas of study for social scientists because they are associated with a wide range of outcomes, including income, careers, marriage, fertility, migration, and longevity. Using unique multimethod, longitudinal data from Nepal, a low-income setting, we explore how the availability of sanitary pads in restrooms is associated with school achievement. We focus on eighth grade students, marking a pivotal time in which many female students begin menstruating. Multilevel modeling estimates reveal that the availability of sanitary pads in school restrooms have a statistically significant positive association with an increase in language test scores for females. The results also show a positive association between the availability of sanitary pads and an increase in math and science test scores, but these associations are not statistically significant. These findings reveal that school hygiene facilities can play an important role in school achievement, which can shape students’ matriculation into the next academic year and potential longer-term consequences including both academic and social outcomes, such as school dropout, income, and marriage timing.
This paper explores how the socioeconomic composition of individuals' social networks shapes attitudes towards inequality in contemporary societies. While limited cross-national research has examined the location of people's acquaintances within the social structure, this study fills a crucial gap by jointly analyzing their heterogeneity and concentration. Using data from the ISSP 2017 Social Networks Module, the study investigates how socioeconomic network diversity and status relate to perceptions of inequality and preferences for redistribution across 32,310 individuals in 31 countries. Country fixed-effects regression results indicate that individuals with more socioeconomically diverse networks show modestly higher national inequality perceptions, whereas having higher-status acquaintances is strongly associated with lower support for redistribution. These effects are not uniform: network diversity often acts as a catalyst and network status as a buffer of distributive attitudes. Social networks thus operate simultaneously as comparative frames (highlighting disparities) and reference frames (anchoring norms), amplifying or dampening individuals' views on inequality. Notably, network effects operate consistently across all socioeconomic groups, regardless of individuals' SES. Overall, the study advances meso-level perspectives on distributive justice by demonstrating how everyday social relationships shape perceptions and beliefs about inequality in stratified societies.
Over the past several decades, a growing body of research has applied a group threat framework to the understanding of hate crimes. This research has been limited, however, by both a narrow focus on racially-motivated hate crimes, and an overreliance on officially recorded hate crimes, which have been shown to be subject to significant bias, largely because of differences in the institutional recording of hate crimes. To address these limitations, the present study uses a restricted version of the National Crime Victimization Survey (NCVS) to examine patterns in hate crime victimization based on perceived religion. Two key findings emerge. First, religiously motivated hate crime victimizations are more likely in areas with a larger Arab population. Second, these victimizations are more likely to align with official FBI religious hate crime data in those same areas. We interpret these patterns within the context of group threat and power differential frameworks, indicating that hate crime—as a behavior—may be associated with group threat theories, while societal pressures for more effective institutional responses are also likely to be stronger in those places with larger minority group populations, leading to responses that more accurately reflect actual hate crime offending.
In this hyperconnected world, misinformation spreads faster and farther than ever before. Adolescents, navigating the challenges of growing up, may be especially vulnerable to the overwhelming flow of digital information, including content on globally shared social media platforms. This study examines adolescents’ resistance to misinformation across 49 countries, using novel measures from the 2022 Program for International Student Assessment (PISA) dataset. We construct a composite index of misinformation resistance, capturing aspects such as checking accuracy, comparing sources, and sharing unverified content. Descriptive patterns show that most students compare sources and check accuracy, but many still trust or share unverified information, with notable socioeconomic status (SES) and regional disparities. Multilevel regression models show a positive, significant association between reading achievement and misinformation resistance. Importantly, the SES gap in misinformation resistance is largest among students at the bottom of the reading achievement distribution, but is not significant among those at the top of the achievement distribution. We argue that closing digital literacy gaps between advantaged and disadvantaged students requires investments in both literacy skills and awareness of misinformation.
