Randomized paired comparisons (RPC) for social values have various advantages over a matrix format of multiple items; however, their use cannot exhaust all possible pairs if there are too many items to compare one-to-one. This article proposes (1) applying a dimension reduction method, structural topic modeling (STM), to RPC survey data by restructuring answers into ordered pairs to estimate latent answering patterns, (2) visualizing them into directed graphs, and (3) interpreting them as respondents’ preference structures among social values. For empirical validation, we randomly divided 920 respondents into RPC and matrix-format groups and asked about the seriousness of ten social problems. Our STM from the RPC group revealed five preference structures beyond a linear order among the 10 items, which are interpretable and incorporate statistical tests with respondents’ traits as covariates. We also discuss how to improve topic modeling with RPC and contribute to various research streams, such as cultural value networks and gamification, by pairwise wiki survey.

A series of papers uses administrative data on school students’ grades to assess whether teachers discriminate against certain demographic groups. Often, differences in teacher and test grades are regressed on student-level variables. However, it is unclear under what circumstances such an estimation strategy is valid. We conceptualize teacher bias as a direct causal effect of student-level attributes on teacher grades, fixing student ability. Standardized tests merely proxy for student ability; additionally, there may be confounders of ability and teacher grade. Accordingly, teacher bias is nonparametrically unidentified. However, we suggest substantive and parametric assumptions that ensure identification using difference-in-grades estimators. Estimators based on regression control for test grades are shown to be inconsistent even under these strong assumptions. We then develop a parametric sensitivity analysis that allows researchers to investigate the consequences of departures from critical assumptions. We illustrate our methodology using administrative data from Denmark.

Inducing semantic relations in word vector spaces and analyzing how other words or entire documents discursively engage these relations is a popular form of cultural analysis. The authors propose a reliability metric that is easily interpretable and agnostic to the type of relation. The metric, which the authors call the anchor reliability coefficient ( relco ), is found by creating an artificial document-term matrix of simulated documents that sequentially shift more of their tokens from relation-relevant anchor terms to nonanchor terms and then regressing the documents’ similarity to an induced relation on the anchor inclusion score of the documents. The authors validate the metric at the word level with both expert- and crowdsourced dictionaries and at the document level with expert-annotated social media posts. The authors also provide some heuristic baselines for assessing reliability effect sizes and null hypothesis testing.

Highlights Our study found little evidence of gender differences in health reporting. Reporting bias was negligible among the youngest White respondents, whereas White men and women aged 60 or older evaluated their health optimistically. Reporting bias among Black adults was inconsistent across age groups. Hispanic men and women tended to evaluate their health more pessimistically. Overall, individuals appear to evaluate their health more optimistically as they age.

Paradoxical to numerous theories and policies, adolescents attending schools high in student cohesion do not reliably exhibit less involvement in offending. Coleman’s arguments about status and norms and the broader literature on adolescent networks suggest that this paradox is partially attributable to high-status males having more leeway to offend when attending schools high in student cohesion, which over time reinforces their engagement in offending. Here, an initial test of this “code of cohesion” hypothesis is conducted using data from high school students in Waves I–III of the National Longitudinal Study of Adolescent to Adult Health. Adolescents’ status positions at school are identified using multiple measures of network centrality, and high-cohesion schools are identified as those high on measures of student friendship network density or school social climate. Consistent with the hypothesis, results indicate that the positive association of male adolescents’ network centrality with offending becomes increasingly evident over time when they attend high-cohesion schools and diminishes over time after exiting these schools. Among schools of low/moderate-cohesion, the association of males’ network centrality with offending is less evident and does not persist after exiting schooling whatsoever. These findings support Coleman’s arguments about status and norms enabling leeway and illuminate a key reason why efforts to reduce offending by fostering student cohesion have had limited success. Additionally, this study highlights the need for research assessing how school relational dynamics among teachers, administrators, and parents, in addition to adolescent social systems, shape adolescents’ behavioral trajectories.

This paper considers the relationship between market composition and temporal dynamics of ethno-racial stratification in mass-participatory asset bubbles. Beginning from the insight that bubbles possess the social structure of a Ponzi scheme, we first ask to what extent timing of entry is structured by racial status and associated resource disparities among participants. We then consider to what extent homophilous social cues and segregation amplify disparate rates of late-stage entry. We focus on the 2000–2008 US residential housing market, using neighborhood-level data from the Home Mortgage Disclosure Act and the Federal Housing Finance Authority. Conditional on mortgage lending terms, credit availability, and other tract-level characteristics, trends in housing purchase rates for Blacks and Hispanics diverged from those of Whites as prices neared their peaks. Residential segregation and processes of homophilous diffusion amplify price-bidding and exacerbate racial disparities in rates of late-stage entry by rendering actors less attuned to the emergent risks of buying at elevated prices, and more vulnerable to the overtures of predatory agents. The analysis links the sociology of financial markets to studies of racialized predatory inclusion by considering heterogenous market timing, a mechanism which has become increasingly salient as financialization draws more diverse actors into volatile asset markets.

Scholars of social movements and culture are increasingly interested in the dynamic relationship between online discourse and offline mobilization. Although prior work typically frames online discourse as a precursor to offline activism, emerging research on right-wing extremist (RWE) mobilization hints at a reciprocal relationship. Recognizing the group-based organizational structure of many contemporary social movements, I put forward a framework for understanding online-offline mobilization cycles by extending status theories of collective action to organizational processes and networks. An analysis of the white supremacist movement White Lives Matter provides preliminary support for the theory, with implications for wider debates in scholarship on social movements, online discourse, and organizational processes.

US Americans are divided along party lines, both in politics and personal relationships. While prior research shows a preference for dating co-partisans, it remains unclear why people prefer to date co-partisans and for whom the effects are strongest. Therefore, we examine whether perceived character quality of a potential partner, perceived similarity, and expected social approval mediate the effects of matching and non-matching political party on romantic interest and whether effects vary by gender and party combinations. We conducted an online dating experiment with US American partisans and partisan leaners aged 20–33 (N = 1,097). Respondents viewed six fictional dating profiles that randomly included ‘Democrat’, ‘Republican’, or no political information. Results show that romantic interest was largely driven by preference against out-partisans, rather than for co-partisans. Out-partisan rejection was driven by perceived similarity, followed by expected social approval and perceived character quality, while co-partisan preference was driven by perceived similarity and, to a lesser extent, perceived quality. The in-group preference was strongest for Republican men and women, while out-group rejection was strongest for Democratic women. Overall, our findings show that people use politics as a proxy for non-political characteristics. They also highlight the importance of distinguishing between in-party preference and out-party animosity.

How does inequality of educational opportunity evolve in the context of educational contraction? In Finland, educational attainment declined for cohorts born after the late 1970s, and this study analyses trends in class background inequality in the transition to upper secondary education using register data for total birth cohorts born between 1975 and 1997. Results from a Karlson-Holm-Breen (KHB) decomposition into primary (driven by educational performance) and secondary (choices net of educational performance) effects on educational inequality show different trends for females and males. Class background differences in upper secondary school choice increased steadily among females as a result of an increase in primary effects. For males, educational inequalities between the service and working classes as well as farmers decreased due to decreasing secondary effects. These results challenge traditional assumptions that changes in educational inequality operate predominantly through secondary effects and highlight the importance of examining mechanisms of educational inequality in the intersections of class background and gender.

Early school-leaving is considered a societal problem, but it is unclear which family factors explain this. In this article, we focus on three family resources that may explain early school-leaving: parent’s cognitive skills at age 12, parent’s educational attainment, and household income. By combining the Intergenerational Transmission of Skills dataset and administrative data from the Netherlands Cohort Study on Education, we follow the educational career of almost 22,000 children. Structural equation models allow us to determine the relative importance of children’s skills and family resources. Focusing on the direct effects, we find that children’s cognitive skills at age 12 and initial track placement in secondary education contribute most to the risk of early school-leaving (66 per cent of the explained variance), followed by parent’s household income (25 per cent) and parent’s educational attainment (9 per cent). However, family resources may also indirectly explain early school-leaving by children’s cognitive skills at age 12 and their initial track placement. Here, parent’s cognitive skills at age 12 play the most important role. If we take that into account, we find that the total effect of parental resources increases, with parents’ cognitive skills at age 12 having the strongest contribution (18 per cent), immediately followed by household income (14 per cent) and parental educational attainment (11 per cent). These results demonstrate how important family resources are for protecting against early school-leaving, both directly and indirectly.