Navigation

You are here: Home

Beyond Text: Using AI-Generated Visual Conjoints to Study Gender and Housework Attribution

By on June 16, 2026 in Articles

Léa Pessin, Kevin Munger

Sociological Science June 16, 2026
10.15195/v13.a26

PDF (129 views)
0 Citation


Despite substantial gender convergence in education and employment, women continue to perform a disproportionate share of housework. We employ a novel visual conjoint experiment to isolate the normative mechanisms underlying this persistent inequality. Using AI-generated photorealistic images, we systematically vary the tidiness of domestic spaces, room type, source of mess, socioeconomic status, and the gender and race/ethnicity of occupants, alongside text describing couples’ employment arrangements. A quota sample of 2,994 U.S. respondents each evaluated five vignettes, yielding 14,970 observations. We find that gender effects operate primarily through responsibility attribution rather than through differential perception of messiness or anticipated social judgment. Women are assigned significantly more cleaning responsibility than men, with the gender penalty concentrated among dual-earner couples. Child-caused mess is perceived as messier than adult-caused mess yet carries reduced social consequences, suggesting that it operates as a legitimating excuse. Our findings suggest that gender equality in paid work is necessary for achieving gender equality in housework, but that it is not sufficient, and that this gap will persist absent changes in normative expectations around responsibility for housework.

Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Léa Pessin: Department of Social and Political Science, European University Institute, Fiesole, Italy.
E-mail: lea.pessin@eui.eu.

Kevin Munger: Department of Social and Political Science, European University Institute, Fiesole, Italy.
E-mail: kevin.munger@eui.eu

Acknowledgments: This work has benefited from generous feedback from the Bocconi Dondena Seminar Speaker Series and the LMU Munich Department of Sociology Research Colloquium. Pessin acknowledges funding by the European Union, under the European Research Council grant for the WeEqualize project (grant agreement no. 101117327). Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.


Supplemental Materials

Reproducibility Package: Replication materials including all data and code and documentation necessary to reproduce all empirical results reported in the article can be found at https://github.com/kmunger/Housekeeping_SocSci_Replication.


  • Citation: Pessin, Léa, Kevin Munger, 2026. “Beyond Text: Using AI-Generated Visual Conjoints to Study Gender and Housework Attribution” Sociological Science 13: 661-684.
  • Received: January 20, 2026
  • Accepted: April 14, 2026
  • Editors: Ari Adut, Elizabeth Bruch
  • DOI: 10.15195/v13.a26


0

Changing Opportunity: Rising Local Wealth Inequality and Growing Class Gaps in Income Mobility

By on June 15, 2026 in Articles

Manuel Schechtl, Florencia Torche

Sociological Science June 15, 2026
10.15195/v13.a25

PDF (76 views)
0 Citation


Recent research documents widening class gaps in intergenerational income mobility in the United States. Children from low-income families in more recent cohorts attain lower incomes than their counterparts in earlier cohorts, while no comparable decline is observed among children from high-income families. This study examines whether rising local wealth inequality contributes to this growing class divide in mobility. To do so, it combines newly published estimates of local wealth inequality from GEOWEALTH-US with cohort-based measures of upward mobility from Opportunity Insights. First-difference models reveal a consistent negative association between rising local wealth inequality and declining upward income mobility for children from low-income families, but no comparable association for their high-income peers. These associations are robust to economic and demographic changes, including, critically, changes in income inequality. A decomposition exercise suggests that rising local wealth inequality accounts for roughly one-fifth of the observed increase in class gaps in mobility. Together, the findings identify local wealth inequality as a central dimension of stratification shaping children’s economic opportunities above and beyond income inequality.

Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Manuel Schechtl: University of North Carolina at Chapel Hill.
E-mail: schechtl@unc.edu

Florencia Torche: Princeton University.
E-mail: ftorche@princeton.edu

Acknowledgments: This research was funded by the Volkswagen Foundation.


Supplemental Materials

Reproducibility Package All code necessary to replicate
this study is available in an OSF repository at: https://osf.io/a3vfd/


  • Citation: Schechtl, Manuel, Florencia Torche. 2026. “Changing Opportunity: Rising Local Wealth Inequality and Growing Class Gaps in Income Mobility” Sociological Science 13: 645-660.
  • Received: March 13, 2026
  • Accepted: April 27, 2026
  • Editors: Stephen Vaisey, Herman van de Werfhorst
  • DOI: 10.15195/v13.a25


0

Declining Inequality and Persistent Inequality Structures

By on June 10, 2026 in Articles

Soohyun Roh, Nathan Wilmers

Sociological Science June 10, 2026
10.15195/v13.a24

PDF (225 views)
0 Citation


Prior research finds that rising labor market inequality in the United States was abetted by structural changes in the economy: a consolidation of occupation and organizational bases of advantage; rising within-job inequality; and declining pay and employment in middle-earning jobs. In this article, we revisit these structural changes by asking whether they have been reversed as labor market inequality fell over the last decade. Drawing on restricted-use microdata from the Occupational Employment and Wages Statistics, we find that declining inequality is due to declining inequality in occupation premiums. There has been only a small reversal of consolidation and no decrease in inequality within jobs. Low-wage jobs gained on shrinking middle-earning occupations, further eroding union, manufacturing, and public sector wage premiums. These findings demonstrate a novel configuration of labor market inequality, in which pay rose in low-wage jobs, but underlying inequality structures in the economy persisted.

Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Soohyun Roh: Sloan School of Management, MIT
E-mail: rohs@mit.edu.

Nathan Wilmers: Sloan School of Management, MIT
E-mail: wilmers@mit.edu.

Acknowledgments: Thank you for very helpful comments from the MIT Applied Microeconomics Seminar, University of Maryland Strategy Seminar, NYU Sociology Colloquium, Columbia Center for Wealth and Inequality Seminar, Russell Sage Foundation Visiting Scholar Seminar, Frankfurt School of Finance and Management Seminar, and Stockholm University Department of Economics Seminar. This research was conducted with restricted access to Bureau of Labor Statistics (BLS) data. The views expressed here do not necessarily reflect the views of the BLS or the US government. This research was funded by MIT Sloan. Please direct correspondence to wilmers@mit.edu.


Supplemental Materials

Reproducibility Package: Full replication code is available at https://osf.io/8tbwh. In June 2025, the BLS suspended researcher access to its restricted data. As such, data for the bulk of this analysis are no longer accessible for replication (or to Roh and Wilmers). If the BLS restarts its data access program, then data will be accessible through the application as a visiting researcher.


  • Citation: Roh, Soohyun, Nathan Wilmers. 2026. “Declining Inequality and Persistent Inequality Structures” Sociological Science 13: 614-644.
  • Received: October 16, 2025
  • Accepted: February 26, 2026
  • Editors: Arnout van de Rijt, Cristobal Young
  • DOI: 10.15195/v13.a24


0

Family Networks and Childcare Choices: A Predictive Machine Learning Approach

By on June 2, 2026 in Articles

Nicolás Soler, Tom Emery, Agnieszka Kanas

Sociological Science June 2, 2026
10.15195/v13.a23

PDF (315 views)
0 Citation


How first-time parents arrange childcare has critical implications for their careers and the child’s development. Previous research shows that childcare choices are shaped by family care availability, understood as an additive function of a small set of parental and grandparental characteristics. However, research on family networks suggests that care availability is rather a non-linear, non-additive function of large family networks. We compare the predictive ability of these two perspectives using a machine learning framework and register-based family network data. We find that considering how the child’s great-grandparents, aunts, uncles, and cousins shape care availability, and modeling their influence using more flexible models, provides small yet significant improvements in predictive ability, particularly among more disadvantaged parents. Predictions are driven by parents’ and grandparents’ socioeconomic characteristics, but cousins’ age and daycare use are important yet understudied predictors. Other important understudied predictors include parents’ self-employment, healthcare spending, and timing of daycare uptake.

Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Nicolás Soler: Department of Public Administration and Sociology, Erasmus University Rotterdam.
E-mail: soleralvarezmiranda@essb.eur.nl.

Tom Emery: Department of Public Administration and Sociology, Erasmus University Rotterdam.
E-mail: tom@odissei-data.nl.

Agnieszka Kanas: Department of Public Administration and Sociology, Erasmus University Rotterdam.
E-mail: kanas@essb.eur.nl.


Supplemental Materials

Reproducibility Package: Code to reproduce the results can be found at https://doi.org/10.5281/zenodo.19189668. The data are non-public microdata from Statistics Netherlands that are accessible to accredited researchers under certain conditions (see Statistics Netherlands 2026).


  • Citation: Soler, Nicolás, Tom Emery, Agnieszka Kanas, 2026. “Family Networks and Childcare Choices: A Predictive Machine Learning Approach” Sociological Science 13: 589-613.
  • Received: February 18, 2026
  • Accepted: March 23, 2026
  • Editors: Stephen Vaisey, Michael Rosenfeld
  • DOI: 10.15195/v13.a23


0

Echo Chambers Are Defined by Conflict, Not Isolation

By on May 11, 2026 in Articles

Anna Keuchenius, Petter Törnberg, Justus Uitermark

Sociological Science May 11, 2026
10.15195/v13.a22

PDF (947 views)
0 Citation


The influential “echo chamber” hypothesis suggests that social media drive polarization through a mutual reinforcement between isolation and radicalization. The existence of such echo chambers has been a central focus of academic debate, with competing studies finding ostensibly contradictory empirical evidence. This article identifies a fundamental methodological limitation of these empirical studies: they do not differentiate between negative and positive interactions. To overcome this limitation, we develop a method to extract signed network representations of Twitter/X debates using machine learning. Applying our approach to a major Dutch cultural controversy, we show that the inclusion of negative interactions provides a new empirical picture of the dynamics of online polarization. Our findings suggest that conflict, not isolation, is at the heart of polarization.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Anna Keuchenius: Sociology, University of Amsterdam
E-mail: anna@keuchenius.com.

Petter Törnberg: Computational Social Science, University of Amsterdam
E-mail: p.tornberg@uva.nl

Justus Uitermark: Human Geography Planning and International Development Studies
University of Amsterdam, E-mail: j.l.uitermark@uva.nl

Financial Disclosure: This research has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement no. 732942, project ODYCCEUS.


Supplemental Materials

Reproducibility Package: The data and code underlying this article are available as part of our replication materials available at this link: https://doi.org/10.6084/m9.figshare.30948659.v1 or the DOI: 10.6084/m9.figshare.30948659. Due to Twitter/X’s Terms of Service and ethical and legal obligations, we do not share tweet text, user IDs, or any data that could identify individuals or their stance in the debate. The dataset contains sensitive information, including political opinions, and releasing identifiable content would pose ethical risks to users and violate GDPR requirements. To support replication, we provide code, documentation, and non-identifying data sufficient to reproduce all analytical steps, with full analyses possible via rehydration of tweet IDs.


  • Citation: Keuchenius, Anna, Petter Törnberg, and Justus Uitermark. 2026. “Echo Chambers Are Defined by Conflict, Not Isolation” Sociological Science 13: 565-588
  • Received: January 8, 2025
  • Accepted: January 12, 2026
  • Editors: Arnout van de Rijt, Bart Bonikowski
  • DOI: 10.15195/v13.a22


0

How a Seemingly Innocuous and Intuitive Methodological Choice Confused a Generation of Research on Policy Responsiveness

By on May 4, 2026 in Articles

Peter K. Enns

Sociological Science May 4, 2026
10.15195/v13.a21

PDF (2623 views)
2 Citation

The finding that government policy is, “virtually unrelated to the desires of the low- and middle-income citizens” (Gilens 2005:789), is one of the most influential social science results of the last two decades. This article offers a new perspective on this finding. I show that the seemingly innocuous decision to restrict analyses to data where different income groups’ policy support differs (i.e., a preference gap exists) introduced Simpson’s paradox, leading to misleading conclusions about whose preferences policy reflects. The same concerns apply to analyses of responsiveness to men and women and to partisan groups. I also present evidence that other common approaches for evaluating policy responsiveness can produce equally misleading conclusions. These findings suggest a need to reconsider conventional wisdom about political influence. The conclusion offers methodological recommendations and discusses implications related to understanding social and economic inequality and support for populist candidates.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Peter K. Enns: Professor, Department of Government; Professor, Brooks School of Public Policy; Robert S. Harrison Director, Cornell Center for Social Sciences, Cornell University; Co-founder and Chief Data Scientist, Verasight.
E-mail: peterenns@cornell.edu.

Acknowledgments: I thank Gabrielle Sorresso and Claudia Miner for outstanding research assistance, David Bateman, Dan Butler, Scott Desposato, Derek Epp, Chris Faricy, Marty Gilens, Luke Keele, Doug Kriner, Neil Malhotra, Henry Mo, Jacob Montgomery, Tom Pepinsky, Bryn Rosenfeld, Kim Weeden, Ariel White, and Chris Wlezien for extremely helpful comments and feedback, Stephen Parry and Kenneth Tyler Wilcox from Cornell’s Statistical Consulting Unit, and the Cornell Center for Social Sciences for verifying that the data and replication code replicate the numerical results reported in the main text and online supplement to this article. Previous versions of this article were presented at the University of California San Diego Methods Workshop, the 2024 Annual Meeting of the American Political Science Association, the 2024 Annual Meeting of the Society for Political Methodology, the 25th Anniversary of the American Politics Research Group (APRG) George Rabinowitz Seminar at the University of North Carolina at Chapel Hill, and Introduction to Public Policy (PUBPOL 2301), Public Opinion (GOVT 6461), Time Series Analysis (GOVT 6089), and Comparative Political Behavior (GOVT 6594) courses at Cornell University.


Supplemental Materials

Reproducibility Package: Data and code to reproduce all numerical results are available here: https://doi.org/10.7910/DVN/9QADPC.


  • Citation: Enns, K. Peter. 2026. “How a Seemingly Innocuous and Intuitive Methodological Choice Confused a Generation of Research on Policy Responsiveness” Sociological Science 13: 528-564.
  • Received: December 15, 2025
  • Accepted: February 25, 2026
  • Editors: Arnout van de Rijt, Bart Bonikowski
  • DOI: 10.15195/v13.a21


2

Dissecting Taste Distinction: Cultural Tastes and Perceptions of Individuals’ Status and Qualities

By on April 30, 2026 in Articles

Mikkel Haderup Larsen, Mads Meier Jæger

Sociological Science April 30, 2026
10.15195/v13.a20

PDF (1132 views)
0 Citation


A rich literature in sociology argues that familiarity with legitimate culture creates favorable perceptions of individuals’ status and qualities, which in turn yield privilege. Yet, it remains unclear which tastes affect what perceptions by how much. To address these important questions, we designed a survey experiment in Denmark that “dissects” and quantifies the effect of individuals’ tastes across six taste domains (music, food, performing arts, leisure, sport, and literature) on perceptions of status and qualities. Ignoring taste domains, we find that an individual whose taste profile in general includes more legitimate tastes is perceived more favorably in terms of status and qualities but less favorably in terms of sociability. Dissecting taste distinction by domain, we find that tastes in music and food have the strongest effect on perceptions, whereas tastes in other domains have little effect. Finally, we find that the substantive (and not just statistical) effect of tastes is large with regard to perceptions of cultural sophistication and sociability but small with regard to perceptions of social rank, earnings, and respectability. Overall, our results show that not all taste domains matter equally, legitimate tastes elicit both positive and negative perceptions, and tastes are powerful signals.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Mikkel Haderup Larsen: ROCKWOOL Foundation Intervention Unit.
E-mail: mhl@rfintervention.dk.

Mads Meier Jæger: Department of Sociology, University of Copenhagen.
E-mail: mads.jaeger@samf.ku.dk.

Acknowledgments: We have presented this article at the 2024 British Journal of Sociology Conference, the 31st Nordic Sociological Association Conference, the Sixth Annual Conference on Experimental Sociology (ACES), the 2024 European Consortium for Sociological Research (ECSR) Conference, the 2024 CEPDISC Conference, and at workshops at the University of Copenhagen, the University of Iceland, and the ROCKWOOL Foundation. We thank participants at these events for helpful comments and suggestions. We gratefully acknowledge funding from the Carlsberg Foundation (grant no. CF21-325) and the ROCKWOOL Foundation (grant no. 934121). The experiment was approved by the Institutional Review Board at the Department of Sociology, University of Copenhagen (Case ID: UCPH-SAMF-SOC-2023-02).


Supplemental Materials

Reproducibility Package: If you wish to reproduce our results, you can access the data set and accompanying R code at https://tinyurl.com/yjm8f9ce. Please be aware that we provide the data set solely for the purpose of reproducing the results we present in the article. You may not use the data set for any other purpose without written consent from the authors.


  • Citation: Larsen, Mikkel Haderup, and Mads Meier Jæger. 2026. “Dissecting Taste Distinction: Cultural Tastes and Perceptions of Individuals’ Status and Qualities” Sociological Science 13: 501-527.
  • Received: October 10, 2025
  • Accepted: January 13, 2026
  • Editors: Arnout van de Rijt, Elizabeth Bruch
  • DOI: 10.15195/v13.a20


0

More Common, Less Equal: Disparities in College Internship Participation Over Time

By on April 23, 2026 in Articles

Carrie L. Shandra

Sociological Science April 23, 2026
10.15195/v13.a19

PDF (926 views)
0 Citation


Internships play a key role in the production of inequality in the U.S. labor market, yet are often unobserved in analyses of youth employment. This study makes two empirical contributions to the study of internships. First, I use nationwide data from the 1994–2017 College Senior Survey to evaluate the association between internship participation and individual and institutional markers of privilege over time, net of grades, college major, and demographic controls. Second, I test if disparities in internship participation narrowed, persisted, or widened over three decades, independent of controls. Results indicate that internship participation has more than doubled since the mid-1990s, marking a period of rapid internship expansion, but these gains were not equal for all students. Those with the highest family income, with college-educated parents, from the most selective colleges, and from private colleges were consistently more likely to participate. Further, these internship participation gaps persisted or widened over time. Findings indicate that internships follow similar patterns of stratification as formal credentials, despite their more ambiguous nature. They also suggest that persistent barriers to internship participation remain for less-privileged students.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Carrie L. Shandra, State University of New York at Stony Brook
E-mail: Carrie.Shandra@stonybrook.edu.

Acknowledgments: This research was funded by a Presidential Grant and a Visiting Scholar award from the Russell Sage Foundation. Assistance with data acquisition and support was received from Ellen Stolzenberg and the Higher Education Research Institute. This study benefited from conversations with Sarah Damaske, Jessica Halliday Hardie, Dara Shifrer, Angela Frederick, Rachel Fish, Jennifer Pearson, and Linda Blum. All errors are my own.


Supplemental Materials

Reproducibility Package: This article uses proprietary data from the Higher Education Research Institute. Code used for data processing and analysis is available at https://osf.io/n39h2.


  • Citation: Shandra, Carrie L. 2026. “More Common, Less Equal: Disparities in College Internship Participation Over Time” Sociological Science 13:476-500.
  • Received: February 11, 2026
  • Accepted: March 2, 2026
  • Editors: Ari Adut, Cristobal Young
  • DOI: 10.15195/v13.a19


0

Fathers’ Military Service and Children’s College Attainment

By on April 20, 2026 in Articles

Paula Fomby, Patricia van Hissenhoven Flórez

Sociological Science April 20, 2026
10.15195/v13.a18

PDF (800 views)
0 Citation


Men’s early adult experiences shape the life chances of their future children. For Black men in the United States, systemic exclusion from educational and labor market opportunity has long constrained intergenerational mobility. We examine whether military service alters this trajectory, drawing on the US Panel Study of Income Dynamics (1968–2023, N=7,808 father–child pairs) to investigate college completion among adult children whose fathers were born between 1920 and 1976. Since the mid-twentieth century, the Armed Forces have offered Black men racial integration, occupational advancement, economic stability, and educational benefits that were less available in civilian society. Black fathers’ military service increased children’s probability of earning a bachelor’s degree by 53 percent compared with children of Black nonveterans, with larger differences when fathers served before the transition to an all-volunteer force. Gains were attributable to GI Bill benefit receipt and diversion out of limited civilian opportunity in early adulthood. White fathers’ veteran status conferred no educational advantage to their children, reflecting different counterfactuals: service provided greater relative benefits when the alternative was a racially closed civilian opportunity structure rather than an open one.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Paula Fomby, University of Pennsylvania
E-mail: pfomby@sas.upenn.edu.

Patricia van Hissenhoven Flórez, University of Pennsylvania
E-mail: vpatr@sas.upenn.edu.

Acknowledgments: We are grateful to Angela Dixon, Megan Reed, Christine Schwartz, and participants in seminars at Emory University, University of Maryland, and University of Wisconsin for comments on earlier versions of this manuscript and to the University of Pennsylvania Population Studies Center and its NICHD Center Grant (P2C HD044964) for administrative and computing support. All errors and omissions are the responsibility of the authors.


Supplemental Materials

Reproducibility Package: Reproducibility package available at: https://www.icpsr.umich.edu/sites/psid/view/studies/303003


  • Citation: Fomby, Paula, Patricia van Hissenhoven Flórez. 2026. “Fathers’ Military Service and Children’s College Attainment” Sociological Science 13: 441-475.
  • Received: January 6, 2026
  • Accepted: March 9, 2026
  • Editors: Stephen Vaisey, Ellis Monk
  • DOI: 10.15195/v13.a18


0

Making Progress in the Chicago Police Department, 1862–2024

By on April 16, 2026 in Articles

Tony Cheng, Johann Koehler

Sociological Science April 16, 2026
10.15195/v13.a17

PDF (1091 views)
0 Citation


Claims to have made progress are a mainstay of organizational reputation management. However, confusing and contradictory performance expectations can make progress difficult to locate among a police department’s priorities. A case study of the Chicago Police Department’s front-facing pronouncements over more than a century and a half clarifies how a bureaucracy works, stretches, and repackages “progress” to resolve those confusions and contradictions. We find that progress claims featured more prominently and fervently during moments when the department had reason to believe its legitimacy was threatened. Within that general pattern, we also find specific patterns in the form that progress claims took. We observe the stable reliance on two techniques to gesture toward progress the police either promised to enact or that it claimed it had already delivered: the police shifted goalposts by cycling through inconsistent measures of favorable performance from one year to the next, and they drummed crises to dramatize the obstacles that favorable performance required them to overcome. By showing how both techniques reinforced one another, we clarify how a police department “makes” progress.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Tony Cheng: Department of Sociology, Duke University.
E-mail: tony.cheng@duke.edu.

Johann Koehler: Department of Social Policy, London School of Economics and Political Science.
E-mail: j.koehler@lse.ac.uk.

Acknowledgments: We thank the London School of Economics and Political Science Phelan US Centre for support that made this research possible; to Vani Kant and Maryam Auwalu for excellent research assistance; to Eric Monson, Lauren Nichols, and the Duke Center for Data & Visualization Sciences for advice about representing the findings; and to Calvin Morrill, Tim Newburn, Coretta Phillips, Gil Rothschild Elyassi, Tobias Smith, and participants in the LSE’s Criminal Justice Forum for comments that sharpened the analysis. Direct correspondence to Tony Cheng, Department of Sociology, Duke University, 417 Chapel Drive, Box 90088, Reuben-Cooke Building Room 258, Durham, NC 27708.



Reproducibility Package: A memo describing the historical data, coding procedures, and analytic workflow used in this study is available here: https://osf.io/hdfp6/overview.

  • Citation: Cheng, Tony, and Johann Koehler. 2026. “Making Progress in the Chicago Police Department, 1862–2024” Sociological Science 13: 408-440.
  • Received: January 7, 2026
  • Accepted: February 17, 2026
  • Editors: Ari Adut, Kristen Schilt
  • DOI: 10.15195/v13.a17


0

Are Occupations “Bundles of Skills”? Identifying Latent Skill Profiles in the Labor Market Using Topic Modeling

By on April 13, 2026 in Articles

Marie Labussière, Thijs Bol

Sociological Science April 13, 2026
0.15195/v13.a16

PDF (1755 views)
0 Citation


Skills are considered a key determinant of workers’ labor market opportunities, especially in times of rapid technological change. However, existing research rarely conceptualizes and measures skills in their own right, instead relying on occupations as a proxy. How does this limit our understanding of the labor market structure and of wage inequality? In this article, we leverage a unique dataset of millions of online job postings in the United Kingdom to measure the skill profiles of jobs and analyze their similarity within and between occupational categories. Our data-driven approach reveals substantial discrepancies between occupational classifications and the actual skill content of jobs. We further demonstrate that job-level variation in skill content constitutes an independent source of wage inequality—one that is obscured by analyses at the occupational level. These findings challenge the conventional view of occupations as coherent bundles of skills, offering new avenues for analyzing labor market stratification.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Marie Labussière: Sciences Po, Centre for Research on Social Inequalities (CRIS).
E-mail: marie.labussiere@sciencespo.fr.

Thijs Bol: University of Amsterdam, Department of Sociology.
E-mail: t.bol@uva.nl.

Acknowledgments: We are grateful to Luisa Burchartz, Viktor Decker, Thomas A. DiPrete, Fenella Fleischmann, Andreas Haupt, and Wouter Schakel for their helpful feedback on earlier drafts of this manuscript. This research was presented at the 2024 ISA RC28 Spring Meeting, the 2025 TASKS VII Conference, and workshops of the Institutions, Inequalities and Life Courses (IIL) research group at the University of Amsterdam, the Sciences Po Center for Research on Social Inequalities (CRIS), the Center for Research in Economics and Statistics (CREST), and the Federal Institute for Vocational Education and Training (BIBB); we thank participants for their constructive discussions. Marie Labussière gratefully acknowledges Pierre Alquier and Matteo Amestoy for their technical advice.

Funding: This work was supported by the ERC starting grant from School to Career: Towards a Career Perspective on the Labor Market Returns to Education (CAREER) (ID: 950189).


Supplemental Materials

Reproducibility Package: All code necessary to reproduce the results reported in this article is publicly available in a replication package hosted on GitHub (https://github.com/mlabussiere/Occupations-bundles-of-skills.git). The online supplement also contains additional information on the data, methods, and robustness checks. The data are subject to access restrictions and cannot be shared publicly.


  • Citation: Labussière, Marie, Thijs Bol. 2026. “Are Occupations “Bundles of Skills”? Identifying Latent Skill Profiles in the Labor Market Using Topic Modeling” Sociological Science 13: 362-407.
  • Received: December 9, 2025
  • Accepted: March 2, 2026
  • Editors: Ari Adut, Vincent Buskens
  • DOI: 10.15195/v13.a16


0

Socio-Economic Advancement and Long-Term Trends in the Gender Gap in Early Career Occupational Status in France 1860–1960

By on April 6, 2026 in Articles

Wiebke Schulz, Ineke Maas, Marco H.D. van Leeuwen

Sociological Science April 6, 2026
10.15195/v13.a15

PDF (657 views)
0 Citation


The stark reduction in gender inequality on the labor market is one of the most profound social changes over the past century. However, little is known about the development of the gender gap in occupational status. This study provides new evidence on the gap in occupational status early in men’s and women’s careers during a phase of rapid socio-economic change. We use an exceptionally rich dataset that combines French marriage certificates containing data on almost 50,000 brides and grooms from 1860 to 1960 with time-varying data on socio-economic advancement for the hundred French departments. From 1910 onwards, the gender gap in occupational status at marriage declined. Around 1940, the gap turned around in favor of women. As expected, labor market opportunities as well as technological development were associated with a reduction of the gender gap in status. Reaching gender equity, however, depends on a specific interplay of socio-economic forces. Technological developments only reduce the gender status gap when paired with expanded occupational opportunities for women.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Wiebke Schulz: Universität Bremen.
E-mail: wschulz@uni-bremen.de.

Ineke Maas: Utrecht University and VU University Amsterdam.
E-mail: i.maas@uu.nl.

Marco H. D. van Leeuwen: Utrecht University.
E-mail: M.H.D.vanLeeuwen@uu.nl.

Acknowledgments: Versions of this article were presented at WZB Berlin and the University of Bremen. Thanks to Lena Hipp, Lara Minkus, and Philipp M. Lersch for helpful feedback on past drafts of this article. We thank Ruta Daktariunaite and Henri Breuer for excellent research assistance. This study received funding as part of an Advanced Investigator Grant from the European Research Council (TowardsOpenSocieties) awarded to Marco H.D. van Leeuwen.


Supplemental Materials

Reproducibility Package: Code necessary to reproduce the results is available at: osf.io/6dsnt. The data used in this study are available in accordance with the access conditions specified on the respective websites: the TRA data were supplied to us by the Institut National de la Recherche Agronomique, MONA, that subsequently merged with the IRSTEA (Institut national de recherche en
sciences et technologies pour l’environnement et l’agriculture) to create the INRAE (Institut national de recherche pour l’agriculture, l’alimentation et l’environnement), https://www.inrae.fr/. The TRA data are now available from the INED (Institut national d’études démographiques): https://tra.site.ined.fr/en/databases/getting-the-data/


  • Citation: Schulz, Wiebke, Ineke Maas, and Marco H. D. van Leeuwen. 2026. “Socio-Economic Advancement and Long-Term Trends in the Gender Gap in Early Career Occupational Status in France 1860–1960” Sociological Science 13: 332-361.
  • Received: November 19, 2025
  • Accepted: February 17, 2026
  • Editors: Ari Adut, Jeremy Freese
  • DOI: 10.15195/v13.a15


0

Jargonization, Language Development, and Team Performance

By on April 2, 2026 in Articles

Ray E. Reagans, Ronald S. Burt, Donald D. Liu

Sociological Science April 2, 2026
10.15195/v13.a14

PDF (697 views)
0 Citation


The emergence of team-specific vocabulary and language (“team jargon”) is a natural con- sequence of sustained, knowledge-intensive work. We examine how jargonization—the emergence of specialized shorthand—affects both the speed of language development and its implications for team performance. We argue that the explicit and mutually understood nature of team jargon reduces ambiguity, thereby facilitating language development and minimizing misunderstandings that could otherwise hinder coordination. Empirical analysis of language formation among newly formed teams assigned a symbol identification task supports this argument. We operationalize jargonization as the proportion of content words in team communications. Our findings indicate that as jargonization increases, the relationship between experience and language development strengthens, and the positive language effect on team accuracy increases in magnitude.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Ray E. Reagans: MIT.
E-mail: rreagans@mit.edu.

Ronald S. Burt: University of Chicago and Bocconi University.
E-mail: rburt@uchicago.edu.

Donald D. Liu
E-mail: donald.liu7@gmail.com.

Acknowledgments: This research was supported by the University of Chicago Booth School of Business and the MIT Sloan School of Management, which provided funding to transition our earlier experiment to the Empirica platform and to conduct the experimental trials reported here. Under the supervision of the two lead authors, the third author implemented the platform transition and managed the experiment. We are grateful to Linda Argote for her thoughtful comments on the manuscript. Please direct all correspondence to Ray Reagans.


Supplemental Materials

Reproducibility Package: A replication package has been deposited to OpenICPSR (https://www.openicpsr.org/openicpsr/project/239292/version/V1/view) that contains code and data required to reproduce the results presented in the article.


  • Citation: Reagans, Ray E., Ronald S. Burt, Donald D. Liu. 2026. “Jargonization, Language Development, and Team Performance” Sociological Science 13: 314-331.
  • Received: August 4, 2025
  • Accepted: October 30, 2025
  • Editors: Ari Adut, Peter Bearman
  • DOI: 10.15195/v13.a14


0

The Cultural and Symbolic Foundations of Status Hierarchies: A Rejoinder to Biegert, Kühhirt, and Van Lanker

By on March 25, 2026 in Articles

Peter McMahan, Eran Shor

Sociological Science March 25, 2026
10.15195/v13.a13

PDF (583 views)
0 Citation


Interpretations of high-profile status attributions like NBA awards tend to come from one of two theoretical standpoints. Rational/economic models treat status assessments as socially tainted measurements of objective quality. In contrast, symbolic/cultural models interpret such assessments as culturally situated assertions used in the determination of quality and merit. In this rejoinder to Bigert, Kühhirt, and Van Lanker’s (2026) reply to our 2024 article in Sociological Science, we reiterate the implications of these contrasting theoretical stances for the interpretation of NBA player awards. We restate our argument that the contrast between All-Star and All-NBA awards provides the theoretical traction needed to distinguish objective bias from culturally endogenous status mechanisms in cumulative status advantage. The analysis supports our broader claim that cultural/symbolic interpretations of status are better suited for explaining the endogeneity and stratification that define status contests in varied contexts.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Peter McMahan: Department of Sociology, McGill University.
E-mail: peter.mcmahan@mcgill.ca.

Eran Shor: Department of Sociology, McGill University.
E-mail: eran.shor@mcgill.ca.


  • Citation: McMahan, Peter, Eran Shor. 2026. “The Cultural and Symbolic Foundations of Status Hierarchies: A Rejoinder to Biegert, Kühhirt, and Van Lanker” Sociological Science 13: 303- 313.
  • Received: September 8, 2025
  • Accepted: September 8, 2025
  • Editors: Ari Adut
  • DOI: 10.15195/v13.a13


Original article:
Status Ambiguity and Multiplicity in the Selection of NBA Awards

Comment:
There Is Cumulative Status Bias and Status Entrenchment in NBA Awards: Comment on McMahan and Shor (2024)


0

There Is Cumulative Status Bias and Status Entrenchment in NBA Awards: Comment on McMahan and Shor (2024)

By on March 25, 2026 in Articles

Thomas Biegert, Michael Kühhirt, Wim Van Lancker

Sociological Science March 25, 2026
10.15195/v13.a12

PDF (471 views)
0 Citation


Peter McMahan and Eran Shor (MS) published an article in Sociological Science critiquing our study on cumulative status bias in NBA All-Star elections (Biegert, Kühhirt, and Van Lancker 2023). In this article, we affirm the presence of cumulative status bias in NBA Awards. Crucially, MS focus only on the accumulated component of cumulative status bias, ignoring the impact of immediately preceding status signals, which decouple quality and status. Furthermore, we identify theoretical and empirical issues with their model extensions of All-Star elections and their reapplication to All-NBA selections. (1) We deem MS’ argument for legitimate deviations between status and quality deeply problematic. (2) We argue that their inclusion of additional variables is not theoretically plausible in several instances, nor does it improve the models, which still support our findings. (3) We argue that All-NBA selections are a different application, not a better one, with no direct implications for the role of cumulative status bias in NBA All-Star elections. (4) We highlight flaws in MS’ models, such as irrelevant covariates, an indiscriminate approach to confounding and mediation, mismeasurement, and problematic post-treatment and post-outcome controls. (5) Our re-analysis confirms that, even in the All-NBA setting, previous status distinctions cumulatively bias outcomes.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.


Thomas Biegert: Department of Social Policy, London School of Economics and Political Science.
E-mail: t.biegert@lse.ac.uk.

Michael Kühhirt:
E-mail: kuehhirtm@gmail.com.

Wim Van Lancker: Center for Sociological Research, KU Leuven.
E-mail: wim.vanlancker@kuleuven.be.

Acknowledgments: We are grateful to David Brady and two anonymous reviewers for helpful feedback and suggestions.

Supplemental Material

Reproducibility Package: Reproduction package available at https://osf.io/t4n75/.

  • Citation: Thomas Biegert, Michael Kühhirt, Wim Van Lancker. 2025. “There Is Cumulative Status Bias and Status Entrenchment in NBA Awards: Comment on McMahan and Shor (2024)” Sociological Science 13: 287-302.
  • Received: February 24, 2025
  • Accepted: April 18, 2025
  • Editors: Ari Adut, Ray Reagans
  • DOI: 10.15195/v13.a12


Status Ambiguity and Multiplicity in the Selection of NBA Awards

0

The Effect of the Texas Migrant Busing Program on the 2024 U.S. Presidential Election

By on March 10, 2026 in Articles

William Scarborough, Ronald Kwon, David Brady

Sociological Science March 10, 2026
10.15195/v13.a11

Abstract

PDF (1350 views)
0 Citation


From 2022 to 2024, Texas transported more than 100,000 migrants from the U.S.–Mexico border to six cities led by Democratic mayors, creating a unique migration shock far from the border. We use county-level data to estimate the program’s effects on presidential elections. Comparing two elections prior to the program (2016–2020) with one after (2024), we find that the busing program increased Trump’s vote share by more than three percentage points in treated counties. These effects are robust to alternative analyses. To explore mechanisms further, we analyze individual-level data from AP VoteCast. The increase in Trump’s vote share in places receiving buses was driven by swing voters and elevated Republican turnout. Swing voters in busing destinations were moved to Trump by amplified concerns with crime, whereas Republican turnout was linked to heightened concerns over immigration. Our findings highlight the enduring power of minority threat and the growing role of subnational immigration policies.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

William Scarborough: Department of Sociology, University of North Texas.
E-mail: William.Scarborough@unt.edu

Ronald Kwon: Department of Sociology, University of North Texas.
E-mail: Ronald.Kwon@unt.edu

David Brady: Department of Public Policy and Management, University of Southern California.
E-mail: Bradyd@usc.edu

Acknowledgments: This project benefited from support from the Department of Sociology at the University of North Texas. The authors thank editor Ari Adut and the anonymous reviewers at Sociological Science for their helpful feedback in advancing the article.

Supplemental Materials

Reproducibility Package: Replication code for this article can be accessed here: https://osf.io/xfsk6/overview?view_only=bfada169cd534edebc22fea0edb56064

  • Citation: Scarborough, William, Ronald Kwon, David Brady 2026. “The Effect of the Texas Migrant Busing Program on the 2024 U.S. Presidential Election” Sociological Science 13: 273-287.
  • Received: September 23, 2025
  • Accepted: January 7, 2026
  • Editors: Ari Adut, Bart Bonikowski
  • DOI: 10.15195/v13.a11

0

Is College Really “the” Equalizer? New Evidence Addressing Unobserved Selection

By on March 3, 2026 in Articles

Haowen Zheng, Robert Andersen, Anders Holm, Kristian Bernt Karlson

Sociological Science March 3, 2026
10.15195/v13.a10

Abstract

PDF (2203 views)
0 Citation


Influential research shows that college graduates achieve similar labor market outcomes regardless of socioeconomic origin, leading to the view that a college degree is a “great equalizer.” Still, other evidence suggests that family background continues to shape labor market outcomes long after graduation, implying that college’s equalizing effect may largely reflect the characteristics of those who pursue higher education. However, the role of unobserved selection into college has rarely been examined. After formally illustrating how this unobserved selection can bias estimates of the college effect, we present new analyses that correct for this bias using an instrumental-variable approach on white male respondents in the 1979 cohort of the National Longitudinal Survey of Youth. The selection-corrected results suggest that intergenerational mobility is similar among college graduates and nongraduates. Although college yields substantial returns for all, these returns do not differ by family background. We conclude that for higher education to serve as a true equalizer, it must become both less selective and more accessible to students from disadvantaged backgrounds.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Haowen Zheng: Stone Center for Inequality Dynamics, University of Michigan.
E-mail: zhenghw@umich.edu

Robert Andersen: Ivey Business School, Western University.
E-mail: bob.andersen@ivey.ca

Anders Holm: Department of Sociology, Western University.
E-mail: aholm@uwo.ca

Kristian Bernt Karlson: Department of Sociology, University of Copenhagen.
E-mail: kbk@soc.ku.dk

Acknowledgments: The authors thank Richard Breen, Wenhao Jiang, Robert Manduca, Kim Weeden, Ang Yu, and participants at PAA 2025, the Sociological Science 2025 Conference, and ASA 2025 for their helpful comments and support. This research was conducted with restricted access to Bureau of Labor Statistics (BLS) data. We thank the staff at BLS and NORC at the University of Chicago for their assistance with accessing restricted data. The views expressed here do not necessarily reflect the views of the BLS.

Supplemental Materials

Reproducibility Package: A replication package is available at https://osf.io/ne23f/. It includes all code used for data cleaning and analysis as well as a cleaned data set derived from the public-use NLSY79 data. Part of the analysis relies on restricted geographic data obtained through a data contract with the BLS (see https://www.bls.gov/nls/request-restricted-data/nlsy-geocode-data.htm). This is not included in the replication package but can be accessed through a BLS application. The instrumental variables were drawn from the replication package of Carneiro, Heckman, and Vytlacil (2011) (see https://www.openicpsr.org/openicpsr/project/112467/ version/V1/view).

  • Citation: Zheng, Haowen, Robert Andersen, Anders Holm, and Kristian Bernt Karlson. 2026. “Is College Really “the” Equalizer? New Evidence Addressing Unobserved Selection” Sociological Science 13: 242-272.
  • Received: October 6, 2025
  • Accepted: December 12, 2025
  • Editors: Arnout van de Rijt, Jeremy Freese
  • DOI: 10.15195/v13.a10

0

Early Childhood Investments and Women’s Work Outcomes across the Life Course

By on February 24, 2026 in Articles

Vida Maralani, Camille Portier, Berkay Özcan

Sociological Science February 24, 2026
10.15195/v13.a9

Abstract

PDF (944 views)
0 Citation


This study investigates variability in women’s experiences balancing work and family, focusing on the association between early childhood investments and work trajectories. Using longitudinal data and event study models, we examine work participation from two years before to 10 years after first birth across different early childhood investment levels. Although sustained intensive investment is associated with the largest reduction in paid work, the relationship between child investment and work outcomes does not follow a simple “more investment, less work” pattern. Instead, investment intensity and duration both shape work trajectories. Women with more intensive short-term practices or moderate longer-term ones work at similar levels as women making lower investments. Patterns also differ by work outcome: not working is most differentiated by sustained intensive child investment, whereas hours worked are similar across a range of investment levels. Finally, women with constrained family resources consistently work more than those married to college-educated spouses.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Vida Maralani: Cornell University.
E-mail: vida.maralani@cornell.edu.

Camille Portier: European University Institute.
E-mail: camille.portier@eui.eu.

Berkay Özcan: New York University Abu Dhabi.
E-mail: berkay.ozcan@nyu.edu.

Acknowledgments: We thank Isadora Milanez, Douglas McKee, Douglas Miller, Samuel Stabler, Kim Weeden, Kelly Musick, Patrick Ishizuka, Stephen Jenkins, Peter Rich, Lucinda Platt, Seth Sanders, Duncan Thomas, Zhipeng Zhou, and Alvaro Padilla Pozo for their valuable feedback and support on this project. We are grateful for research support from the Cornell Center on the Study of Inequality. After completing the study and drafting this manuscript, we used ChatGPT (OpenAI) to check grammar and clarity in several sections of dense prose.

Supplemental Materials

Reproducibility Package: Replication code for this article can be accessed here: https://osf.io/j8ymw/overview.

  • Citation: Maralani, Vida, Camille Portier, and Berkay Özcan. 2026. “Early Childhood Investments and Women’s Work Outcomes across the Life Course” Sociolog- ical Science 13: 214-241.
  • Received: August 31, 2025
  • Accepted: January 13, 2026
  • Editors: Ari Adut, Maria Abascal
  • DOI: 10.15195/v13.a9

0
SiteLock