Navigation

You are here: Home

How Do (Human) Child Welfare Workers Respond to Machine-Generated Risk Scores?

By on January 6, 2026 in Articles

Martin Eiermann, Maria Fitzpatrick, Katharine Sadowski, Christopher Wildeman

Sociological Science January 6, 2026
10.15195/v13.a1

Abstract

PDF (1007 views)
0 Citation


Algorithmic risk scoring tools have been widely incorporated into governmental decision making, yet little is known about how human decision makers interact with machine-generated risk scores at the street level. We examined such human–machine interactions in the child welfare system, a high-stakes setting where caseworkers ascertain whether government interventions in family life are warranted. Using novel data—verbatim transcripts of caseworker discussions—we found that decision makers: (1) disregarded scores in the middle of the distribution while paying attention to extremely high or low risk scores and (2) rationalized divergences between human decisions and machine-generated scores by highlighting the algorithm’s overemphasis on historical data and specific risk factors and its lack of contextual knowledge. This meant that caseworkers were unlikely to modify their decisions so that they aligned with risk scores. However, we did not find evidence of principled resistance to algorithmic tools. Our findings advance research on such tools by specifying how human perceptions of the utility and limitations of novel technologies shape discretionary decision making by state officials; and they help to explain their uneven and potentially modest impact on the bureaucratic management of social vulnerability.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Martin Eiermann: Department of Sociology, University of Wisconsin-Madison.
E-mail: meiermann@wisc.edu.
Maria Fitzpatrick: Books School of Public Policy, Cornell University; National Bureau of Economic Research.
E-mail: maria.d.fitzpatrick@cornell.edu.
Katharine Sadowski: Graduate School of Education, Stanford University.
E-mail: ksadow@stanford.edu.
Christopher Wildeman: Department of Sociology, Duke University; Sanford School of Public Policy, Duke University; ROCKWOOL Foundation Research Unit.
E-mail: christopher.wildeman@duke.edu.

Acknowledgments: The authors are grateful to Ruby Richards and Nicole Adams for feedback on earlier drafts of this manuscript and the Douglas County Department of Human Services for providing data throughout this project.

No supplemental materials.

Reproducibility Package: The terms of our Data Use Agreement with the Douglas County Department of Human Services (DCDHS) legally prohibit us from sharing the original data, which are temporarily stored on a secure Cornell University research server, cannot be shared externally, and must be destroyed at the end of the agreement period. These restrictions reflect the presence of highly sensitive child welfare data in verbatim transcripts of caseworker discussions. All analysis code and documentation of qualitative coding workflows are publicly available at OSF. Researchers with questions about Douglas County Decision Aide (DCDA) data that were generated during the randomized controlled trial may contact: Ruby Richards, Director of Human Services, Douglas County (303-688-4825).

  • Citation: Eiermann, Martin, Maria Fitzpatrick, Katharine Sadowski, and Christopher Wildeman. 2025. “How Do (Human) Child Welfare Workers Re- spond to Machine-Generated Risk Scores?” Sociological Science 13: 1-21.
  • Received: September 3, 2025
  • Accepted: November 14, 2025
  • Editors: Ari Adut, Jeremy Freese
  • DOI: 10.15195/v13.a1

0

How Robust Are Country Rankings in Educational Mobility?

By on December 11, 2025 in Articles

Ely Strömberg, Per Engzell

Sociological Science December 11, 2025
10.15195/v12.a36

Abstract

PDF (1254 views)
0 Citation


We investigate the impact of analytical choices on country comparisons in intergenerational educational mobility using a multiverse approach. A literature survey gives rise to 2,880 plausible ways of measuring educational mobility, which we apply to European Social Survey data from 16 countries. Although some countries consistently appear at the top or bottom of the mobility rankings, most show substantial variation. Beyond our methodological contribution, we report two substantive findings. First, some countries often characterized as low-mobility emerge as matching or surpassing the egalitarian Nordic countries, reinforcing the view that wider mobility differences cannot be attributed solely to the education system but must be sought elsewhere, such as the labor market. Second, the choice of parameter—such as regression coefficients, correlations, or categorical measures—is the single most influential factor that shifts country rankings. As different parameters carry distinct theoretical meanings, researchers should treat parameter choice not merely as a robustness check but as an opportunity to test and refine competing theories.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Ely Strömberg: Department of Sociology, University of Amsterdam.
E-mail: e.o.stromberg@uva.nl
Per Engzell: UCL Social Research Institute, University College London; Swedish Institute for Social Research, Stockholm University.
E-mail: p.engzell@ucl.ac.uk

Acknowledgments: Per Engzell acknowledges funding from the European Research Coun- cil, grant no. 101165962 (MaMo). Earlier versions of this work were presented at the 2023 Spring Meeting of the ISA Research Committee 28 on Social Stratification and Mobility (RC28) in Paris, the 2024 Conference of the European Consortium for Sociolog- ical Research (ECSR) in Barcelona, and in seminars at the Swedish Institute for Social Research (SOFI) and the Amsterdam Institute for Social Science Research (AISSR). For comments that improved the manuscript, we thank Editor-in-Chief Arnout van de Rijt, Deputy Editor Kristian Karlson, two external reviewers, as well as Adam Altmejd, Krzysztof Czarnecki, Harry Ganzeboom, Jan Helmdag, Mike Hout, Linda Kridahl, Liliya Leopold, Silke Schneider, Edvin Syk, Max Thaning, Jens-Peter Thomsen, An- dreas Videbæk Jensen, Kim Weeden, Herman van de Werfhorst, and Daniel Wilhelm. Any errors remain our own.

Supplemental Materials

Reproducibility Package: The microdata underlying our analyses are available to download from the European Social Survey. Code necessary to reproduce the results is available at: https://doi.org/10.17605/OSF.IO/VCDSX

  • Citation: Strömberg, Ely, Per Engzell. 2025. “How Robust Are Country Rankings in Educational Mobility?” Sociological Science 12: 891-922.
  • Received: July 9, 2025
  • Accepted: October 13, 2025
  • Editors: Arnout van de Rijt, Kristian B. Karlson
  • DOI: 10.15195/v12.a36

0

The Causal Impact of Segregation on a Disparity: A Gap-Closing Approach

By on December 6, 2025 in Articles

Ian Lundberg

Sociological Science December 9, 2025
10.15195/v12.a35

Abstract

PDF (881 views)
0 Citation


Segregation—whether across schools, neighborhoods, or occupations—is regularly invoked as a cause of social and economic disparities. However, segregation is a complicated causal treatment: what do we mean when we appeal to a world in which segregation does not exist? One could take societal contexts as the unit of analysis and compare across societies with differing levels of segregation. In practice, it is more common for studies of segregation to take persons or households as the unit of analysis within a single societal context, focusing on what would happen if particular individuals were counterfactually assigned to social positions in a more equitable way. Taking this latter framework, this article shows how to study segregation as a cause. The first step is to theorize a counterfactual assignment rule: what would it mean to assign people to social positions equitably? The second step is to identify the causal effect of those social positions and simulate counterfactual outcomes. The third step is to interpret results as the impact of a unit-level (rather than society-level) intervention. A running example and empirical analysis illustrates the approach by studying the causal effect of occupational segregation on a racial health gap.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Ian Lundberg: Department of Sociology, UCLA.
E-mail: ianlundberg.org, ianlundberg@ucla.edu.

Acknowledgments: For helpful discussions and feedback relevant to this project, I thank Brandon Stewart, Matthew Salganik, Dalton Conley, Sara McLanahan, Rebecca Johnson, Gillian Slee, and participants in presentations at the Princeton Department of Sociology, the Cornell Center for the Study of Inequality, and the UCSF Department of Social and Behavioral Sciences, as well as the editor and anonymous reviewers. The author benefited from facilities and resources provided by the California Center for Population Research at UCLA (CCPR), which receives core support (P2C-HD041022) from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD). The content is solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy Shriver National Institute of Child Health & Human Development or the National Institutes of Health.

Supplemental Materials

Reproducibility Package: doi.org/10.7910/DVN/TB5Q4N.

  • Citation: Lundberg, Ian. 2025. “The Causal Impact of Segregation on a Disparity: A Gap-Closing Approach” Sociological Science 12: 871-890.
  • Received: July 15, 2025
  • Accepted: August 31, 2025
  • Editors: Arnout van de Rijt, Maria Abascal
  • DOI: 10.15195/v12.a35

0

What You Need to Know When Estimating Monthly Impact Functions: Comment on Hudde and Jacob, “There’s More in the Data!”

By on December 4, 2025 in Articles

Josef Brüderl, Ansgar Hudde, Marita Jacob

Sociological Science December 4, 2025
10.15195/v12.a34

Abstract

PDF (640 views)
0 Citation


In life course research, it is common practice to analyze the effects of life events on outcomes. This is usually done by estimating “impact functions.” To date, most studies have estimated yearly impact functions. However, Hudde and Jacob (2023) (hereafter H&J) pointed out that most panel data sets include information on the month of events. Consequently, they proposed exploiting this information by estimating monthly impact functions. In this adversarial collaboration, we address two issues regarding H&J’s work. First, H&J did not provide sufficient guidance on how to estimate monthly impact functions. We will provide a step-by-step description of how to do so. Second, the procedure H&J proposed for smoothing monthly estimates produces confidence intervals (CIs) that are likely too narrow. This can lead to misleading conclusions. Therefore, we suggest using more appropriate bootstrapped CIs.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Josef Brüderl: Department of Sociology, LMU Munich. E-mail: bruederl@lmu.de
Ansgar Hudde: Department of Sociology and Social Psychology, University of Cologne.
E-mail: hudde@wiso.uni-koeln.de
Marita Jacob: Department of Sociology and Social Psychology, University of Cologne.
E-mail: marita.jacob@uni-koeln.de

Acknowledgments: We thank Katrin Auspurg for her helpful comments. This article uses data from the German Family Panel pairfam, coordinated by Josef Brüderl, Sonja Drobniˇc, Karsten Hank, Johannes Huinink, Bernhard Nauck, Franz J. Neyer, and Sabine Walper. From 2004 to 2022, pairfam was funded as a priority program and a long-term project by the German Research Foundation (DFG).


Reproducibility Package: Stata replication code is available on the Open Science Framework (OSF), https://osf.io/kx9ne/ (file: “Monthly Impact Functions-Replication File.zip”). The replication file includes the prepared pairfam data that we used for all of our analyses. If you would like to reproduce our data preparation (also included in the replication file), you can order the pairfam data at https://www.pairfam.de/en/data/data-access/.

  • Citation: Brüderl, Josef, Ansgar Hudde, Marita Jacob. 2025. “What You Need to Know When Estimating Monthly Impact Functions: Comment on Hudde and Jacob, “There’s More in the Data!”” Sociological Science 12: 862-870.
  • Received: May 16, 2025
  • Accepted: August 31, 2025
  • Editors: Arnout van de Rijt, Kristian B. Karlson
  • DOI: 10.15195/v12.a34

0

Pathways to Independence: The Dynamics of Parental Support in the Transition to Adulthood

By on November 25, 2025 in Articles

Ramina Sotoudeh, Ginevra Floridi

Sociological Science November 25, 2025
10.15195/v12.a33

Abstract

PDF (1324 views)
0 Citation


In the United States, the financial and co-residential dependence of young adults on parents has increased for decades. This study provides the first comprehensive analysis of economic support trajectories, their contextual, family, and individual determinants, and temporal relation to other transition to adulthood milestones. Using data from the Panel Study of Income Dynamics’ Transition to Adulthood Study (2005–2021), we identify trajectories of financial and co-residential support between ages 18 and 28 and relate them to economic and partnership trajectories and events. We study how macro-economic crises (the Great Recession and COVID-19), family characteristics, and individual traits within sibships predict trajectory membership. We find three distinct pathways: first, prolonged education and financial support are more common among advantaged families and, within siblings, among those exposed to the Great Recession. Second, early employment and prolonged co-residence are the most prevalent among disadvantaged families and children. Third, economic independence through marriage is most common among white people living outside metropolitan areas.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Ramina Sotoudeh: Department of Sociology, Yale University and Department of Political
and Social Sciences, European University Institute.
E-mail: ramina.sotoudeh@yale.edu.
Ginevra Floridi: School of Social and Political Science, University of Edinburgh.
E-mail: Ginevra.Floridi@ed.ac.uk.

Acknowledgments: We acknowledge funding from British Academy/Leverhulme Small Research Grant SRG2324\240432. Versions of this article were presented at the European Population Conference in Edinburgh, Population Association of America Meeting in Washington DC, and PopDays in Cagliari. We are grateful for the helpful feedback that we received. We thank our parents for their economic and non-economic support throughout our lives and the U.S. highway system for providing us with ample time to discuss our ideas for this article.

Supplemental Materials

Reproducibility Package: Replication code for this article can be accessed here: https://github.com/raminasotoudeh/pathways_to_independence/tree/main

  • Citation: Sotoudeh, Ramina, Ginevra Floridi. 2025. “Pathways to Independence: The Dynamics of Parental Support in the Transition to Adulthood” Sociological Science 12: 833-861.
  • Received: July 9, 2025
  • Accepted: August 25, 2025
  • Editors: Arnout van de Rijt, Michael Rosenfeld
  • DOI: 10.15195/v12.a33

0

Public Support for the Legalization of Undocumented Immigrants during the 2016 Presidential Campaign

By on November 21, 2025 in Articles

Mariano Sana

Sociological Science November 21, 2025
10.15195/v12.a32

Abstract

PDF (1370 views)
0 Citation


I investigate whether the political ascent of Donald Trump, an adamant immigration restrictionist, during the 2016 presidential campaign was accompanied by decreasing support for the legalization of undocumented immigrants. Compiling survey data from 2012 to 2016, I show consistent support for legalization throughout the period. However, support was on the decline until Trump entered the presidential race in June 2015, rising thereafter. I use two Pew Research Center surveys, fielded in May 2015 and October 2016, to document that the increase in support for legalization was spearheaded by females, suburban residents, and self-identified Democrats. No demographic group, however defined, recorded a significant decline in their support for legalization. The political ascent of Donald Trump between mid-2015 and the presidential election of November 2016 was not associated with a decline in support for the legalization of undocumented immigrants but the opposite, consistent with similar trends recorded in Europe following the rise of right-wing parties. I discuss the implications of these findings for research on immigration attitudes.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Mariano Sana: Department of Sociology, Vanderbilt University.
E-mail: mariano.sana@vanderbilt.edu.

Acknowledgments: I benefited from comments and suggestions from Guy Stecklov, Jenny Trinitapoli, and Alex Weinreb as well as those of the editor and anonymous reviewers. My gratitude also goes to Yu-Ri Kim and Alyssa Davis for their research assistance.


Reproducibility Package: Data and code necessary for full replication are publicly available here: https://www.openicpsr.org/openicpsr/project/238445/version/V1/view. Original raw data were downloaded from the Roper iPoll database managed by the Public Opinion Research Archive at Cornell University (https://ropercenter.cornell.edu/ipoll/) and from the Pew Research Center (https://www.pewresearch.org/tools-and-datasets/).

  • Citation: Sana, Mariano. 2025. “Public Support for the Legalization of Undocumented Immigrants during the 2016 Presidential Campaign” Sociological Science 12: 804-832.
  • Received: August 18, 2025
  • Accepted: October 9, 2025
  • Editors: Ari Adut, Maria Abascal
  • DOI: 10.15195/v12.a32

0

The Hardcore Brokers: Core-Periphery Structure and Political Representation in Denmark’s Corporate Elite Network

By on November 18, 2025 in Articles

Lasse Folke Henriksen, Jacob Aagard Lunding, Christoph Houman Ellersgaard, Anton Grau Larsen

Sociological Science November 18, 2025
10.15195/v12.a31

Abstract

PDF (978 views)
0 Citation


Who represents the corporate elite in democratic governance? In his seminal work on the corporate “inner circle,” Useem (1986) studied three network-related mechanisms from corporate interlocks that together shaped the ideology and political organization of American and British corporate elites during the postwar era in crucial ways: corporate brokerage, elite social cohesion, and network centrality. Subsequent research has found similar dynamics at play across a variety of democratic capitalist societies. However, all existing studies on corporate elite representation in democratic governance rest on analyses of the top ranks at very large corporations. We cast a wider net. Analyzing new population data on all members of corporate boards in the Danish economy (∼200,000 directors in ∼120,000 boards), we locate ∼1,500 directors who operate as brokers between local corporate networks and measure their network coreness using k-core detection. We find a highly connected network core of ∼275 directors, half of whom are affiliated with smaller companies or subsidiaries and then document the power of director coreness in predicting government committee attendance, a key form of political representation in Denmark’s social-corporatist model of governance. We find a large political premium for directors in very large companies but show that within the network core the gap between directors of smaller and large companies is closed, suggesting that the network core levels the playing field in corporate access to the legislative process.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Lasse Folke Henriksen: Department of Organization, Copenhagen Business School.
E-mail: lfh.ioa@cbs.dk
Jacob Aagard Lunding: Department of Social Sciences and Business, Roskilde University.
E-mail: jaagaard@ruc.dk
Christoph Houman Ellersgaard: Department of Organization, Copenhagen Business School.
E-mail: che.ioa@cbs.dk.
Anton Grau Larsen: Department of Social Sciences and Business, Roskilde University.
E-mail: agraul@ruc.dk

Acknowledgments: We would like to thank Leonard Seabrooke, Felix Bühlman, Donald Tomaskovic-Devey, Thomas Lyttelton, and Megan Neely for commenting on an earlier draft of this article. We also thank participants at the Political Economy Group Seminar at the Copenhagen Business School for engaging with an earlier draft. We are grateful to the Independent Research Fund Denmark (grant 5052-00143b), the Carlsberg Foundation (grant CF19-0175), and the Velux Foundation (grant 00048306) for generously supporting this research.

Supplemental Materials

Reproducibility Package: Because our data-use agreement prohibits direct sharing of our analytic data, we share only the analysis code here: https://github.com/JacobLunding/hardcore_brokers_replication. Interested parties may apply to Statistics Denmark (https://www.dst.dk/en/TilSalg/Forskningsservice/Dataadgang/) for access to the data (project 706264) and can run the full replication package from the folder named “/replication,” which includes all data generating steps of the analysis and the analytical code.

  • Citation: Henriksen, Lasse Folke, Jacob Aagard Lunding, Christoph Houman Ellersgaard, Anton Grau Larsen. 2025. “The Hardcore Brokers: Core-Periphery Structure and Political Representation in Denmark’s Corporate Elite Network” Sociological Science 12: 769-803.
  • Received: December 19, 2024
  • Accepted: September 24, 2025
  • Editors: Arnout van de Rijt, Michael Rosenfeld
  • DOI: 10.15195/v12.a31

0

The Intergenerational Reach of Maternal Adverse Childhood Experiences: Associations with Children’s Emotional Support and Cognitive Stimulation

By on November 6, 2025 in Articles

Lawrence Stacey, Kristi Williams

Sociological Science November 6, 2025
10.15195/v12.a30

Abstract

PDF (1311 views)
0 Citation


Adverse childhood experiences (ACEs)—such as abuse, neglect, and household dysfunction before age 18—pose substantial risks to individual health and well-being throughout life, but relatively less research has examined how ACEs are associated with parenting behaviors or children’s home environments. We use linked mother–child data from the National Longitudinal Survey of Youth 1979, a U.S. longitudinal cohort study, to investigate how maternal ACEs are associated with the emotional support and cognitive stimulation of children. Regression results demonstrate an inverse relationship between maternal ACE exposure and the degree of emotional support and cognitive stimulation in children’s home environments. Children born to mothers with four or more ACEs had, on average, 4.9 percentile-unit lower emotional support scores and 5.6 percentile-unit lower cognitive stimulation scores relative to mothers with no ACE exposure, net of maternal and child sociodemographic characteristics. Further results document the importance of emotional neglect and physical abuse, both of which were independently and negatively related to the emotional support and cognitive stimulation of children. Our article builds on a growing body of literature by documenting links between maternal ACE exposure and children’s home environments and by illuminating the lengthy intergenerational reach of parental ACEs.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Lawrence Stacey: Department of Sociology, Vanderbilt University.
E-mail: lawrence.stacey@vanderbilt.edu.
Kristi Williams: Department of Sociology, Ohio State University.
E-mail: williams.2339@osu.edu.

Acknowledgments: We wish to thank Melissa Alcaraz, John Casterline, Reanne Frank, Sarah Hayford, Jake Hays, and Alec Rhodes for feedback on an earlier draft of this article. We also wish to thank the editors of Sociological Science.

Supplemental Materials

Reproducibility Package: A replication package with instructions, data, and Stata code has been
made publicly available on the Open Science Framework (OSF): https://osf.io/gb29u/

  • Citation: Stacey, Lawrence, and Kristi Williams. 2025. “The Intergenerational Reach of Maternal
    Adverse Childhood Experiences: Associations with Children’s Emotional Support and Cognitive Stimulation” Sociological Science 12: 743-768.
  • Received: June 17, 2024
  • Accepted: April 24, 2025
  • Editors: Arnout van de Rijt, Michael Rosenfeld
  • DOI: 10.15195/v12.a30

0

Wide Social Influence and the Emergence of the Unexpected: An Empirical Test Using Spotify Data

By on October 23, 2025 in Articles

Martin Arvidsson, Peter Hedström, Marc Keuschnigg

Sociological Science October 23, 2025
10.15195/v12.a29

Abstract

PDF (2091 views)
0 Citation


Social-influence processes not only affect the rate at which behaviors spread but can also decouple adoption behavior from individual preferences, and thereby bring about unexpected collective outcomes that cannot be predicted on the basis of the initial likes and dislikes of the individuals involved. However, the conditions under which social influence can lead to such decoupling are not well understood. We identify a social-influence mechanism that widens individuals’ behavioral repertoires and breaks the link between individuals’ initial preferences and the collective outcomes they jointly bring about. We test the micro-level assumptions of the mechanism in the context of cultural choices on Spotify, combining topic modeling with traditional statistical matching to cultural change. agent-based simulation estimate peer-to-peer influence effects from digital trace data. We then use agent-based simulations to examine the macro-level consequences of “wide” social influence and its importance for explaining cultural change.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Martin Arvidsson: The Institute for Analytical Sociology, Linköping University. E-mail: martin.arvidsson@liu.se.
Peter Hedström: The Institute for Analytical Sociology, Linköping University. E-mail: peter.hedstrom@liu.se.
Marc Keuschnigg: The Institute for Analytical Sociology, Linköping University and Institute of Sociology, Leipzig University. E-mail: marc.keuschnigg@liu.se.

Acknowledgments: For helpful comments, we thank James Evans, Jacob Habinek, Mark Lutter, Arnout van de Rijt, and Duncan Watts. We are grateful for financial support from Riksbankens Jubileumsfond (M12-0301:1) and the Swedish Research Council (2013-7681, 2018-05170, 2019-00245, and 2024-01861). This research was carried out at the Swedish Excellence Center for Computational Social Science, which is also funded by the Swedish Research Council (2022-06611). Resources provided by the Swedish National Infrastructure for Computing (2024/22-1012) enabled computations.

Supplemental Materials

Reproducibility Package: A replication package has been deposited to OSF (https://osf.io/grsyt/?view_only=133867f728644ba596eb104890cb018f ) that contains code and data required to reproduce the results presented in the article.

  • Citation: Arvidsson, Martin, Peter Hedström, Marc Keuschnigg. 2025. “Wide Social Influence and the Emergence of the Unexpected: An Empirical Test Using Spotify Data.” Sociological Science 12: 715-742.
  • Received: December 16, 2024
  • Accepted: September 10, 2025
  • Editors: Ari Adut, Peter Bearman
  • DOI: 10.15195/v12.a29

0

Equalization through Deterioration: The Shrinking Gender Gap in Swedish School Grades

By on September 30, 2025 in Articles

Carina Mood

Sociological Science September 30, 2025
10.15195/v12.a27

Abstract

PDF (2094 views)
0 Citation


This article documents a surprising reversal in the long-standing gender gap in academic achievement: between 2021 and 2024, Swedish girls’ school grades declined sharply, whereas boys’ grades remained stable, narrowing the gender gap by over a third. Using full-population data on official school grades and national test scores, the analysis shows that the decline is broad based, affecting nearly all subjects and concentrated among previously high-performing girls. Changes in grading policy or long-term mental health trends are unlikely to account for the sudden downturn. Instead, the evidence points to behavioral changes in girls’ school engagement, possibly triggered by external shocks. Among the most plausible shocks are the rise of TikTok, which increased in popularity among girls just before the grade decline, and Covid-19 disruptions, which may have acted as a catalyst. Although the causes remain uncertain, the pattern signals a sociologically significant disruption, challenging assumptions of stable female academic advantage and inviting international replication.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Carina Mood: Swedish Institute for Social Research, Stockholm University, Institute for Futures Studies E-mail: carina.mood@sofi.su.se.

Acknowledgments: I am grateful to Janne Jonsson for helpful comments and Vetenskapsrådet (grant number: 2022-02036) and Riksbankens Jubileumsfond (grant number: P24-0170) for financial support.

Reproducibility Package: A reproducibility package is available at https://osf.io/y8vmk/files/osfstorage. Parts of the article are based on publicly available aggregate data. These data, links to sources, and code to reproduce the analyses are provided in the reproducibility package. Parts of the article are based on analyses of microdata accessed through Statistics Sweden’s secure MONA server within the framework of the Inequality project with ethical approval from the Swedish Ethical Review Authority (Dnr 2019-02761). Data can be accessed only after ethical approval from the Swedish Ethical Review Authority (see https://etikprovningsmyndigheten.se/en/) and after secrecy assessment of Statistics Sweden (see https://www.scb.se/en/services/ordering-data-and-statistics/ ). A list of necessary registers and variables and code to reproduce the analyses are provided in the reproducibility package.


  • Citation: Carina Mood. 2025. ““Equalization through Deterioration: The Shrinking Gender Gap in Swedish School Grades” Sociological Science 12: 670-684.
  • Received: July 30, 2025
  • Accepted: August 30, 2025
  • Editors: Arnout van de Rijt, Jeremy Freese
  • DOI: : 10.15195/v12.a27

0

The Political Economy of Optimal Taxation

By on September 18, 2025 in Articles

Elias Nosrati

Sociological Science September 18, 2025
10.15195/v12.a26

Abstract

PDF (2334 views)
0 Citation


The question of how to design socially optimal tax policies is located at the epicenter of today’s inequality debate. However, the field of optimal policy design is dominated by a distinctive brand of economic analysis which suffers from a blinkered theoretical lens and weak empirical foundations. This article offers an alternative framework for studying optimal policy design that replaces the dominant economic model’s emphasis on individual utility with a sociological emphasis on unequal life chances. Cross-national data are mobilized to study the architecture of contemporary tax systems across different institutional contexts. The dual taxation of labor and capital income coupled with the preferential treatment of socially concentrated wealth accumulation is shown to have generated sharp tax regressivity at the top of the resource distribution in all countries under consideration, from the (neo)liberal Anglosphere to social-democratic Scandinavia. Rationales for and possible designs of a progressive wealth tax, for which there is renewed international interest, are then explored in detail. A tractable formal model of optimal policy design is presented in which the net welfare effect of a tax policy reform emerges as a weighted sum of how the reform impacts aggregate life chances, inequality in life chances, and the ambient ecosphere. Under common normative and analytical assumptions, a socially optimal annual wealth tax levied on society’s most affluent—defined above a high exemption threshold—is shown to be positive and, in empirically realistic scenarios, lies upward of 10 percent. The corresponding top income tax rate exceeds 65 percent and can, on “limitarian” grounds, approach the confiscatory rate.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Elias Nosrati: : Department of Sociology and Human Geography, University of Oslo E-mail: eliasno@uio.no.

Acknowledgments: The author wishes to thank the members of the Power, Elites, and Social Class (MEK) seminar group at the Department of Sociology and Human Geography, UiO for the opportunity to present an early version of the paper and for their helpful feedback. Thank you also to Simon Szreter for reading and commenting on the penultimate draft and to the journal’s editors and reviewers for their constructive critiques.

  • Citation: Elias Nosrati. 2025. “The Political Economy of Optimal Taxation” Sociological Science 12: 634-669.
  • Received: June 30, 2025
  • Accepted: August 12, 2025
  • Editors: Arnout van de Rijt, Cristobal Young
  • DOI: : 10.15195/v12.a26

0

An Unreliable Ladder: Top–Bottom Self-Placement, Subjective Social Status, and Political Preferences

By on September 11, 2025 in Articles

Lewis Robert Anderson

Sociological Science September 11, 2025
10.15195/v12.a25

Abstract

PDF (2170 views)
0 Citation


Research on right populist support and redistribution preferences increasingly argues for the explanatory power of subjective over objective social position. However, scrutiny of a widely used measure underlying such findings is lacking. I provide a multifaceted assessment of the Top–Bottom Self-Placement question (“Topbot”), which is primarily used in the International Social Survey Programme. Through 36 cognitive interviews and analysis of secondary data sets, I evaluate Topbot’s psychometric qualities, how it is interpreted by respondents, and how far this corresponds to the (contradictory) interpretations assumed by researchers. Consonant with findings of low reliability and high, non-random non-response when a “Don’t know” option is available, the interviews highlight that Topbot is worded ambiguously, leading to varied interpretations and often puzzlement. The most frequently mentioned bases of self-placement represent economic resources. Clustering of responses in the middle is widely known; interviews reveal explanations beyond misestimation. As additionally evidenced by convergent validity analyses, interpretations of Topbot as measuring perceived income decile or subjective social status in a specifically Weberian sense are untenable, and empirical claims made on these bases should be revisited.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Lewis Robert Anderson: Department of Social Policy and Intervention, University of Oxford; Institute for New Economic Thinking, University of Oxford. E-mail: lewis.anderson@spi.ox.ac.uk.

Acknowledgments: First and foremost, I wish to thank the 36 individuals who made this research possible by participating in an interview. For their valuable comments and suggestions, I am grateful to Noah Bacine, Geoff Evans, John Goldthorpe, Henning Lohmann, Brian Nolan, Patrick Präg, David Weisstanner, various anonymous reviewers, and participants at three venues where I presented earlier versions: a 2024 meeting of the Inequality and Policy Research Group at the Department of Social Policy and Intervention, University of Oxford; the European Consortium for Political Research Joint Sessions of Workshops 2025 at Charles University, Prague; and the 2025 Sociological Science Conference at Cornell University. I also gratefully acknowledge funding from the European Research Council (Grant 856455, DINA) and the support of the Nuffield College Centre for Experimental Social Sciences (CESS) in facilitating the interviews.

Supplemental Materials

Reproducibility Package: Stata code and anonymized interview transcripts are available on the Open
Science Framework repository (https://osf.io/q4sjr/ ). The online supplement includes information about accessing the secondary data sets analyzed.

  • Citation: Anderson, Lewis Robert. 2025. “An Unreliable Ladder: Top–Bottom Self-Placement, Subjective Social Status, and Political Preferences” Sociological Science 12: 601-633.
  • Received: April 21, 2025
  • Accepted: July 15, 2025
  • Editors: Arnout van de Rijt, Kristian B. Karlson
  • DOI: 10.15195/v12.a25

0

Streaming Platforms, Filter Bubbles, and Cultural Inequalities. How Online Services Increase Consumption Diversity

By on September 4, 2025 in Articles

Samuel Coavoux, Abel Aussant

Sociological Science September 4, 2025
10.15195/v12.a24

Abstract

PDF (4617 views)
0 Citation


Do digital technologies affect diversity in cultural tastes? Digital sociologists have warned of “filter bubbles,” whereas sociologists of culture have shown that diversity in consumption is valued as a marker of upper-middle-class status. We estimate the effect of using streaming platforms on the diversity of cultural consumption using a matching technique applied to 2018 survey data from France. We find a statistically significant positive effect of using streaming platforms on the diversity of cultural consumption as well as on cosmopolitanism, on three domains, music, movies, and TV shows. The magnitude of this effect is much higher for TV shows. The study brings new evidence against the filter bubble thesis; it shows that platforms do reinforce cultural inequalities by increasing the social gap in consumption diversity. It further suggests that the effect of technology on cultural consumption might mainly operate through its impact on cultural markets rather than changes in cultural experience.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Samuel Coavoux: CREST, ENSAE, Institut Polytechnique de Paris, Paris, France. E-mail: samuel.coavoux@ensae.fr.
Abel Aussant: Sciences Po, CRIS, Paris, France. E-mail: abel.aussant@sciencespo.fr.

Acknowledgments: This article benefited greatly from comments by Quentin Mazel, Patrick Präg, Léa Pessin, and anonymous reviewers, as well as from the audiences of AFS 2023, ESA-RN05 Midterm 2023, ECSR 2023, Culture in a digital context conferences, and the CREST sociology seminar.

Supplemental Materials

Reproducibility Package: A replication package containing all scripts necessary to reproduce the results presented in the article is available at OSF. The data are available on demand from the Progedo-Adisp repository.

  • Citation: Coavoux, Samuel and Abel Aussant. 2025. “Streaming Platforms, Filter Bubbles, and Cultural Inequalities. How Online Services Increase Consumption Diversity” Sociological Science 12: 572-600.
  • Received: May 29, 2025
  • Accepted: July 6, 2025
  • Editors: Arnout van de Rijt, Bart Bonikowski
  • DOI: 10.15195/v12.a24

0

Workplace Networks and the Dynamics of Worker Organizing

By on August 28, 2025 in Articles

Hana Shepherd, Rebecca Roskill, Suresh Naidu, Adam Reich

Sociological Science August 28, 2025
10.15195/v12.a23

Abstract

PDF (1473 views)
0 Citation


A rich literature has established the importance of social networks for explaining participation in contentious politics but has typically treated networks as existing outside the awareness or influence of movement actors themselves. A separate literature has long recognized the importance of “organizing” for successful collective action but has not conceived of organizing in relation to network structure. Bridging these literatures, we develop the concept of “network-driven organizing” (NDO), where organizers allocate relational activity based on perceived social network structure. Using the case of labor organizers in a campaign at Walmart, we analyze more than 80,000 unstructured organizer field notes from almost 120 store-level campaigns between 2010 and 2015 and find that our measure of NDO is positively and robustly correlated with campaign success; going from 0 to 1 on the measure of NDO more than doubles the number of cards signed. We discuss the implications of our results in light of sociological theories of action and the practice of movement organizing.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Hana Shepherd: Sociology, Rutgers University, E-mail: hshepherd@sociology.rutgers.edu
Rebecca Roskill: E-mail: beccaroskill@gmail.com
Suresh Naidu: Economics and SIPA, Columbia University, E-mail: sn2430@columbia.edu
Adam Reich: Sociology, Columbia University, E-mail: ar3237@columbia.edu

Acknowledgments: Authorship is equal and the order is randomized. We thank OUR Walmart for sharing their data and for their time and insights, the Robert Wood Johnson Foundation, Evidence for Action Program for funding, and Jeff Jacobs, Easton Schindler, and Rachel Springer for research assistance.

Supplemental Materials

Reproducibility Package: The data used in this article are proprietary data from the organization
OUR Walmart. More information about this is provided in the Data section. All code used for data processing and analysis is available at https://osf.io/wejb5/. The researchers will make the processed and anonymized data available for replication purposes upon request and subject to review of a plan to keep the data secure and to delete after use.

  • Citation: Shepherd, Hana, Rebecca Roskill, Suresh Naidu, and Adam Reich 2025. “Workplace Networks and the Dynamics of Worker Organizing” Sociological Science 12: 537-571.
  • Received: February 20, 2025
  • Accepted: April 28, 2025
  • Editors: Arnout van de Rijt, Filiz Garip
  • DOI: 10.15195/v12.a23

0

One Sentiment, Multiple Interpretations: Contrasting Official and Popular Anti-Americanism in China

By on August 21, 2025 in Articles

Yinxian Zhang, Di Zhou

Sociological Science August 21, 2025
10.15195/v12.a22

Abstract

PDF (2366 views)
0 Citation


This study contrasts official and popular expressions of anti-Americanism in China by comparing narratives from People’s Daily and Zhihu between 2011 and 2022. Using computational and qualitative methods, we examined sentiment trends, topics, and opinions in official and popular discourses. We find that although both discourses have become increasingly negative toward the United States, they diverge significantly in specific expressions: official discourse mirrors Western liberal critiques of American social problems but attributes these issues to American democracy, whereas popular discourse blends left- and right-wing populism and blames liberal elites and capitalism for the American decline. These findings highlight both the limits of state control over public opinion and the pluralistic nature of nationalist expressions. The study also situates Chinese anti-Americanism within a global zeitgeist, discussing how populism transcends borders and shapes local political discourse in unexpected contexts.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Yinxian Zhang: Department of Sociology, CUNY Queens College. E-mail: yinxian.zhang@qc.cuny.edu.
Di Zhou: Department of Sociology, New York University. E-mail: di.zhou@nyu.edu.

Acknowledgments: This study was financially supported by the 2025 CUNY Faculty Fellowship Publication Program (FFPP) and a PSC-CUNY Research Award (68208-00 56). We are deeply grateful to Yinxian Zhang’s FFPP mentor and fellow participants— Sarah Hoiland, Cindy Bautista-Thomas, Philippe Marius, Nicole McKenna, Douglas Medina, and Prash Naidu—for their invaluable comments and suggestions on earlier drafts.

Conflicts of Interest Statement: The authors declared no potential conflicts of interest with respect to the research and publication of this article.

Author Contributions: YZ: research design, data collection, data analysis and visualization, and writing and editing. DZ: data collection, sentiment classification, and writing (data and methods).

Supplemental Materials

Reproducibility Package: The reproducibility package is available in an OSF repository (Zhang and Zhou 2025; https://osf.io/wxjnr/). Although the original Zhihu posts and People’s Daily articles cannot be shared for legal reasons, we have provided the complete code and derivative data (without text content) for colleagues to replicate the quantitative/computational analyses. Full Zhihu data can be collected via GitHub APIs and the People’s Daily database can be accessed through institutional subscriptions.

  • Citation: Zhang, Yinxian, and Di Zhou. 2025. “One Sentiment, Multiple Interpretations: Contrasting Official and Popular Anti-Americanism in China” Sociological Science 12: 511-536.
  • Received: April 6, 2025
  • Accepted: June 17, 2025
  • Editors: Ari Adut, Kieran Healey
  • DOI: 10.15195/v12.a22

0

Rising Educational Divides in Attitudes: How Polarization across Cohorts Can Mask Age-Related Polarization

By on August 19, 2025 in Articles

Fabian Kratz

Sociological Science August 19, 2025
10.15195/v12.a21

Abstract

PDF (1603 views)
0 Citation


The question of whether attitudes become more polarized over time has stimulated significant scientific and political debate. This study is the first to show that polarization processes can occur both across cohorts and with rising age and that cohort-based polarization may obscure age-related polarization. I introduce the age polarization and cohort polarization hypotheses, which propose that attitudes become increasingly polarized both as individuals age and across successive cohorts. I use multi-cohort panel data from the German Socio-Economic Panel and leverage one of its longest-running attitude measures: concerns about immigration. I show that education-specific differences in immigration concerns intensify both across cohorts and with rising age and that age related polarization only becomes apparent when cohort-based polarization is taken into account. These findings contribute to debates on polarization processes in attitudes over time and advance the literature on heterogeneity in the liberalizing effect of education.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Fabian Kratz: Department of Sociology, University of Munich, LMU. E-mail: fabian.kratz@lmu.de
Acknowledgments: I am grateful to Daniel Krähmer, Madison Garrett, Lena Jost, Philipp Lersch, Josef Brüderl, and participants at the RC28 Conference in Milan (2025) for their helpful comments.

Supplemental Materials

Reproducibility Package: STATA code for replication is available on the author’s Open Science Framework page: https://osf.io/um8f7/. The data sets were provided by the Socio-Economic Panel (SOEP) Study at the German Institute for Economic Research (DIW). Access to the SOEP data requires signing a data assignment contract, which can be requested here: https://www.diw.de/en/diw_01.c.601584.en/data_access.html. For more information, visit https://www.diw.de/en/diw_01.c.838578.en/edition/soep-core_v37eu__data_1984-2020__eu-edition.html.

  • Citation: Kratz, Fabian. 2025. “Rising Educational Divides in Attitudes: How Polarization across Cohorts Can Mask Age-Related Polarization” Sociological Science 12: 486-510.
  • Received: May 23, 2025
  • Accepted: July 6, 2025
  • Editors: Arnout van de Rijt, Peter Bearman
  • DOI: 10.15195/v12.a21

0

How to Make a Functionalist Argument

By on August 14, 2025 in Articles

Andrés Castro Araújo, Nicolás Restrepo Ochoa

Sociological Science August 14, 2025
10.15195/v12.a20

Abstract

PDF (2161 views)
0 Citation


Sociologists have an awkward relationship with functionalist explanations. Despite having declared “functionalism” to be obsolete, some form of functionalist argument still remains cryptically present in much substantive research. We argue that the resulting inability to talk plainly about functions is a major hindrance for theory building in the discipline. As such, this article has two goals. The first is disambiguation. What does it mean to attribute a function to something? We answer this question by elaborating on the distinction between proper functions (responding to why-is-it-there questions) and role functions (responding to how-does-it-work questions). The second is to introduce a typology of functional arguments that builds upon this distinction, allowing us to recast “functionalism” as a set of general explanatory strategies and not as a substantive theory about society. Importantly, these forms of argument are not burdened by the problems with the organicist framework that many sociologists associate with functionalism.
Creative Commons LicenseThis work is licensed under a Creative Commons Attribution 4.0 International License.

Andrés Castro Araújo: Department of Sociology, Duke University. E-mail: andres.castro.araujo@duke.edu.
Nicolás Restrepo Ochoa: Department of Anthropology, University of California-Davis. E-mail: nrestrepoochoa@ucdavis.edu.

Acknowledgments: We would like to thank Kieran Healy, Braulio Güémez, Turgut Keskintürk, Juan R. Loaiza, Gunnar Babcock, Elizaveta Sheremet, Martin Ruef, and Steve Vaisey for all the helpful feedback given throughout the long amount of time it took to write this.

  • Citation: Araújo, Andrés Castro, and Nicolás Restrepo Ochoa. 2025. “How to Make a Functionalist Argument” Sociological Science 12:456-485.
  • Received: May 1, 2025
  • Acceptedd: June 9, 2025
  • Editors: Arnout van de Rijt, Elizabeth Bruch
  • DOI: 10.15195/v12.a20

0
SiteLock