PNAS 123/32 pp. e2619078123-e2619078123 2026-08-05 Original Hyperedge approximation for stochastic processes on higher-order networks
Anzhi Sheng, Alex McAvoy, Ye Tian, Silun Zhang, Angela Fontan, Joshua B. Plotkin
APSR — pp. 1-12 2026-07-29 Original Is Blood Thicker Than Water(-and-Earth)? Partisanship Geography in Imperial China and the Limits of Quantitative History
YAO LIN
Abstract
Using Wang Anshi’s Reform as a case study, Yuhua Wang’s “Blood is Thicker Than Water” ( American Political Science Review , 2022) argues that geographically dispersed kinship networks incentivize political elites to support the building of a strong state, whereas geographically concentrated kinship networks undermine state building. In this article, I improve the sample with the help of additional historical records and show that the reported correlation fails to replicate when either the control issue or the outlier issue is considered. I then propose two alternative hypotheses on the role of kinship network dispersal in Wang Anshi’s Reform, based on two common explanations of its partisanship geography in Chinese historical scholarship, that is, status competition and policy adaptiveness. After examining those hypotheses respectively, I reflect on the limits of quantitative history that this case illustrates.
APSR — pp. 1-9 2026-07-28 Original Still Instrumentally Inclusive
STUART J. TURNBULL-DUGARTE, ALBERTO LÓPEZ ORTEGA
Abstract
Do individuals in Western democracies shift their views on LGBT+ inclusion when exposed to opposition from Muslim out-groups? And is this “homonationalist” responsiveness stronger among anti-immigration individuals? Our article “Instrumentally inclusive [...]” addressed these questions using representative data from Britain (Study 1) and crowd-sourced data from Spain (Study 2). We revisit Study 2 in response to concerns about post-stratification weighting a convenience sample and erroneously inconsistent use of robust standard errors. Varying population reference parameters, weight caps, variance estimators, subgroup definitions, and covariate adjustments, we find strong support for the original findings. Neither reasonable alternative weighting schemes nor corrected variance estimators substantively change the results. In fact, the originally published point estimates are equal to or smaller than the median across alternative specifications. Consistent with the Article, we find that treatment increases support for LGBT+ inclusion in Spain, without consistent identifiable variation by immigration attitudes.
PNAS 123/31 pp. e2606267123-e2606267123 2026-07-27 Original When coordination is avoidable: A monotonicity analysis of organizational tasks
Harang Ju
Abstract
Organizations devote substantial resources to coordination, yet which tasks actually require it for correctness remains unclear. The problem is acute in multiagent AI systems, where coordination cost is directly measurable and can exceed the cost of the work itself. Distributed systems theory provides a precise criterion: Coordination is required when a task specification is nonmonotonic, meaning that as histories grow, new information can invalidate prior conclusions. Here we show that Thompson's classic taxonomy of interdependence maps to that criterion, yielding a decision rule for when coordination is required for correctness. We formalize the correspondence in a bridge theorem, apply the rule to 65 workflows from the American Productivity & Quality Center (APQC), and (with a calibrated large language model (LLM), 13,417 Occupational Information Network (O*NET tasks), and illustrate it in multiagent AI simulations. Under our decompositions, 74% of workflows and 42% of O*NET tasks are monotonic, implying that up to 24 to 57% of coordination spending is unnecessary for correctness.
AJPS — 2026-07-23
Original Correcting measurement error bias in conjoint survey experiments
Katherine Clayton, Yusaku Horiuchi, Aaron R. Kaufman, Gary King, Mayya Komisarchik
Abstract
Abstract Conjoint survey designs are spreading across the social sciences due to their unusual capacity to estimate many causal effects from a single randomized experiment. Unfortunately, by their ability to mirror complicated real‐world choices, these designs often generate substantial measurement error and thus bias. We replicate both the data collection and analysis from eight prominent conjoint studies, all of which closely reproduce published results, and reveal high levels of measurement error in all. We then discover a common empirical pattern in how measurement error appears in conjoint studies and, with it, introduce an easy‐to‐use statistical method to correct the bias. Along the way, we provide a much simpler and simultaneously more powerful approach to designing, organizing, understanding, and analyzing conjoint data analyses.
PNAS 123/30 pp. e2509757122-e2509757122 2026-07-20 Original There is no free benchmark: An institutional view of legal AI benchmarking
Neel Guha, Andy K. Zhang, Christine Tsang, Christopher D. Manning, Julian Nyarko, Daniel E. Ho
PNAS 123/30 pp. e2509765122-e2509765122 2026-07-20 Original Measuring disparate impact in human and machine decisions
Jongbin Jung, Sam Corbett‐Davies, Johann Gaebler, Ravi Shroff, Sharad Goel
PNAS 123/30 pp. e2509763123-e2509763123 2026-07-20 Original Continuity and change in US legal tradition: Evidence from judicial citation communities
Elliott Ash, Caterina Chiopris, Robert Mahari, Suresh Naidu
PNAS 123/30 pp. e2606704123-e2606704123 2026-07-20 Original At the boundary: Law and AI
Daniel E. Ho, Julian Nyarko, Vanessa Parli, Christopher D. Manning
Abstract
(原文未提供摘要)
中文 在边界之上:法律与人工智能
Daniel E. Ho, Julian Nyarko, Vanessa Parli, Christopher D. Manning
JOP — 2026-07-17
Original Open-Ended Survey Responses and Political Conceptualizations in a Polarized Era
Maxwell Allamong, Benjamin Beutel, Jongwoo Jeong, Paul Kellstedt
Abstract
(原文未提供摘要)
中文 极化时代的开放式调查回答与政治概念化