Dear all,



On behalf of this year's poster award committee (Lonna Atkeson, Jake Bowers, Naoki Egami, Ted Enamorado, Jeff Gill, Walter Mebane, Brandon Stewart, Michelle Torres, Ariel White, Yiqing Xu, and Luwei Ying), I am excited to announce this year's winners of our Summer meeting poster awards!



- Best Theory Poster: Melody Y. Huang (UC Berkeley).

In “Variance-based Sensitivity Models for Weighting Estimators Result in More Informative Bounds,” Huang and Pimentel take a variance adjustment approach to estimate valid confidence intervals of the average treatment effect on the treated under unobserved confounding. Huang's proposed method not only produces more stable, and often narrower, bounds than existing sensitivity analysis methods, but also allows researchers to evaluate the robustness of the estimated treatment effects by benchmarking plausible values of residual imbalance against observed covariates. It contributes to the fast-growing literature on sensitivity analysis and significantly improves the accessibility of sensitivity analysis for applied researchers.



- Best Applications Poster: Thomas Cao (Stanford).

Thomas Cao’s poster is titled "The Weaknesses of Selective Sanctions Against Authoritarian Regimes." In this study, and using the case of U.S. selective sanctions on Surveillance Technology Firms in China, Cao finds that sanctioned firms received morecontracts (6.5 million US dollars) due to selective sanctioning --- the opposite of the intended effect of these sanctions. To identify the causal nature of his findings, Cao considers a series of identification strategies (DID, Triple DID, and Generalized Synthetic Controls) and consistently shows a positive causal effect. In addition to defining credible identification strategies to study an important theory and its implications, Cao offers a comprehensive and clear discussion of his substantive findings.



- Best Virtual Poster:  Elisa Wirsching (NYU).

In  “High Quality Social Science Embeddings for 157 Languages,” and relying on Wikipedia corpora, Wirsching et al. outline a promising way to produce word embeddings for a wide array of languages, including many using logographic systems (e.g., Mandarin, Arabic, and Japanese). It has the potential to be a huge public good, as they provide pre-trained embeddings, their corresponding transformation matrices, and subword embeddings --- potentially saving researchers considerable computational and time costs. In addition to giving an impressive presentation of the work, Wirsching has clearly thought a fair amount about validation and dissemination of the measures.



Although there were many amazing posters being presented during all sessions, these three stood out for their rigor, creativity, and clarity. Competition was hard, however, and we would not have been able to identify these outstanding projects without the incredible commitment of designated poster discussants, as well as that of committee members.



Thank you all for your continued support of our students, and congratulations to the winners! We hope you can join us during this year's (virtual) awards ceremony!



Best,



SO

-----------------------

Santiago Olivella

Associate Professor

Department of Political Science

University of North Carolina at Chapel Hill















**************************************************************

               Political Methodology E-Mail List

   Editors:  Dominique Lockett and Gechun Lin

                  <[log in to unmask]>

**************************************************************

     Send messages to [log in to unmask]

  To join the list, cancel your subscription, or modify

                 your subscription settings visit:



  https://www.cambridge.org/core/membership/spm/mailing-list



**************************************************************