Political Analysis

Measuring the Quality of Answers in Political Q&As with Large Language Models | Political Analysis | Cambridge Core Measuring the Quality of Answers in Political Q&As with Large Language Models

While some answers have only a weak semantic connection to questions, they are generally relevant. They also find meaningful correlations between the quality of answers and the party affiliation of the MPs asking the questions. You can read the paper here: www.cambridge.org/core/journal...

Currently in FirstView: In “Measuring the Quality of Answers in Political Q&As with Large Language Models,” @rmichaelalvarez.bsky.social and Jacob Morrier develop an approach for measuring the quality of answers in Q&A sessions using data from the Question Period in the Canadian House of Commons.

What to Observe When Assuming Selection on Observables | Political Analysis | Cambridge Core What to Observe When Assuming Selection on Observables

The authors discuss diagnostic tools to identify problems with estimation methods. Using two applied examples, they recommend researchers consider many estimation methods and model specifications and encourage open reporting. You can read the paper here: www.cambridge.org/core/journal...

Currently in FirstView: In “What to Observe When Assuming Selection on Observables,” Kevin M. Quinn, Guoer Liu, Lee Epstein, and Andrew Martin clarify how most estimators justified by a selection-on-observables assumption are special cases of a general weighting estimator.

Accessibility and Equity in the Research Process: Gender Bias in Elite Interview Recruitment | Political Analysis | Cambridge Core Accessibility and Equity in the Research Process: Gender Bias in Elite Interview Recruitment

Counter to expectations, they find that elites are more likely to schedule an interview when outreach comes from a female alias. This suggests that qualitative interviews may be limited by gender biases. You can read the paper here: www.cambridge.org/core/journal...

Currently in FirstView: In “Accessibility and Equity in the Research Process: Gender Bias in Elite Interview Recruitment,” Margaret A. T. Kenney and John Salchak study how researcher identity affects the research process. Specifically, they look at how this influences elite interview recruitment.

Political Analysis: Volume 33 - Issue 4 | Cambridge Core Cambridge Core - Political Analysis - Volume 33 - Issue 4

(2/2) @zengchen.bsky.social, @eliasdinas.bsky.social, @redatamtam.bsky.social, and more! Most papers are currently open access. You can read the new issue here: www.cambridge.org/core/journal...

The latest issue of PA is out now. We have a great collection of papers by @yamilrvelez.bsky.social, @sysilviakim.bsky.social, @mattblackwell.bsky.social, @sophieehill.bsky.social, @dwlee.bsky.social, @melissazrogers.bsky.social, @kaipingchen.bsky.social, @samuelbaltz.bsky.social (1/2)

This method tightens non-parametric partial identification bounds in settings where outcomes are endogenously missing, enabling more informative and assumption-agnostic inference about treatment effects. You can read the full paper here: www.cambridge.org/core/journal...

Currently in FirstView: “Generalizing Trimming Bounds for Endogeneously Missing Outcome Data Using Random Forests." @cdsamii.bsky.social, Ye Wang, @jlzhou.bsky.social‬ present a partial identification approach that avoids strong assumptions. This is illustrated using a simulation and replication.

Using a Monte Carlo simulation, the authors explore conditions where the bracketing property holds and demonstrate that treatment effects cannot be bracketed when unobserved heterogeneity is correlated with the regressors. You can read the full paper here: www.cambridge.org/core/journal...

Currently in FirstView: “Fixed Effects, Lagged Dependent Variables, and Bracketing: Cautionary Remarks” by Matei Demetrescu, Manuel Frondel, Lukas Tomberg, and Colin Vance investigates a bracketing property used to yield bounds on treatment effects from fixed effects and lagged DV models.

Logo of "Political Analysis" in white letters on a red background, positioned above a yellow section with the hashtag #OpenAccess in white.

#OpenAccess from @polanalysis.bsky.social -

Generative AI and Topological Data Analysis of Longitudinal Panel Data - cup.org/460fgFn

- Badredine Arfi

#FirstView

The authors discuss several potential uses for eye tracking data, including identifying inattention in surveys and measuring an item’s importance in a decision. They also provide starter code for analyzing eye tracking data. Read the full paper here: www.cambridge.org/core/journal...

Currently in FirstView: In “Attention and Political Choice: A Foundation for Eye Tracking in Political Science,” Libby Jenke and Nicolette Sullivan explain what eye tracking allows researchers to measure and how these measures are relevant to political science questions.

Logo of "Political Analysis" in white letters on a red background, positioned above a yellow section with the hashtag #OpenAccess in white.

#OpenAccess from @polanalysis.bsky.social -

Decomposing Network Influence: Social Influence Regression - cup.org/3UKZjwv

"provides a statistical mechanism for explaining actor influence based on observable traits"

- Shahryar Minhas & Peter D. Hoff

#FirstView

They validate their findings using expert-, manifesto-, and poll-based estimates and show that ideological scores produced by LLMs closely map those obtained through expert-based evaluation. You can read the full paper here: www.cambridge.org/core/journal...

Currently in FirstView: “Mapping (A)Ideology: A Taxonomy of European Parties Using Generative LLMs as Zero-Shot Learners.” Riccardo Di Leo, @zengchen.bsky.social, @eliasdinas.bsky.social‬, and @redatamtam.bsky.social see if ML can obtain measures of party ideology that match the validity of experts.

They use moral content in tweets as a case study, highlighting the vec-tionary's ability to process texts missed by conventional dictionaries and its ability to produce measurements more aligned with crowdsourced human assessments. You can read the paper here: www.cambridge.org/core/journal...

Currently in FirstView: “Constructing Vec-tionaries to Extract Message Features from Texts: A Case Study of Moral Content.” @kaipingchen.bsky.social and and colleagues introduce “vec-tionaries” which are embedding-based tools for measuring latent features of messages.

They show that PCRDs estimate the local average treatment effects for districts, not the effects of politician attributes. The paper also addresses confusion regarding PCRDs and offers tools for researchers using PCRDs. You can read the full paper here: www.cambridge.org/core/journal...

Currently in FirstView: In “Seeing Like a District: Understanding What Close-Election Designs for Leader Characteristics Can and Cannot Tell Us,” Andrew Bertoli and Chad Hazlett examine the limitations of politician characteristic regression discontinuity (PCRD) designs.

They find that F-tests can mask weak instruments and understate the uncertainty surrounding estimates from two-stage least squares. IV estimates are often larger than OLS estimates, whose biases they are designed to correct. The paper is open access here: www.cambridge.org/core/journal...

We are pleased to announce the 2025 Editors’ Choice Award for the paper “How Much Should We Trust Instrumental Variable Estimates in Political Science? Practical Advice Based on 67 Replicated Studies” by @apoorvalal.com‬, @maclockhart.bsky.social, @yiqingxu.bsky.social‬, and @garyzu.bsky.social.

Logo of Political Analysis featuring the text 'Political Analysis' in white letters on a red background, above 'Editor's choice articles' in white letters on a yellow background.

The latest winning article of the @polanalysis.bsky.social Editors' choice has been announced, find out more - cup.org/4551W22

The articles represent papers that the Editors see as providing an especially significant contribution to political methodology.

Logo for Political Analysis with the text 'Political Analysis' in white font on a red background, above the hashtag 'OpenAccess' on a yellow background.

#OpenAccess from @polanalysis.bsky.social -

Detecting Formatted Text: Data Collection Using Computer Vision - cup.org/3IMLbjB

"This letter describes a workflow process for structured text extraction using free models and software"

- Jonathan Colner

#FirstView

Banner reading "POLITICAL ANALYSIS" in white capital letters on a red background, with a yellow strip bearing the hashtag "OpenAccess" below.

#OpenAccess from @polanalysis.bsky.social -

Bin-Conditional Conformal Prediction of Fatalities from Armed Conflict - cup.org/4mjbJHl

- David Randahl, Jonathan P. Williams & Håvard Hegre

#FirstView

They use three experimental designs to derive bounds for interactions between the treatment and the moderator. These bounds allow researchers to assess the sensitivity of findings to both priming and post-treatment bias. You can read the full paper here: www.cambridge.org/core/journal...

Currently in FirstView: In “Priming Bias Versus Post-Treatment Bias in Experimental Designs,” @mattblackwell.bsky.social, Jacob Brown, @sophieehill.bsky.social, Kosuke Imai, and Teppei Yamamoto analyze the trade-off between post-treatment and priming biases in survey experiments.

They illustrate this method by studying the relative importance of partisan, racial, gender, and religious identities. 30% of respondents offered random responses and not accounting for this can affect substantive conclusions. You can read the full paper here: www.cambridge.org/core/journal...