Read Methodology Articles with Replication Datasets

Why Do People Draw Neighborhoods Along Race and Party Lines?

i

2024

Subjective Neighborhoods

Map Survey

Spatial Model

Partisanship

Race

i

2023

Fixing Crosswise Survey Bias From Inattentive Respondents

Crosswise Model

Inattentiveness

Bias Correction

Anchor Question

CWise

i

Improve Proportion Estimates From Text With Matching And Continuous Features

Nonparametric

Content Analysis

Matching

Text Features

Classification

Train Once, Classify Across: Party Platforms Predict Speech Topics

i

Cross-domain

Topic Classification

Supervised Learning

Party Platforms

Parliamentary Speeches

Why AMCE Misleads When Studying Preferences in Forced‑Choice Conjoint Experiments

i

Conjoint

AMCE

Preferences

Forced-choice

Causal Inference

i

How Partisan Symmetry Metrics Fail in Practice

Partisan Symmetry

Mean-median

Partisan Bias

Redistricting

Mathematical Analysis

i

How Noisy, Privacy-Protected Facebook Data Still Yield Valid Results

Differential Privacy

Measurement Error

Facebook

Regression

Scalability

i

Fatal Terror Attacks Create Social Desirability Bias in Survey Responses

Social Desirability Bias

Terror Attacks

Electoral Participation

Field Experiment

i

2022

A Permutation Test That Spots Real Changes in Effect Sizes

Changepoint

Permutation

Time-series

Monte Carlo

Alliances

i

Turn Open-Ended Answers Into Continuous Attention Checks

Manipulation Checks

Attention Checks

Text-as-data

Survey Experiments

R Package

i

Skewed Measurement Error Can Bias Regressions Even When Small

Measurement Error

Skewness

Heteroskedasticity

Stochastic Frontier

Nonlinear Estimation

i

How Adaptive Fuzzy Matching Fixes Messy Single-Field Data Merges

Record Linkage

Fuzzy Matching

Adaptive Learning

Entity Resolution

Data Merging

i

New Analysis Uncovers Flaw in Standard ANES Ideological Scales

Ideological Measurement

ANES Scales

Survey Research

JOP

37 R files

1 text files

How Good Are Corruption Measures? A Practical Rubric for Data Quality

i

Data Quality

Validity

V-Dem

Corruption

Measurement

How to Balance Preferences and Predicted Outcomes in Refugee and School Assignments

i

Matching

Refugees

School Choice

Machine Learning

Strategy-proofness

i

Mixing Integration Orders in Time Series Doesn't Increase Error Risk, New Study Shows

Error Correction Model (ECM)

ARDL

Orders Integration

Time Series Analysis

i

Names Signal More Than Race: SES and Migration Bias Audit Experiments

Names

Information Equivalence

Correspondence Audit

Race

Socioeconomic Status

i

Fully Randomized Conjoint Reduces Social Desirability Bias by Two-Thirds

Conjoint Analysis

Social Desirability

AMCE

Experimental Design

Survey Experiments

i

How to Recover Individual Preferences From Conjoint Experiments

Conjoint

Marginal Effects

Individual Preferences

Survey Experiment

Immigration

When Time Matters, Use Cox: A Better Approach for Binary Panel Data

i

Cox

Logit

Duration

BTSCS

Transition Probabilities

New Method Shows How Ordinal Scale Choices Can Shift Empirical Estimates

i

Ordinal

Partial Identification

Monotonic Transformation

Robustness

Sub-Saharan Africa

i

How Populist Are Parties? New Machine Learning Scores From Manifestos

Populism

Manifestos

Supervised Learning

Text Analysis

Europe

i

Machine vs Human Geocoding: Similar Validity, Different Errors in Colombian Rights Data

Geocoding

Colombia

Human-rights

Spatial-probit

INLA

i

Why Conjoint Results Depend on Profile Distribution

Conjoint

AMCE

External Validity

Profile Distribution

Experimental Design

i

How to Estimate Multidimensional Ideal Points on Massive Datasets

Ideal Points

Multidimensional

Scalability

Sparsity

R

i

How CNNs Read Handwritten Vote Tallies for Social Science Research

CNNs

Image Classification

Handwriting

Vote Tallies

Machine Learning

ZIP Codes Work for White/Black Imputation; Blocks Matter for Asian and Hispanic

i

BISG

Geocoding

ZIP Codes

Census Blocks

Georgia

i

Why Predicting Multiple Labels Together Improves Political Text Coding

Multi-label

Text-as-data

Supervised Learning

Mexico

Human Rights

i

Drop Both or One? New Take on Missing Data in Paired Experiments

Attrition

Pairwise Randomization

Deletion Estimator

Variance Estimator

Causal Inference

i

How to Measure Sexism Quickly and Reliably in Political Surveys

Sexism

Hostile Sexism

Measurement

Surveys

Validity

Why Thousands of AI-Generated Placebos Improve Survey Experiments

i

Placebo

Nonspecific Effects

GPT-2

Survey Experiments

Generative Models

i

How Fast Answers Reveal Unseen Inattentiveness in Online Surveys

Attentiveness

Response Time

Clustering

Web Surveys

Survey Quality

Why Standard Interaction Tests Mislead and a Better Lasso-Based Fix

i

Interactions

Regularization

Lasso

Post-double Selection

Heterogeneity

i

Big Data & Rare Events: Standard Fixes Can Backfire

Separation

Rare Events

Dyadic Data

Cauchy Prior

Google Searches Forecast U.S. Protests More Accurately Than Twitter

i

Google Trends

Protests

Forecasting

Crowd Counting

Time Series

i

Sponsorship Shapes Survey Responses — A New Look at Online Polling Biases

Online Survey

Sampling Bias

Response Bias

Political Engagement

i

This study reveals how word-of-mouth alters survey responses on sensitive topics.

Grapevine Effect

Sahel Region

Survey Responses

Experimental Measures

i

MIDAS: Deep Learning That Fixes Missing Data Fast

Multiple Imputation

Denoising Autoencoder

Missing Data

Surveys

Deep Learning

i

List Experiments Can Be Less Accurate Than Direct Questions on Turnout

List Experiments

Direct Questions

Turnout

Partition Validation

Misreporting

i

American Politics Is One-Dimensional, But Politicians Are More Constrained

Constraint

Dimensionality

Ideal Points

Cross-validation

Public Opinion

i

Crowdsourcing Better Validation for Topic Models Used as Measures

Topic Models

Validation

Crowdsourcing

Measurement

U.S. Senators

i

New Estimator Solves Dynamic Panel Dilemma for Small-T Studies

Dynamic Panel Models

Orthogonal Reparameterization

GMM Estimators

Small T Panels

How Multilevel Models Mirror Fixed Effects — And Why Corrections Matter

i

Multilevel Models

Fixed Effects

Cluster-robust

Random Effects

Observational Data

Senate Filibusters End Suddenly? Not Always: A Counterintuitive Look at the War of Attrition

i

2021

War Of Attrition

MLE Estimation

Hazard Rate

Senate Filibusters

i

Beyond Words: Vocal Tone Reveals Hidden Political Signals

Speech Dynamics

Vocal Tone

Oral Arguments

R Programming Language

Fast Estimator for Binary Choice Models With Space and Time Dependence

i

Binary Choice

Spatial Dependence

Temporal Dependence

Pseudo-MLE

Monte Carlo

How AI Helps Better Define Where Civil Wars Actually Happen

i

Civil War

Machine Learning

Geospatial Zones

Regression Discontinuity

2 other files

Hidden Control Variables Give Many Papers Their Statistical Significance

i

Suppression Effects

Control Variables

Transparency

Observational Studies

Statistical Significance

i

Political Science Researchers Face Surprising Challenges with Code Obsolescence

Code Reproducibility

Long-term Maintenance

ISPS Data Archive

Data Preservation

PS

1 text files

1 datasets

Adaptive Research Designs Could Transform How Political Scientists Test Ideas

i

Adaptive Design

Political Science Experiments

Treatment Discovery

Estimation Precision

i

How Firth's Method Solves Duration Model Problems in Political Science

Duration Modeling

Failure Time Data

Covariate Monotonicity

Firth's Estimation

Empirical Guidance

i

Decision Records vs. Voting: A New Approach to Analyzing Committee Outcomes

Bayesian Structural Models

Gibbs Sampler

US Supreme Court

UN Security Council

i

Foreign Workers Not the Only Problem on MTurk: An Indian Quality Control Study

Mechanical Turk

Indian Workers

Data Reliability

Response Accuracy

New Method Creates Political Event Dictionaries with Less Effort

i

Political Event Coding

Nlp

Machine Learning

Dictionary Generation

17 other files

Small Choices, Big Differences: How Corpus and Coding Change News Tone Measures

i

Text Classification

Supervised Learning

Corpus Selection

Coding Design

New York Times

i

More Attributes, More Responses? Conjoint Surveys Can Handle Complexity Better Than Expected

Conjoint Experiment

Survey Satisficing

Filler Attributes

US Senate

Response Quality

i

Even Wrong Network Models Beat Ignoring Spatial Dependence

Spatial Dependence

Spatial Weights

Network Misspecification

Omitted Variable

Spatial Econometrics

i

Detecting When Political Effects Change: A New Change-Point Toolkit

Change Points

Temporal Heterogeneity

Democratization

V-Dem

Panel Data

Minimally-Supervised Corpus-Based Sentiment Analysis Validated for Specialized Legal Language

i

Sentiment Analysis

Corpus-Based Approach

Specialized Legal Language

Benchmarks Datasets

Minimally-Supervised Methods

List Experiments' Achilles Heel: How Placebo Controls Fall Short, Requiring Mixed Lists

i

List Experiments

Measurement Error

Mixed Controls

Non-strategic Errors

Qualitative Transparency Deliberations: Insights from Hundreds of Political Scientists

i

Qualitative Transparency

Deliberative Process

Methodological Guidance

Research Practice

POP

163 PDF files

7 other files

Manski's Method Delivers Reliable Belief Distributions in Online Surveys

i

Belief Distributions

Elicitation

Manski Method

Online Surveys

Expectations

i

Proving Regression Discontinuity Credibility With Equivalence Tests

Regression Discontinuity

Equivalence Testing

Falsification Tests

Density Test

Close Elections

i

Survey Scales Don't Explain Ideology Gaps — Perceptions Do

Anchoring Vignettes

Aldrich-McKelvey

Ideology

Partisanship

Measurement

i

Small Samples Don't Mean Big Problems: PML Improves Logit Model Estimates

Penalized Maximum Likelihood

Logit Model Bias

Firth's Method

Separation Problem

i

Elite Ethics Experiments: What Officials and Voters Truly Think

Elite Experimentation

Ethics Perception

Governance Innovation

i

From the Lab to the Poll: The Use of Survey Experiments in Political ResearchIPSR

i

Match Millions in Minutes: New Method Balances Six-Treatment Experiments

Full-matching

Causal Inference

Approximation Algorithm

Voter Mobilization

Large-N

Just a Few Checks Can Match Full Survey Quality Control

i

Survey Quality

AmericasBarometer

Geo-tracking

Audio Capture

Classification

i

Do More Surveys Mean Better Results? New Study Finds Precision Without Distortion

Survey Experiment

Repeated Measures Design

Between Subjects

Precision

i

Understanding Spatial Effects in Political Science: More Than Just Location

Spatial Analysis

Democracy Diffusion

Lag Terms

Geographic Variation

Why ADL(2,1) Isn't a Universal Fix for Serial Dependence

i

ADL(2

1)

Lagged Variables

Serial Correlation

Wald Test

Dynamic Models

Why Plugging ML Predictions Into Regressions Misleads — And How to Fix It

i

Missing Data

Machine Learning

Regression

Prediction Error

Text Analysis

i

How Lobbying Networks Reveal Hidden Communities in Congress

Community Detection

Bipartite Networks

Lobbying

Interest Groups

Congress

i

Correcting False Positive Rates in Interaction Model Testing

False Positive Rate

Benjamini Hochberg

Interaction Models

Brambor Procedure

CPS

1 PDF files

New Technique Maps American Public Ideology at District Level

i

Bayesian Kriging

US States

2010 Census

Congressional Districts

How Legislative Language Reveals Bureaucratic Discretion and Executive Delegation

i

Delegation

Discretion

Syntactic Parsing

Text-analysis

US States

i

Economic Games Miss Ethnic Bias — Misattribution Tests Reveal It in Nairobi

Ethnic Bias

Misattribution

Nairobi

Dictator Game

Public Goods

New Way to Spot Meaningful Government Policies

i

PU Learning

UK Govt Regs

Expert Validation

Automated Classification

MTurk's 'Don't Know' Quandary: A Surprising Caveat to Online Political Knowledge Studies

i

Mechanical Turk

Attention Checks

Don't Know Option

Response Behavior

PS

i

Media Exposure of Portfolios as a Measure of Relevance, Published in Bpsr, Vol. 15, N. 1, 2021BPSR

i

Facebook Ads Fall Short: How Survey Recruitment Claims Mislead

Facebook Ads

Demographic Targeting

Survey Recruitment

Success Rates

i

Enhancing Pooled Event History Analysis Through Multilevel Modeling

Multilevel Modeling

Pooled Event History Analysis

Heterogeneity

Random Coefficients

Antiabortion Policies

Few Upper-Level Units Don't Doom Multilevel Models: ML Estimates Remain Unbiased

i

Multilevel Models

Restricted ML Estimation

Coefficient Estimates

Monte Carlo Simulation

BJPS

2 other files

When Lawmakers Shift: Revealing Different Ideal Points Across Policy Domains

i

Ideal Points

Roll-call

Partial Pooling

U.S. House

Issue Space

i

Why Do List Experiments In Singapore Inflate Responses For Low Education Respondents?

List Experiments

Singapore

Mechanical Inflation

Control Statement

i

Why Multinomial Logit Coefficients Mislead: Focus on Predicted Probabilities

Multinomial Logit

Predicted Probabilities

Reference Category

Nominal Outcomes

i

Aligning Votes and Speeches Reveals Shared Ideology and Distinct Issue Signals

Scaling

Latent Factor

Resampling

Text Analysis

Senate

i

Unseen Networks Can Skew Spillover Estimates — New Sensitivity Tools Show How

Spillover

Networks

Sensitivity

Twitter

Field Experiment

i

Beyond Siloed Analysis: How Multilayer Models Solve Political Network Inference Problems

Multilayer Networks

Exponential Random Graph Model

ERGM Modeling

Policy Communication

i

How a 'Binary Snowball' Method Boosts Precision in News Topic Coding

Text Classification

Comparative Agendas

Machine Learning

Binary Classification

Hungary

When the Wald Test Misleads Spatial Model Choice

i

Spatial Dependence

Spatial Durbin

Wald Test

Likelihood Ratio

Bootstrap

i

The Wild Bootstrap With a Small Number of Large ClustersRESTAT

i

Eye-Tracking Reveals How Respondents Simplify Conjoint Choices

Conjoint

Eye-tracking

AMCE

Bounded Rationality

Survey Experiment

i

Screeners Could Enhance Political Science Data Quality

Screeners For Attention

Item-response Theory

Survey Standardization

i

2020

Why UK Backbenchers Aren't Becoming More Alike — Except the Senior Ones

Hansard

Stylometry

Backbenchers

Panel Regression

i

Divided Government Causality? Two New Methods Address Election Challenges

Regression Discontinuity Design

Probability Restricted Design

Electoral Distance

Divided Government

Central Bank Preferences Uncovered Through Text Analysis Alone

i

Text Analysis

Scaling Methods

FOMC

Financial Crisis

i

Small Corpora, Big Insights: Word2Vec Tracks 161 Years of Shifting Meanings

Word2vec

Semantic Change

Newspapers

Bootstrap

Pretraining

i

Active Learning Cuts Text Labeling Costs, Especially for Rare Categories

Active Learning

Text Classification

Intercoder Reliability

Simulation

Text Matching Offers New Solutions for Addressing Confounding Observational Gaps

i

Observational Studies

Text Confounders

Matching Methods

Causal Inference

i

Advocacy Output: Automated-coding Documents from Human Rights OrganizationsJHR

Automated Text Analysis Reveals Radical Shifts in Animal Rights Movements

i

Automated Text Analysis

Ideological Shifts

Direct Actions

Animal Liberation Movement

1 text files

New Approach Makes Statistical Duration Findings Easier for Political Scientists

i

Cox Model

Duration Analysis

Hazard Ratio

Expected Durations

i

Beyond the Unit Root Question: New Approach Handles Uncertainty in Long-Run Economic Relationships

Unit Root Uncertainty

Long Run Relationship Inference

Regression Model Application

i

Less Deception, More Data: Field Experiment Shows Monitoring Reduces Discrimination Against Hispanics

Field Experiments

State Legislative Offices

Deception Reduction

Discrimination Mitigation

New CNA Method Handles Multi-Value Outcomes & Fuzzy Sets

i

Fuzzy Set Qca

Causal Inference Methods

Membership Scores

R Software

i

Compare Mediation Effects Without Assuming No Confounding — An Audience Costs Case

Mediation

Audience Costs

International Law

Comparative Mediation

Causal Inference

i

Why Clustered SEs Underestimate Uncertainty — A Better Method

Clustered SE

Bootstrapping

Simulation

Small Clusters

i

How Crowdsourcing Makes Local Government Data Accurate and Cheap

Local Government

Crowdsourcing

Data Validation

Consensus Measures

United States

New Dataset Reveals Everyday Racial/Ethnic Environments Are Less Extreme Than Traditional Measures Suggest

i

Geolocation Data

Measurement Gap

Daily Life Exposure

Political Representation

i

Why Averaging Hides Politics: A Binary Quantile Model Reveals Hidden Heterogeneity

Binary Quantile

Discrete Choice

Heterogeneity

IIA

Counterfactual

i

Allowing Contextual Effects Tightens Ecological Inference Bounds by About 44%

Ecological Inference

Duncan-Davis

Contextual Effects

Partial Identification

2x2 Datasets

How Political Scientists Can Properly Estimate Direct Causal Effects

i

Marginal Structural Models

Controlled Direct Effects

Political Participation

i

Dropped Groups Explain Why Logit and OLS Diverge

Fixed Effects

Grouped Data

Logit

OLS

Binary Outcome

i

Fixed Effects Can Skew Rare Event Predictions, New Method Fixes It

Fixed Effects

Rare Events Data

Penalized MLE

Monte Carlo

Civil War

i

New Sampling Method Enters Political Science Mainstream

Respondent Driven Sampling

RDS Method

Syria

Hidden Populations

Social Network

POP

i

Why Log-Log Plots Mislead: Rethinking Power Laws in Public Budgets

Power Law

Stochastic Process

Public Budgets

Log-log Plot

Statistical Testing

i

Which Text-Matching Methods Actually Work? Evidence From 100+ Tests

Text Matching

Match Quality

Text Representation

Distance Metric

Causal Inference

2 other files

When Spatial Weights Are Mismeasured: A Practical Test for Spatial Models

i

Spatial Weights

Measurement Error

Panel Data

Monte Carlo

Endogeneity

i

Field Guide to Measuring Political Networks

Network Measurement

Political Science

Relational Data

Node Ties Tradeoff

i

When Subgroup Comparisons Mislead: The Problem With Conditional AMCEs

Conjoint

AMCE

Interactions

Marginal Means

F-test

i

Context Matters: A Dirichlet-Process Model for Hidden Effect Heterogeneity

Dirichlet Process

Latent Heterogeneity

Generalized Linear Models

Comparative Analysis

Hamiltonian Monte Carlo

Using Multiple Modes to Spot Measurement Error in Experiments

i

Measurement Error

Experimental Modes

Heterogeneous Effects

Nonparametric

MTurk

i

Women's March cellphone data reveals news/social media accurately reflect protest size

Cellphone Data

Womenâ€™s March

Geolocated Tweets

Protest Size

i

Online Coders Revolutionize Political Video Analysis: Speed & Cost Efficiency Unlocked

Content Analysis

Online Coding

Crowd Aggregation

Political Communication

How Text Mining Predicts When Rival Parties Will React (Evidence From Slovakia)

i

Partisan Responsiveness

LDA

Time Series

Slovakia

Classification

Flawed Controls in Political Science Models

i

Interactive Effects

Quadratic Terms

Control Strategies

Regularized Estimators

Facebook Ads Can Approximate Representative US Public Opinion on Climate Change

i

Quota Sampling

Facebook Ads

Us Public Opinion

Acs Validation

i

Online Recruitment in India vs. U.S.: Not All Samples Can Make Valid Inferences

Representative Samples

Convenience Samples

Partisanship Effects

US Benchmarks

India Recruitment

i

A Better Alternative to Inverse Probability Weighting for Time-Varying Causal Effects

Residual Balancing

MSM

IPW

Causal Inference

Continuous Treatments

Shadowing Political Elites: A Scalable Way to Observe Daily Behavior

i

Shadowing

Elites

Fieldwork

Observational Methods

i

Do Sponsor Logos Bias Online Surveys? Evidence Says No

Sponsorship

Online Surveys

Satisficing

Social Desirability

Survey Experiment

i

Spatial Interdependence in IV Studies: A Surprising Blind Spot for Researchers

Spatial Dependence

Instrumental Variable

OLS Bias

Monte Carlo Simulation

i

Geographic Data Points to Better Case Selection and Richer Theory Development

Spatial Autocorrelation

Mixed Methods

Case Selection

Theory Development

i

Text Alone Can Bridge the Gap Between Spatial Voting Theory and Policy Positions

Spatial Voting

Policy Positions

Text Scaling

Experimental Design

i

Stacked Regression Beats MRP for Local Public Opinion Estimation

SRP

MRP

Poststratification

Ensemble Methods

Public Opinion

Geographical Correlation Matters: CAR Models Improve State Politics Analysis

i

Spatial Autocorrelation

Multilevel Modeling

United States States

Bayesian Estimation

i

Why Party Intercepts Can't Pin Down Valence in Spatial Models

Valence

Discrete Choice

Party Intercepts

Spatial Theory

Germany

i

How Reliable Are Sensitivity Analyses?

Sensitivity Analysis

Model Averaging

Robustness Criteria

Variables Influence

i

When Networks Are Wrong, Spatial Models Mislead — BMA Fixes It

Spatial Econometrics

Network Uncertainty

BMA

Spatial Autoregression

Monte Carlo

Amazon Mechanical Turk's Data Quality Crisis: New Insights into Detection

i

Mechanical Turk

VPNs

Data Quality

Fraud Detection

i

When Surveys Get Interrupted: How Unexpected Events Reveal Causal Effects

Unexpected Events

Natural Experiment

Causal Inference

ESS

Charlie Hebdo

When Word Order Matters: LSTMs Improve Political Text Classification

i

LSTM

Text Classification

Word Order

Social Media

Newspapers

How Voter Registration Records Sharpen MRP Estimates to County and District Levels

i

Voter Registration

Multilevel Regression

Poststratification

Exit Polls

2012 Election

i

BLUE Not Best? Non-Normal Errors Challenge Least Squares in Political Data

BLUE Estimator

Least Squares Method

Non-Normal Errors

Political Data

Expert Disagreement: Mean vs Median Responses Lead to Different Survey Outcomes

i

Mean Response

Median Response

Modal Response

EU Integration

Expert Surveys

i

Why Better Measuring Populist Attitudes Matters for Understanding Democracy's Health

Non Compensatory Measures

Multi Dimensional Constructs

Populist Attitudes

Five Populism Scales

i

Sensitivity Bias in Surveys: New Insights from 30 Years of List Experiments

Social Reference Theory

Sensitivity Bias

List Experiment

Measurement Approach

Party Embeddings Reveal Ideology Hidden in Parliamentary Speech

i

Word Embeddings

Party Embeddings

Ideology

Parliamentary Corpora

Manifestos

i

2019

How Misclassified Failures Distort Survival Studies — And a Bayesian Fix

Misclassification

Survival

Bayesian

Split-population

Civil War

i

How To Test Long-Run Relationships When Stationarity Is Unknown

Cointegration

Long-Run Multiplier

Unit Root

Bounds Testing

Simulations

DiD vs Lagged Outcomes: Why Estimates Bracket the Truth

i

Difference-in-differences

Lagged-outcome

Parallel-trends

Ignorability

Semiparametric

i

Edges Make Edges: Rethinking How Political Networks Form

LSGM

Nodes-as-actions

Network Dynamics

Alliances

Cosponsorship

Multinomial Ideal-Point Models: Reclassifying Nonresponse Reveals Hidden Conservatism

i

Ideal Points

Multinomial

EM Algorithm

ANES

Nonresponse

i

Dropping Manipulation-Check Failures Can Produce Serious Bias

Manipulation Check

Experimental Design

Bounds

Causal Inference

Treatment Effects

When Standard G-Estimation Misses: Regression-With-Residuals Handles Effect Moderation

i

CDE

Mediation

G-estimation

RWR

Moderation

i

Matching Meets Stratified Sampling: A Practical Theory of Inference

Matching

Stratified Sampling

Causal Inference

Propensity Score

Treatment Effects

i

Non-Random Measurement Error: A New Tool for Political Scientists

Non-random Error

Sensitivity Analysis

Political Scientists

Statistical Models

i

Flexible Stats Challenge Survey Data Gold Standard

Multilevel Regression Post-stratification

Bayesian Additive Regression Trees

Survey Extrapolation

Statistical Methods

i

Accurate Causal Estimates: Bivariate Probit Models Using Bayesian vs Maximum Likelihood Estimation

Bivariate Probit Models

Bayesian Estimation

Maximum Likelihood Estimation

Causal Inference

i

Why Algorithms Still Beat Crowds at Predicting Recidivism

Recidivism

Statistical Learning

MTurk

Predictive Accuracy

Algorithmic Fairness

i

Dynamic Pies Reveal More Than Just Budget Breakdowns Across US States

Compositional Data

Time Series

Dynamic Visualization

US States

i

Comparing Random Forest With Logistic Regression for Predicting Class-Imbalanced Civil War Onset Data: A CommentPol. An.

Network Effects Cause Interference In State Legislature Experiments

i

Field Experiments

Spillover Effects

Network Analysis

Legislative Politics

Clearing Up Confusion in Political Research Design

i

Research Design Framework

Monte Carlo Techniques

Analytical Features

Diagnosands

Preregistration

Survey Experiments Resistant to Demand Effects: A Counterintuitive Look

i

Demand Effects

Survey Experiments

Online Political Science Research

i

Recovering Time-Varying Population Distributions From Sparse Marginal Data

Ecological Inference

Poststratification

Dirichlet

Time-series

Estsubpop

How Bayesian Models Reconstruct Country-Year Public Opinion on Democracy

i

Bayesian

Latent Trait

Public Opinion

Democracy Support

Panel Smoothing

i

Robust Dynamic Model Reveals Sudden Political Shocks Missed by Standard Models

Latent Variables

Dynamic Models

Judicial Ideology

Democracy

Time Series

How Do Survey Experiments on Mechanical Turk Generalize? New Replication Evidence Provides Insights.

i

Experimental Generalization

Mechanical Turk Replicates

Demographic Variation

Treatment Effects

i

Linking Survey Items to Real Preferences: Hierarchical Models That Reveal Polarization

Item Response

Latent Preference

Hierarchical IRT

Polarization

HIRT

i

How Cross-Validation Can Go Wrong and What to Do About It.Pol. An.

i

Interview Mode Shifts Responses: Self-Completing vs Face-to-Face for Cognitively Diverse Audiences

Item Non-response

Self-complete Surveys

In-person Interviews

Cognitive Skills

U.S.

Randomized Experiment

Response Variation

i

Newspapers Reveal Political Power Through Coverage Space

Newspaper Coverage

Word Frequency

Historical US

State Party Committees

i

A Simpler, Better Way to Model Network Data: AME Beats ERGM and LSM

Bayesian

Dyadic Data

Network Analysis

ERGM

Latent Space

When List Experiments Mislead: Fixing Measurement Error in Sensitive Surveys

i

List Experiment

Measurement Error

NLSreg

MLreg

Misspecification Test

Lost in Translation: How Aggregation Masks Key Insights in Event Data

i

Report Level Data

Aggregation Bias

MMAD Dataset

Measurement Error

i

Lost in Translation? New Algorithm Fixes Geolocation Accuracy

Geolocation

Machine Learning

Supervised Algorithm

Location Mentions

N-gram Patterns

News Articles

i

How Machine Learning Detects Shifts in Human Rights Standards

Human Rights

Supervised Learning

Text Analysis

Measurement

i

How Uncertainty Changes What Spatial Models Say About Party Positions

Spatial Econometrics

Measurement Uncertainty

Bayesian IRT

Party Competition

Germany

How Do We Measure Political Sophistication? New Methods Solve an Old Problem

i

Political Communication

Statistical Modeling

Google Books Ngrams

Text Complexity

i

How Unit Choices Warp Results — A New Measure That Reduces MAUP

MAUP

Spatial Distribution

Measurement

Monte Carlo

Subnational

How Party Labels Improve Polarization Measurement in UK Commons Speeches (1811–2015)

i

Polarization

Text Analysis

Supervised Methods

House Of Commons

Party Affiliation

i

BigKRLS Makes Kernel Models Fast Enough for Real Political Data

KRLS

BigKRLS

Kernel Methods

Computational Efficiency

Marginal Effects

How To Reveal Direction In Symmetric Networks

i

Network Reconstruction

Directed Networks

Investment Treaties

Regression

Simulation

i

How More Zero Location Mentions Impairs Geolocation Accuracy

Location Accuracy

Zeros

Contextual Features

Machine Learning

New Approach Tamest Multilingual Sentiment Analysis For Measuring Parliamentary Conflict

i

Sentiment Analysis

Translatable Dictionary

Parliamentary Conflict

Legislative Debates

LSQ

i

Factual Checks Reveal Attention: A New Tool for Experimental Clarity

Factual Manipulation Checks

US Politics

Attention Measurement

Experimental Methods

i

Why Fixed-Effects Models Can Make Estimates More Biased Than OLS

Fixed Effects

Dynamic Misspecification

Hausman Test

Omitted Variables

Panel Data

i

Ignoring Inattentive Respondents Biases Political Surveys — Even List Experiments

Inattentiveness

Instructed-Response

List Experiment

Survey Bias

Online Survey

i

Small Cluster Numbers? New Tools Help Political Scientists Get More Accurate Results

Cluster Robust Standard Errors

Ibragimov Mueller Method

Stata Software

R Packages

Simulate Events to Spot Bad Fits in Relational Event Models

i

Relational Events

Survival Analysis

Event Simulation

Model Fit

Network Dynamics

i

Who Are IV Compliers? A Simple Way to Profile Them

Instrumental Variables

Compliers

LATE

Noncompliance

Stata

i

Heavy-Tailed Policy Changes: Why Normal Distributions Might Mislead Researchers

Heavy-tailed Distribution

Policy Change Models

Within-time Variance

Normal Distribution

When 'No Liars' Fails: Bounding Sensitive Answers in List Experiments

i

List Experiment

No Liars

Bounds

Survey Experiment

California

i

Replications in Context: A Framework for Evaluating New Methods in Quantitative Political SciencePol. An.

Mapping Nigeria's Elite Connections with Scraping Algorithms: A New Approach to Political Networks

i

Scraping Algorithms

Nigeria

Elite Networks

Patronage Politics

i

New Method for Simulating Duration Data Improves Cox Model Accuracy

Duration Analysis

Cox Proportional Hazards

Hazard Function Simulation

Non-parametric Survival

i

Can Speech Recognition Replace Human Transcripts in Political Text Analysis?

Speech Recognition

Text-as-data

Bag-of-words

WERSIM

R

i

Material Actions Outweigh Words in Diplomatic Crises: A Signal Clarity Study

Signal Credibility

Berlin Crisis

US-Soviet Union

Statistical Learning Techniques

i

Student Samples Vary Demographically but Produce Similar Experimental Effects

Student Samples

External Validity

Survey Experiment

Crisis Bargaining

Treatment Heterogeneity

i

Real-Time Financial Stress Insights Could Reshape Political Behavior Research

Financial Stress

Digital Measurement

Voting Behavior

Public Opinion

BJPS

Machine Coded Data? Not Always Better Than Human Coding for Underreporting Bias

i

Underreporting Bias

State Repression Events

Agence France-Presse

Associated Press

Machine Coding

i

Venezuela Land Reforms: Individual vs Aggregate Data Matters for Testing Distributive Politics

Venezuela Land Reforms

Individual-level Data

Macro-level Analysis

Distributive Targeting

i

Probabilistic Model Streamlines Large-Scale Data Merging

Probabilistic Model

Record Linkage

Administrative Data

Large-scale Merging

i

Survey Incentives Increased Responses but Weights Failed to Remove Nationalist Bias

Nonresponse

Instrumental Variables

Survey Weights

Switzerland

Field Experiment

i

There’s No Magic Formula: Expert Reliability Falls Flat Across Many Characteristics

Expert Reliability

Latent Traits Estimation

Ordinal IRT Model

Coder-level Data

1 PDF files

Unit Fixed Effects Models: A Heads-Up on Their Hidden Limitations

i

Unit Fixed Effects

Nonparametric Matching

Regression Models

Dynamic Causality

i

When Propensity Score Matching Hurts: Why Better Matching Beats PSM

Propensity Scores

Matching

Blocking

Causal Inference

Bias

When Imputation Hurts: Listwise Deletion Outperforms Under MNAR

i

2018

Multiple Imputation

Listwise Deletion

MNAR

Simulation

i

New Statistical Method Challenges Causal Inference Standards in Political Science

Equivalence Testing

Causal Inference

Balance Tests

Placebo Tests

Natural Experiments

i

Identify Direct Effects in Survey Experiments Without Risky Assumptions

Mediation

Factorial Design

Survey Experiments

Selection-on-observables

Direct Effect

i

New Method Shows U.S. Supreme Court Arguments' Quality Highly Susceptible to Unseen Confounders

Confounder

Supreme Court Justice Voting

Simultaneous Sensitivity Analysis

Oral Argument Quality

i

How Classifier Accuracy Measures Polarization in Westminster Politics

Classification Accuracy

Polarization

Westminster

Text Classification

House Of Commons

i

How Do We Know What People Think Locally? New Insights on Survey Accuracy

Post-stratification

Constituency-level Predictors

Geographic Smoothing

British 2010 Election

Detect Hidden Actor Roles in Dynamic Networks With ego-TERGM

i

Ego-TERGM

Latent Roles

Longitudinal Networks

Expectation-maximization

Kapferer

i

WTO Membership Linked to Unrelated Outcomes: A Problem of Selection Bias

Selection Bias

Regression Analysis

Environmental Treaties

Informed Research

i

Neurotics Avoid Ideological Questions Due to Error Aversion

Bayesian Hierarchy

Neuroticism Avoidance

CCES Survey

Ideological Uncertainty

i

A New Approach to Boosting Credibility in Political Science Observational Studies

Observational Studies

Predictive Validity

Generalization Error

Regularization

i

How Senate Speeches and Votes Reveal Hidden Ideology and Leadership

Spatial Preferences

Gaussian Copula

Factor Model

Text

US Senate

i

Flexible Likelihood Models Slash Bias in Political Causal Inference

Causal Inference

Endogeneity

Matching

Heckman

IMF

i

ARDL Models Offer More Robust Approach to Cointegration Testing in Time Series Analysis

ARDL Model

Bounds Test

Cointegration Testing

Dynamic Modeling

i

Why Your Experimental Data Might Be Messed Up - And How to Fix It

Regression Discontinuity

Conditioning Bias

Posttreatment Control

Causal Inference

i

When Interaction Models Go Wrong: Simple Checks to Avoid Misleading Results

Interaction

Moderation

Common Support

Nonlinear

Replication

Correcting Causal Inference Biases in Time-Series Cross-Sectional Studies

i

Time Series Cross Section

Selection Bias

Welfare Spending

Lagged Effects

i

Measure Individual Ideology From Social Media Short Texts With a Dynamic Lexicon

Ideology

Social Media

Text-scaling

Lexicon

Voting Intentions

i

Fixed Effects Models: How Researchers Misinterpret Their Impact

Fixed Effects

Regression Analysis

Counterfactuals

i

Low Information Amplifies Gender Effects; High Information Dampens Them

Information

Gender

Laboratory Experiment

Dynamic Presentation

Candidate Evaluation

i

Why Survey Vignettes Mislead: The Hidden Information Equivalence Problem

Information Equivalence

Survey Experiments

Exclusion Restriction

Democratic Peace

Placebo Tests

i

New Methods Reveal How Freedom of Expression Leads to Democratic Change

Ordinal Sequences Analysis

V-Dem Dataset

Freedom Of Expression

Election

i

When Expert Averages Fail: IRT Fixes Cross-National Coding

IRT

Expert Coding

V-Dem

DIF

Panel Data

i

How Labels Versus Photos Change Voter Reactions to Candidate Race

Treatment Mode

Conjoint

Race

United States

i

When List-Experiment Estimators Break Down: Small Errors, Big Biases

List Experiments

ICT-MLE

Measurement Error

Monte Carlo

Survey Design

i

Japan's 2014 Vote Didn't Give LDP a Clear Policy Mandate

Conjoint Analysis

Manifestos

Japan

Electoral Mandate

Public Opinion

i

Frailty ERGM Reveals How Hidden Traits Warp Network Conclusions

ERGM

Frailty

Networks

Monte Carlo

Multilevel Estimation

1 Archives

Google Translate Works for Bag-of-Words Text Analysis

i

Machine Translation

Bag-of-words

Topic Models

Europarl

LDA

How to Estimate Ideal Points Without Strong Assumptions

i

Ideal Points

Nonparametric

Hypothesis Testing

Supreme Court

Multidimensionality

i

Adding Covariates Can Worsen Bias in Political Studies?

Omitted Variables Bias

Unobserved Confounders

Countervailing Effects

Overadjustment

i

Why Spatial Instruments Rarely Work

Spatial Instruments

Exclusion Restriction

Simultaneity

Cross-unit Dependence

i

QCA's Blind Spot: Low Power Can Hide True Causal Paths

QCA

Permutation Testing

Power Analysis

Truth Table

Consistency

i

Time-Varying Coefficients Mislead. Survivor Functions Reveal the Real Effects

Duration Analysis

Time-varying

Survivor Function

Event History

Interstate Conflict

i

New Measure Reveals How Relationships Between Politicians Are Quantified

Latent Factor Models

Network Analysis

Western Europe

Coalition Politics

i

Realists Vs Idealists: Do Necessary Conditions Matter in QCA?

QCA

Necessity

ESA

Data-mining

Idealism

i

Navigating Population-Based Municipal Rules in France, Germany & Italy

Regression Discontinuity Design

France

Germany

Italy

Boosting Political Science Experiments with an Easy Motivation Technique

i

Seriousness Technique

Experiment Motivation

Student Experiments

Political Science

2 Stata files

Scholars vs Subjects: A Clash on Ethics for Political Field Experiments

i

Ethics Debate

Informed Consent

Deception Norms

Survey Study

Political Methods

POP

i

Why Preprocessing Choices Can Make or Break Unsupervised Text Analysis

Preprocessing

Unsupervised Learning

Sensitivity Analysis

Feature Selection

i

Party Cues Shape Political Text Coding — A Surprising New Study

Party Cues

Coder Bias

Immigration Policy

Manifesto Texts

i

Conjoint Surveys Withstand Dozens of Tasks, New Experiments Show

Conjoint

Satisficing

MTurk

SSI

Survey Experiment

i

LDV Misconception Corrected: Include More Lags

Lagged Dependent Variables

Regression Models

Autocorrelation

Monte Carlo Simulation

i

Unlock Insights in Political Science with Flexible Tree-Based Models

Machine Learning

Nonparametric Methods

Decision Trees

Random Forests

i

Partial Observability Model Flawed? Minor Errors Can Cause Big Inference Problems

Partial Observability Models

Model Critique

Functional Form Misspecification

Monte Carlo Simulations

How to Map Radio Reach: Signal Models Reveal Media Exposure During Mali's 2012 Coup

i

Propagation Models

Media Exposure

Radio

Mali

Signal Strength

i

Multiple Imputation Isn’t Always Better: When Complete-Case Regression Works

Multiple Imputation

Listwise Deletion

Missing Data

Regression

Complete-case

i

When to Weight Survey Experiments: Why Unweighted Estimates Often Suffice

Survey Weights

Survey Experiments

SATE

PATE

Poststratification

i

2017

New Framework Ties Human Insight with Computer Efficiency for Political Text Analysis

Pairwise Comparison

Human Coding

Political Texts

Crowdsourcing

Racial Names Mostly Signal Race, Not Socioeconomic Clues

i

Excludability

Racial Signaling

Audit Study

Legislators

Public Records

i

Voluntary Voting Didn't Boost Turnout? The Surprising Bounds on Chilean Election Data

Electoral Law

Voter Turnout

Causal Bounds

Chilean Elections

i

Humans Flail at Keyword Search; A Better Computer-Based Approach Emerges

Keyword Selection

Human Bias

Text Analysis

Classifier Errors

i

Voter Files Aren't Always Reliable: Assessing Differential Registration Bias

Differential Registration Bias

Voter Files

United States

Sensitivity Analysis

i

Ensembles Beat Single Models for Estimating Heterogeneous Treatment Effects

Heterogeneous Effects

Ensemble Methods

Randomized Experiments

Causal Inference

i

Bridging the Gap: MrsP Extends Subnational Poll Predictions Beyond Census Limits

Multilevel Regression & Poststratification

Syntheticpoststratification Method

Subnational Public Opinion Estimation

Political Representation Accuracy

i

New Approach Uncovers Positive Early Voting Effects Contrasting Previous Findings

Difference-in-differences

Front-door Estimators

Bounds Estimation

Florida Elections

i

Find Structural Breaks in Noisy Count Data—No Need to Guess How Many

Changepoints

Dirichlet Process

Bayesian

Overdispersion

Campaign Contributions

i

How to Do Mixed-Methods Research When Cases Aren't Independent

Mixed Methods

Spatial Dependence

Spatial Econometrics

Case Selection

Homicide

i

Long-Term Effects for Models With Temporal DependencePol. An.

i

How to Track Influence in Incomplete Social Networks

Centrality

Indegree

Network Sampling

Longitudinal

Twitter

i

Data Reporting Reveals a Distinct Form of Government Transparency

Transparency

Data Dissemination

Item Response

World Bank

ICRG

i

Who Benefits From Guessing on Political Knowledge Tests?

Political Knowledge

Guessing

Bayesian IRT

Taiwan

Gender Gap

i

Beyond Simple Numbers: Powerful Network Analysis Techniques for Political Researchers

Network Inference

Political Networks

ERMG Modeling

Latent Space Analysis

i

Time-Varying Fixes Often Misapplied in Duration Models

Duration Models

Proportional Hazards

Time-varying

Nonproportional Hazards

Survival Analysis

i

Historical Weather Reveals Bigger Uncertainty in Rainfall’s Effect on Turnout

Rainfall

Turnout

Randomization Inference

Variance Estimation

United States

i

Newspaper Text Variation Predicts Conflict Before It Happens

Topic Modeling

Text Analysis

Conflict Prediction

Panel Regressions

Online Social Ties Mirror Offline Relationships: Implications for Political Science Research

i

Social Networks

Experimental Methods

External Validity

Tie Similarity

i

How New Matching Methods Maximize Balance and Sample Size in Causal Analysis

Matching Methods

Matching Frontier

Sample Size Optimization

Balance Metrics

i

How Four Simple Design Choices Make Survey Field Experiments More Efficient and Ethical

Survey Experiments

Sampling Frame

Field Experiment

Placebo Control

Canvassing

i

Actual vs Perceived Voting Intentions: A Family Study in Belgium

Pcss Dataset

Belgium Flemish Region

Perceived Voting Intention

Actual Voting Intention

i

Facebook Ads as a Cheap, Reliable Way to Sample Political Activists Worldwide

Facebook

Sampling

Activists

Poststratification

Time-series

i

Detecting Who Hides Sensitive Opinions: A Regression-Based Fix for List Experiments

List Experiment

Misreporting

Prejudice

Regression

Ideology

i

Why Unbiased Coefficients Can Still Mislead: Transformation-Induced Bias

Transformation Bias

Marginal Effects

Predicted Probabilities

Simulation

Small-sample

i

When Absences Matter: New Tools for Asymmetric Hypotheses

Asymmetry

Set-theory

Large-N

Stochastic Frontier

DEA

i

How Predictive Models Complement Regression in Political Research

Predictive Modeling

Regression

Forecasting

Interstate Conflict

Benchmarks

i

Unknown Delay? Here's How Political Scientists Can Still Analyze Time Lags Effectively

Bayesian Model Averaging

Distributed Lag Model

Time-Series Data

American Politics Examples

i

2016

How Multistate Survival Models Reveal Political Phase Dynamics

Multistate

Survival Analysis

Event-history

Duration

Transitions

i

Bayesian Analysis Reveals Democracy's Boost on Social Spending in Latin America

Bayesian Estimation

Latin America

Social Spending

Democratic Regimes

i

Direct Questions Fall Short, But Indirect Techniques Offer Better Results

Sensitive Questions

List Experiment

Endorsement Experiment

Randomized Response

Bridge Decisions Let Scholars Track Lawmakers Across Time and Chambers

i

Ideal Points

Roll Calls

Cosponsorship

Congress

Bridge Decisions

i

When Controlling Backfires: How Fixed Effects Amplify and Unmask Bias

Omitted Variable

Bias Amplification

Fixed Effects

Sensitivity Analysis

Placebo Studies

i

Parametric Models Flawed: Inaccurate Estimates and Confidence Intervals

Parametric Models

Specification Uncertainty

Statistical Inference

Monte Carlo Simulation

i

Why Binary Treatments Can Triple IV Estimates: The Hidden Bias From Coarse Measurement

Instrumental Variables

Coarsening

Education

Great Britain

Causal Inference

i

Comparing Random Forests With Logistic Regression for Predicting Class-Imbalanced Civil War Onset DataPol. An.

i

Does Your Interaction Effect Really Need a Product Term? New Evidence on Logistic Regression

Product Term

Logistic Regression

Interaction Testing

Conditional Effects

i

How to Project Data When Geographic Boundaries Change

Areal Interpolation

Boundary Change

Spatial Aggregation

GIS

Crosswalks

i

How Much Is the Prior Driving Rare-Event Logistic Results?

Separation

Logistic Regression

Jeffreys Prior

Partial Prior

Rare Events

1 Archives

Representative Samples Don't Guarantee Regression Estimates: A Surprising Flaw

i

Causal Inference

Regression Weights

ANES Data

External Validity

i

Beyond Simple Trade-offs: A New Model for Dynamic Political Compositions

Compositional Variables

Dynamic Modeling

Fixed-sum Constraints

Regression Techniques

Party Support

Budgeting Decisions

i

How Combinatorics Exposes Sensitivity Blind Spots in QCA

QCA

Sensitivity

Combinatorics

Enumeration

Parameters

i

Balancing Equations and Fractional Integration in Dynamic Political Models

Equation Balance

Error Correction

Fractional Integration

Monte Carlo

ARFIMA

i

Faster Ideal Point Estimates with Less Wait for Political Scientists

Ideal Point

Expectation Maximization Algorithm

Massive Data

Variational Inference

i

Fractional Differencing Improves Long-Run Forecasts—Until Short Dynamics and Small Samples Intervene

Fractional Integration

Fractional Differencing

GECM

Simulation

Time Series

1 R files

Why Simple ADL Trumps ARFIMA for Short Fractionally Integrated Series

i

Fractional Integration

ARFIMA

ADL

ECM

Simulation

1 R files

Accounting for Guessing Raises Estimated Learning by 13% and Eliminates Gender Gap

i

Guessing

Latent Class

Measurement

Deliberative Polls

Gender Gap

i

Joint Ideological Scaling for Citizens & Elites? New Methods Address Long-Neglected Challenges

Ideal Point Models

Estimation Method

US House Members

Political Elites

i

Nearly Half of CIPE Results Vanish When Missing Data Is Imputed

Multiple Imputation

Listwise Deletion

Missing Data

Comparative Political Economy

Reanalysis

i

Predicting Race from Voter Rolls Cuts Bias in Turnout Estimates

Ecological Inference

Voter Registration

Bayesian Prediction

Surname List

Turnout

i

When Randomization Isn't Best: Assign Treatments to Minimize Mean Squared Error

Experimental Design

Randomization

Average Treatment Effect

Mean Squared Error

Matlab

13 Text

Speeches, Not Votes: New Way to Map Lawmakers' Positions

i

Legislative Speech

Polarization

Text Analysis

Ireland

United States

i

When Tactics Trump Policy: Roll Call Voting in Parliaments

Ideal Points

Tactical Incentives

Roll Call

Britain

Germany

i

A More Powerful Test for Interference on Social Networks

Interference

Spillover

Randomization

Networks

KS Test

How Machine Learning Identifies Which Anti-Recidivism Policies Work in Colombia

i

Causal Inference

Machine Learning

Ensembles

Recidivism

Colombia

i

What If Legislatures Looked Different? Simulating Counterfactual Representation

MRP

Post-stratification

Proportional Representation

Polarization

Same-sex Marriage

i

Binary Outcomes Aren't Independent—But How Do We Account for That?

Spatiotemporal Interdependence

Binary Outcomes Probit

Maximum Simulated Likelihood

Civil War

i

Theory-Guided QCA Produces Conservative, Not Enhanced, Solutions

QCA

Necessity

T/ESA

Truth Tables

Boolean Minimization

i

When Cheap Surveys Work: Google Consumer Surveys for Causal Experiments

GCS

Survey Experiments

Randomization

Causal Inference

Replication

i

Error-Correction Models Work When Time Series Are Properly Balanced

ECM

Time-series

Fractional Integration

Overfitting

Low Power

i

Mixing Protected and Direct Survey Answers Shows When Sensitive Techniques Are Needed

Crosswise Model

Indirect Questioning

Prevalence Estimation

Corruption

Costa Rica

25 Text

A Practical Fix for Right-Censored Spatial Duration Data

i

2015

Right Censoring

Spatial-lag

Duration Models

Multiple Imputation

Monte Carlo

Novel Approach Tackles Autocorrelation in Repeated Cross-Sections

i

Repeated Cross-sections

ARFIMA Modeling

Multilevel Models

Autocorrelation Correction

A Flawed 1969 Draft Lottery Still Functions as a Random Instrument—With Caveats

i

Draft Lottery

Instrumental Variable

Randomization

ANES

Natural Experiment

i

How to Get Robust Standard Errors With Dyadic Data

Dyadic Data

Sandwich Estimator

Cluster-robust

Logistic Regression

Interstate Disputes

i

Citation Bias in IR Research Depends on Your Statistical Model, Study Finds

Poisson Inverse Gaussian

Citation Data

International Relations

Statistical Specification

Quantitative Precision Meets Qualitative Insight in Comparative Politics

i

Synthetic Control

Quantitative Methods

Comparative Politics

Reunification

i

Bring Multilingual Texts Into Comparative Politics With Topic Models

STM

Machine Translation

Multilingual

Text Analysis

Comparative Politics

i

Measure Election Fraud From Digits Without P-Values

Digit Forensics

Election Fraud

Latent Class

Mixture Index

Beber-Scacco

i

How ‘Zeroing’ Ongoing Years Warps Onset Estimates in Panel Data

Event Onset

Panel Data

Duration

Bias

Civil War

i

How To Spot Measurement Bias In Categorical Latent Models

Measurement Invariance

EPC-interest

Latent Variables

Categorical Data

World Values Survey

MRP Reconstructs Country Means Well but Fails on Rankings and Ranges

i

MRP

Eurobarometer

Poststratification

Multilevel Modeling

Census

17 text files

How QCA Can Produce False Positives — And a Simple Fix

i

QCA

Permutation Test

False Positives

Type I Error

Arab Spring

1 Text

When Robust Standard Errors Reveal Bigger Problems — And How to Fix Them

i

Robust SE

Misspecification

Information Matrix

Simulation

Software

i

Hidden Voter Groups Alter What Explains Immigrant Turnout and Party Choice in Germany

Dirichlet Process

Bayesian Nonparametrics

Unobserved Heterogeneity

Immigrant Voters

Germany

i

When Voters Don't Add Up: Non-Separable Preferences Change Vote Models

Spatial Voting

Nonseparability

Weighted Euclidean

Vote Choice

Monte Carlo

1 Archives

Non-Ignorable Non-Response: A Hidden Variable in Roll-Call Analysis?

i

Non-response Bias

Roll Call Analysis

Legislative Behavior

US Congress

i

RD Design Still Valid for Many Elections: A Global Comparison

Regression Discontinuity Design

Electoral Effects

Incumbency Advantage

Elections

i

Reconsidering Time Transforms in Political Survival Analysis

Proportional Hazards

Survival Analysis

Outlier Times

Censoring Levels

Beyond Statistics: Prioritizing Real-World Significance in Political Research

i

Practical Significance Testing

Context Driven Approach

Descriptive Representation

i

Survey Experiments Often Omit Conditions and Outcomes, Inflating False Positives

Underreporting

Multiple Testing

Preregistration

TESS

Survey Experiments

i

New Scaling Technique Reveals Greater Ideological Polarization

Bayesian Scaling

Differential Item Functioning

ANES Dataset

CCES Survey

Three Ways Qualitative Data Enhances Quantitative Causal Inference

i

Rosenbaum Approach

Qualitative Outcome Data

Measurement Error Reduction

Confidence Intervals Effect Size

Unmeasured Confounders Africa

i

2014

Beyond Flimsy Assertions: A Better Way to Show Variables Have No Political Impact

Statistical Significance

Confidence Intervals

Negligible Effect

i

Testing Political Science Theory Through Confidence Regions

Multidimensional Scaling

Confidence Regions

Bootstrap Method

American National Election Study

i

List vs Endorsement Experiments Converge in Afghanistan: New Statistical Methods Validate Findings

List Experiments

Endorsement Experiments

Representation Gap

Afghanistan

i

PULSAR Teaches Machines to Read Human Rights Reports and Spot Judgments

Human Rights

Machine Learning

Text-as-data

PULSAR

Amnesty International

JHR

Categorical Data Favor Conditional Multiple Imputation Over MVN

i

Multiple Imputation

Multivariate Normal

Conditional Imputation

Categorical Data

ANES

i

How Correcting Latent Class Bias Can Underestimate Uncertainty

Latent Class

Classification Error

Standard Errors

Three-step

Monte Carlo

i

Attention Not Enough? Why Single Survey Screeners Fall Short in Political Studies

Attention

Representation

Self-administered Surveys

i

Semiaautomating Survey Response Analysis

Structural Topic Model

Open-Ended Surveys

Experimental Political Science

Automated Text Analysis

i

Does How We Ask About Political Knowledge Matter? New Insights from Examining Time-Specific vs. General Facts

Political Knowledge

Typology

Gender Differences

Media Effects

Why Common ANES Knowledge Scales Mislead Group Comparisons

i

2013

ANES

Political Knowledge

Measurement Invariance

DIF

CFA

i

Choosing Between Average Effects and Observed-Value Estimates in Political Science Research

Limited Dependent Variable Models

Average Effect

Observed Value Approach

LDV Analysis

2 datasets

Deepening MRP Analysis: How Voter Turnout Changes Differ Across Complex Demographic Subgroups

i

MRP (Multilevel Regression And Poststratification)

Subgroup Interactions

Voter Turnout Patterns

Survey Weighting Techniques

Nonlinearity Modeling

i

How Blogs Chose Links in 2004: A Dynamic Network Logistic Choice Study

Network Dynamics

Logistic Choice

Blogs

2004 Election

Temporal ERGM

i

Small Country Samples: Why Bayesian Wins Over Frequentist Multilevel Models

Multilevel Modeling

Monte Carlo Experiment

Bayesian Estimation

Cross-Level Interactions

i

When Algorithms Read Politics: Promise and Pitfalls of Text-as-Data

Text-as-data

Validation

Machine Learning

Natural Language Processing

Political Texts

Surveys Misread Global Views on Democracy: How Nonsubstantive Responses Skew Results

i

2012

Democracy

Surveys

Response Rates

Data Accuracy

PS

2 Stata files

Better Policy Measurement: Log Odds-Ratio Scaling vs. Current Word Count Approaches

i

2011

Log Odds Ratio

Political Texts Scaling

Comparative Manifesto Project

Policy Dimension Measurement

LSQ

i

Beyond Assumptions: New Paths to Causal Understanding

Causal Mechanism

Minimum Assumption

Estimation Algorithm

Sensitivity Analysis

Multicollinearity Mess-Up? Don't Drop Your Variables!

i

2007

Multicollinearity

Regression Models

Simulated Data

Interaction Terms

When Can History Be Our Guide? The Pitfalls of Counterfactual InferenceISQ

i

i

2004

New Method Tames Roll Call Data: A Bayesian Twist on Spatial Voting Models

Bayesian Methods

Roll Call Data Analysis

Spatial Voting Models

A Simple Regression That Predicts Multiparty Vote Shares As Well As Katz‑King

i

2002

Vote Shares

SUR

Katz-King

Multiparty Elections

Clarify