We present a hierarchical Dirichlet regression model with Gaussian process priors that enables accurate and well-calibrated forecasts for U.S. Senate elections at varying time horizons. This Bayesian model provides a balance between predictions based on time-dependent opinion polls and those made based on fundamentals. It also provides uncertainty estimates that arise naturally from historical data on elections and polls. Experiments show that our model is highly accurate and has a well calibrated coverage rate for vote share predictions at various forecasting horizons. We validate the model with a retrospective forecast of the 2018 cycle as well as a true out-of-sample forecast for 2020. We show that our approach achieves state-of-the art accuracy and coverage despite relying on few covariates.

Topic models, as developed in computer science, are effective tools for exploring and summarizing large document collections. When applied in social science research, however, they are commonly used for measurement, a task that requires careful validation to ensure that the model outputs actually capture the desired concept of interest. In this paper, we review current practices for topic validation in the field and show that extensive model validation is increasingly rare, or at least not systematically reported in papers and appendices. To supplement current practices, we refine an existing crowd-sourcing method by Chang and coauthors for validating topic quality and go on to create new procedures for validating conceptual labels provided by the researcher. We illustrate our method with an analysis of Facebook posts by U.S. Senators and provide software and guidance for researchers wishing to validate their own topic models. While tailored, case-specific validation exercises will always be best, we aim to improve standard practices by providing a general-purpose tool to validate topics as measures.

Political elites sometimes seek to delegitimize election results using unsubstantiated claims of fraud. Most recently, Donald Trump sought to overturn his loss in the 2020 US presidential election by falsely alleging widespread fraud. Our study provides new evidence demonstrating the corrosive effect of fraud claims like these on trust in the election system. Using a nationwide survey experiment conducted after the 2018 midterm elections – a time when many prominent Republicans also made unsubstantiated fraud claims – we show that exposure to claims of voter fraud reduces confidence in electoral integrity, though not support for democracy itself. The effects are concentrated among Republicans and Trump approvers. Worryingly, corrective messages from mainstream sources do not measurably reduce the damage these accusations inflict. These results suggest that unsubstantiated voter-fraud claims undermine confidence in elections, particularly when the claims are politically congenial, and that their effects cannot easily be mitigated by fact-checking.

Bivariate probit models are a common choice for scholars wishing to estimate causal effects in instrumental variable models where both the treatment and outcome are binary. However, standard maximum likelihood approaches for estimating bivariate probit models are problematic. Numerical routines in popular software suites frequently generate inaccurate parameter estimates and even estimated correctly, maximum likelihood routines provide no straightforward way to produce estimates of uncertainty for causal quantities of interest. In this note, we show that adopting a Bayesian approach provides more accurate estimates of key parameters and facilitates the direct calculation of causal quantities along with their attendant measures of uncertainty.

Scholars are increasingly utilizing online workforces to encode latent political concepts embedded in written or spoken records. In this letter, we build on past efforts by developing and validating a crowdsourced pairwise comparison framework for encoding political texts that combines the human ability to understand natural language with the ability of computers to aggregate data into reliable measures while ameliorating concerns about the biases and unreliability of non-expert human coders. We validate the method with advertisements for U.S. Senate candidates and with State Department reports on human rights. The framework we present is very general, and we provide free software to help applied researchers interact easily with online workforces to extract meaningful measures from texts.

Electoral forensics involves examining election results for anomalies to efficiently identify patterns indicative of electoral irregularities. However, there is disagreement about which, if any, forensics tool is most effective at identifying fraud, and there is no method for integrating multiple tools. Moreover, forensic efforts have failed to systematically take advantage of country-specific details that might aid in diagnosing fraud. We deploy a Bayesian additive regression trees (BART) model–a machine-learning technique–on a large cross-national data set to explore the dense network of potential relationships between various forensic indicators of anomalies and electoral fraud risk factors, on the one hand, and the likelihood of fraud, on the other. This approach allows us to arbitrate between the relative importance of different forensic and contextual features for identifying electoral fraud and results in a diagnostic tool that can be relatively easily implemented in cross-national research.

Political science researchers typically conduct an idiosyncratic search of possible model configurations and then present a single specification to readers. This approach systematically understates the uncertainty of our results, generates fragile model specifications, and leads to the estimation of bloated models with too many control variables. Bayesian model averaging (BMA) offers a systematic method for analyzing specification uncertainty and checking the robustness of one's results to alternative model specifications, but it has not come into wide usage within the discipline. In this paper, we introduce important recent developments in BMA and show how they enable a different approach to using the technique in applied social science research. We illustrate the methodology by reanalyzing data from three recent studies using BMA software we have modified to respect statistical conventions within political science.

Survey researchers avoid using large multi-item scales to measure latent traits due to both the financial costs and the risk of driving up nonresponse rates. Typically, investigators select a subset of available scale items rather than asking the full battery. Reduced batteries, however, can sharply reduce measurement precision and introduce bias. In this article, we present computerized adaptive testing (CAT) as a method for minimizing the number of questions each respondent must answer while preserving measurement accuracy and precision. CAT algorithms respond to individuals' previous answers to select subsequent questions that most efficiently reveal respondents' positions on a latent dimension. We introduce the basic stages of a CAT algorithm and present the details for one approach to item selection appropriate for public opinion research. We then demonstrate the advantages of CAT via simulation and empirically comparing dynamic and static measures of political knowledge.

We present ensemble Bayesian model averaging (EBMA) and illustrate its ability to aid scholars in the social sciences to make more accurate forecasts of future events. In essence, EBMA improves prediction by pooling information from multiple forecast models to generate ensemble predictions similar to a weighted average of component forecasts. The weight assigned to each forecast is calibrated via its performance in some validation period. The aim is not to choose some “best” model, but rather to incorporate the insights and knowledge implicit in various forecasting efforts via statistical postprocessing. After presenting the method, we show that EBMA increases the accuracy of out-of-sample forecasts relative to component models in three applied examples: predicting the occurrence of insurgencies around the Pacific Rim, forecasting vote shares in U.S. presidential elections, and predicting the votes of U.S. Supreme Court Justices.

In this article, we challenge the conclusion that the preferences of members of Congress are best represented as existing in a low-dimensional space. We conduct Monte Carlo simulations altering assumptions regarding the dimensionality and distribution of member preferences and scale the resulting roll call matrices. Our simulations show that party polarization generates misleading evidence in favor of low dimensionality. This suggests that the increasing levels of party polarization in recent Congresses may have produced false evidence in favor of a low-dimensional policy space. However, we show that focusing more narrowly on each party caucus in isolation can help researchers discern the true dimensionality of the policy space in the context of significant party polarization. We re-examine the historical roll call record and find evidence suggesting that the low dimensionality of the contemporary Congress may reflect party polarization rather than changes in the dimensionality of policy conflict.

John Aldrich is a positive scientist—in both the scholastic and colloquial sense. A progeny of the Rochester school, Aldrich's research displays a commitment to the tenets of positive political theory. He derives internally consistent propositions and subjects those claims to empirical testing, all in an attempt to explain scientifically phenomena and institutions at the heart of democratic theory. As a mentor and builder of academic institutions, Aldrich has shown unswerving kindness, modesty, and a commitment to foster new generations of political scientists. He is a positive influence on his students, his colleagues, and the political science discipline at large.

Perhaps the major story in forecasting the 2012 election is the growing awareness of the benefits of aggregating multiple sources of data for improving prediction. Most prominently, polling analysts including Simon Jackman, Drew Linzer, and Nate Silver made the strong case, ultimately validated by the election results, that combining information from multiple polls gives a better picture of the electorate than any poll analyzed in isolation.

For more than two decades, political scientists have created statistical modelsaimed at generating out-of-sample predictions of presidential elections. In 2004 and2008, PS: Political Science and Politics published symposia of thevarious forecasting models prior to Election Day. This exercise serves to validatemodels based on accuracy by garnering additional support for those that mostaccurately foretell the ultimate election outcome. Implicitly, these symposia assertthat accurate models best capture the essential contexts and determinants ofelections. In part, therefore, this exercise aims to develop the “best” model of theunderlying data generating process. Scholars comparatively evaluate their models bysetting their predictions against electoral results while also giving some attentionto the models' inherent plausibility, parsimony, and beauty.