At the microlevel, comparative public opinion data are abundant. But at the macrolevel—the level where many prominent hypotheses in political behavior are believed to operate—data are scarce. In response, this paper develops a Bayesian dynamic latent trait modeling framework for measuring smooth country–year panels of public opinion even when data are fragmented across time, space, and survey item. Six models are derived from this framework, applied to opinion data on support for democracy, and validated using tests of internal, external, construct, and convergent validity. The best model is reasonably accurate, with predicted responses that deviate from the true response proportions in a held-out test dataset by 6 percentage points. In addition, the smoothed country–year estimates of support for democracy have both construct and convergent validity, with spatiotemporal patterns and associations with other covariates that are consistent with previous research.

Localized network processes are central to the study of political science, whether in the formation of political coalitions and voting blocs, balancing and bandwagoning, policy learning, imitation, diffusion, tipping-point dynamics, or cascade effects. These types of processes are not easily modeled using traditional network approaches, which focus on global rather than local structures within networks. We show that localized network processes, in which network edges form in response to the formation of other edges, are best modeled by shifting from the traditional theoretical framework of nodes-as-actors to what we term a nodes-as-actions framework, which allows for zeroing in on relationships among network connections. We show that the proposed theoretical framework is statistically compatible with a local structure graph model (LSGM). We demonstrate the properties of LSGMs using a Monte Carlo experiment and explore action–reaction processes in two empirical applications: formation of alliances among countries and legislative cosponsorships in the US Senate.

Political scientists are increasingly interested in the geographic distribution of political and economic phenomena. Unlike distribution measures at the individual level, geographic distributions depend on the “unit question” in which researchers choose the appropriate political subdivision to analyze, such as nations, subnational regions, urban and rural areas, or electoral districts. We identify concerns with measuring geographic distribution and comparing distributions within and across political units. In particular, we highlight the potential for threats to inference based on the modifiable areal unit problem (MAUP), whereby measuring concepts at different unit aggregations alters the observed value. We offer tangible options for researchers to improve their research design and data analysis to limit the MAUP. To help manage the measurement error when the unit of observation is unclear or appropriate data are not available, we introduce a new measure of geographic distribution that accounts for fluctuations in the scale and number of political units considered. We demonstrate using Monte Carlo simulations that our measure is more reliable and stable across political units than commonly used indicators because it reduces measurement fluctuations associated with the MAUP.

Complex models are of increasing interest to social scientists. Researchers interested in prediction generally favor flexible, robust approaches, while those interested in causation are often interested in modeling nuanced treatment structures and confounding relationships. Unfortunately, estimators of complex models often scale poorly, especially if they seek to maintain interpretability. In this paper, we present an example of such a conundrum and show how optimization can alleviate the worst of these concerns. Specifically, we introduce bigKRLS, which offers a variety of statistical and computational improvements to the Hainmueller and Hazlett (2013) Kernel-Regularized Least Squares (KRLS) approach. As part of our improvements, we decrease the estimator’s single-core runtime by 50% and reduce the estimator’s peak memory usage by an order of magnitude. We also improve uncertainty estimates for the model’s average marginal effect estimates—which we test both in simulation and in practice—and introduce new visual and statistical tools designed to assist with inference under the model. We further demonstrate the value of our improvements through an analysis of the 2016 presidential election, an analysis that would have been impractical or even infeasible for many users with existing software.

The fixed-effects estimator is biased in the presence of dynamic misspecification and omitted within variation correlated with one of the regressors. We argue and demonstrate that fixed-effects estimates can amplify the bias from dynamic misspecification and that with omitted time-invariant variables and dynamic misspecifications, the fixed-effects estimator can be more biased than the ‘naïve’ OLS model. We also demonstrate that the Hausman test does not reliably identify the least biased estimator when time-invariant and time-varying omitted variables or dynamic misspecifications exist. Accordingly, empirical researchers are ill-advised to rely on the Hausman test for model selection or use the fixed-effects model as default unless they can convincingly justify the assumption of correctly specified dynamics. Our findings caution applied researchers to not overlook the potential drawbacks of relying on the fixed-effects estimator as a default. The results presented here also call upon methodologists to study the properties of estimators in the presence of multiple model misspecifications. Our results suggest that scholars ought to devote much more attention to modeling dynamics appropriately instead of relying on a default solution before they control for potentially omitted variables with constant effects using a fixed-effects specification.

Does attentiveness matter in survey responses? Do more attentive survey participants give higher quality responses? Using data from a recent online survey that identified inattentive respondents using instructed-response items, we demonstrate that ignoring attentiveness provides a biased portrait of the distribution of critical political attitudes and behavior. We show that this bias occurs in the context of both typical closed-ended questions and in list experiments. Inattentive respondents are common and are more prevalent among the young and less educated. Those who do not pass the trap questions interact with the survey instrument in distinctive ways: they take less time to respond; are more likely to report nonattitudes; and display lower consistency in their reported choices. Inattentiveness does not occur completely at random and failing to properly account for it may lead to inaccurate estimates of the prevalence of key political attitudes and behaviors, of both sensitive and more prosaic nature.

Pesaran, Shin, and Smith (2001) (PSS) proposed a bounds procedure for testing for the existence of long run cointegrating relationships between a unit root dependent variable ($y_{t}$) and a set of weakly exogenous regressors $\boldsymbol{x}_{t}$ when the analyst does not know whether the independent variables are stationary, unit root, or mutually cointegrated processes. This procedure recognizes the analyst’s uncertainty over the nature of the regressors but not the dependent variable. When the analyst is uncertain whether $y_{t}$ is a stationary or unit root process, the test statistics proposed by PSS are uninformative for inference on the existence of a long run relationship (LRR) between $y_{t}$ and $\boldsymbol{x}_{t}$. We propose the long run multiplier (LRM) test statistic as a means of testing for LRRs without knowing whether the series are stationary or unit roots. Using stochastic simulations, we demonstrate the behavior of the test statistic given uncertainty about the univariate dynamics of both $y_{t}$ and $\boldsymbol{x}_{t}$, illustrate the bounds of the test statistic, and generate small sample and approximate asymptotic critical values for the upper and lower bounds for a range of sample sizes and model specifications. We demonstrate the utility of the bounds framework for testing for LRRs in models of public policy mood and presidential success.

We develop a new Bayesian split population survival model for the analysis of survival data with misclassified event failures. Within political science survival data, right-censored survival cases are often erroneously misclassified as failure cases due to measurement error. Treating these cases as failure events within survival analyses will underestimate the duration of some events. This will bias coefficient estimates, especially in situations where such misclassification is associated with covariates of interest. Our split population survival estimator addresses this challenge by using a system of two equations to explicitly model the misclassification of failure events alongside a parametric survival process of interest. After deriving this model, we use Bayesian estimation via slice sampling to evaluate its performance with simulated data, and in several political science applications. We find that our proposed “misclassified failure” survival model allows researchers to accurately account for misclassified failure events within the contexts of civil war duration and democratic survival.

We show that propensity score matching (PSM), an enormously popular method of preprocessing data for causal inference, often accomplishes the opposite of its intended goal—thus increasing imbalance, inefficiency, model dependence, and bias. The weakness of PSM comes from its attempts to approximate a completely randomized experiment, rather than, as with other matching methods, a more efficient fully blocked randomized experiment. PSM is thus uniquely blind to the often large portion of imbalance that can be eliminated by approximating full blocking with other matching methods. Moreover, in data balanced enough to approximate complete randomization, either to begin with or after pruning some observations, PSM approximates random matching which, we show, increases imbalance even relative to the original data. Although these results suggest researchers replace PSM with one of the other available matching methods, propensity scores have other productive uses.

Multiplicative interaction models are widely used in social science to examine whether the relationship between an outcome and an independent variable changes with a moderating variable. Current empirical practice tends to overlook two important problems. First, these models assume a linear interaction effect that changes at a constant rate with the moderator. Second, estimates of the conditional effects of the independent variable can be misleading if there is a lack of common support of the moderator. Replicating 46 interaction effects from 22 recent publications in five top political science journals, we find that these core assumptions often fail in practice, suggesting that a large portion of findings across all political science subfields based on interaction models are fragile and model dependent. We propose a checklist of simple diagnostics to assess the validity of these assumptions and offer flexible estimation strategies that allow for nonlinear interaction effects and safeguard against excessive extrapolation. These statistical routines are available in both R and STATA.

Researchers who generate data often optimize efficiency and robustness by choosing stratified over simple random sampling designs. Yet, all theories of inference proposed to justify matching methods are based on simple random sampling. This is all the more troubling because, although these theories require exact matching, most matching applications resort to some form of ex post stratification (on a propensity score, distance metric, or the covariates) to find approximate matches, thus nullifying the statistical properties these theories are designed to ensure. Fortunately, the type of sampling used in a theory of inference is an axiom, rather than an assumption vulnerable to being proven wrong, and so we can replace simple with stratified sampling, so long as we can show, as we do here, that the implications of the theory are coherent and remain true. Properties of estimators based on this theory are much easier to understand and can be satisfied without the unattractive properties of existing theories, such as assumptions hidden in data analyses rather than stated up front, asymptotics, unfamiliar estimators, and complex variance calculations. Our theory of inference makes it possible for researchers to treat matching as a simple form of preprocessing to reduce model dependence, after which all the familiar inferential techniques and uncertainty calculations can be applied. This theory also allows binary, multicategory, and continuous treatment variables from the outset and straightforward extensions for imperfect treatment assignment and different versions of treatments.

According to spatial models of political competition, parties strategically adjust their ideological positions to movements made by rival parties. Spatial econometric techniques have been proposed to empirically model such interdependencies and to closely convert theoretical expectations into statistical models. Yet, these models often ignore that the parties’ ideological positions are latent variables and, as such, accompanied by a quantifiable amount of uncertainty. As a result, the implausible assumption of perfectly measured covariates impedes a proper evaluation of theoretical propositions. In order to bridge this gap between theory and empirics, the present work combines a spatial econometric model and a Bayesian dynamic item response model. The proposed model accurately accounts for measurement uncertainty and simultaneously estimates the parties’ ideological positions and their spatial interdependencies. To verify the model’s utility, I apply it to recorded votes from the sixteen German state legislatures in the period from 1988 to 2016. While exhibiting a notable degree of ideological mobility, the results indicate only moderate spatial dependencies among parties of the same party family. More importantly, the analysis illustrates how measurement uncertainty can lead to substantively different results which stresses the importance of appropriately incorporating theoretical expectations into statistical models.

Measurement error threatens the validity of survey research, especially when studying sensitive questions. Although list experiments can help discourage deliberate misreporting, they may also suffer from nonstrategic measurement error due to flawed implementation and respondents’ inattention. Such error runs against the assumptions of the standard maximum likelihood regression (MLreg) estimator for list experiments and can result in misleading inferences, especially when the underlying sensitive trait is rare. We address this problem by providing new tools for diagnosing and mitigating measurement error in list experiments. First, we demonstrate that the nonlinear least squares regression (NLSreg) estimator proposed in Imai (2011) is robust to nonstrategic measurement error. Second, we offer a general model misspecification test to gauge the divergence of the MLreg and NLSreg estimates. Third, we show how to model measurement error directly, proposing new estimators that preserve the statistical efficiency of MLreg while improving robustness. Last, we revisit empirical studies shown to exhibit nonstrategic measurement error, and demonstrate that our tools readily diagnose and mitigate the bias. We conclude this article with a number of practical recommendations for applied researchers. The proposed methods are implemented through an open-source software package.

This paper creates a multinomial framework for ideal point estimation (mIRT) using recent developments in Bayesian statistics. The core model relies on a flexible multinomial specification that includes most common models in political science as “special cases.” I show that popular extensions (e.g., dynamic smoothing, inclusion of covariates, and network models) can be easily incorporated whilst maintaining the ability to estimate a model using a Gibbs Sampler or exact EM algorithm. By showing that these models can be written and estimated using a shared framework, the paper aims to reduce the proliferation of bespoke ideal point models as well as extend the ability of applied researchers to estimate models quickly using the EM algorithm. I apply this framework to a thorny question in scaling survey responses—the treatment of nonresponse. Focusing on the American National Election Study (ANES), I suggest that a simple but principled solution is to treat questions as multinomial where nonresponse is a distinct (modeled) category. The exploratory results suggest that certain questions tend to attract many more invalid answers and that many of these questions (particularly when signaling out particular social groups for evaluation) are masking noncentrist (typically conservative) beliefs.

Surveys are a key tool for understanding political behavior, but they are subject to biases that render their estimates about the frequency of socially desirable behaviors inaccurate. For decades the American National Election Study (ANES) has overestimated voter turnout, though the causes of this persistent bias are poorly understood. The face-to-face component of the 2012 ANES produced a turnout estimate at least 13 points higher than the benchmark voting-eligible population turnout rate. We consider three explanations for this overestimate in the survey: nonresponse bias, over-reporting and the possibility that the ANES constitutes an inadvertent mobilization treatment. Analysis of turnout data supplied by voter file vendors allows the three phenomena to be measured for the first time in a single survey. We find that over-reporting is the largest contributor, responsible for six percentage points of the turnout overestimate, while nonresponse bias and mobilization account for an additional 4 and 3 percentage points, respectively.

This paper introduces an instrumental variables framework for analyzing how external factors that affect survey response rates can also affect the composition of the sample of respondents. The method may be useful for studying survey representativeness, and for assessing the effectiveness of some of the conventional corrections for survey nonresponse bias.

The paper applies the method to data collected in the 2011 Swiss Electoral Study (SES), in which survey participation incentives were randomly assigned across members of the original survey sample. The empirical analysis shows that the incentives increased response rates substantially. Estimates of a new instrumental variable parameter called the Complier Average Survey Response (CASR) suggest that the incentives induced participation among people with more nationalist political opinions than those who would have participated without the incentives. Weighting the respondent data to match the covariate distribution in the target population did not account for the discrepancy in attitudes between the two groups, suggesting that the weights would not succeed in removing nonresponse bias.

The analysis of political texts from parliamentary speeches, party manifestos, social media, or press releases forms the basis of major and growing fields in political science, not least since advances in “text-as-data” methods have rendered the analysis of large text corpora straightforward. However, a lot of sources of political speech are not regularly transcribed, and their on-demand transcription by humans is prohibitively expensive for research purposes. This class includes political speech in certain legislatures, during political party conferences as well as television interviews and talk shows. We showcase how scholars can use automatic speech recognition systems to analyze such speech with quantitative text analysis models of the “bag-of-words” variety. To probe results for robustness to transcription error, we present an original “word error rate simulation” (WERSIM) procedure implemented in $R$. We demonstrate the potential of automatic speech recognition to address open questions in political science with two substantive applications and discuss its limitations and practical challenges.

Opinion surveys often employ multiple items to measure the respondent’s underlying value, belief, or attitude. To analyze such types of data, researchers have often followed a two-step approach by first constructing a composite measure and then using it in subsequent analysis. This paper presents a class of hierarchical item response models that help integrate measurement and analysis. In this approach, individual responses to multiple items stem from a latent preference, of which both the mean and variance may depend on observed covariates. Compared with the two-step approach, the hierarchical approach reduces bias, increases efficiency, and facilitates direct comparison across surveys covering different sets of items. Moreover, it enables us to investigate not only how preferences differ among groups, vary across regions, and evolve over time, but also levels, patterns, and trends of attitude polarization and ideological constraint. An open-source R package, hIRT, is available for fitting the proposed models.

Are student subject experiment pools comparable across institutions? Despite repeated concerns over the “college sophomore problem,” many experiment-based studies still rely on student subject pools due to their convenience and accessibility. In this paper, I investigate whether student subject pools are comparable across universities by examining how respondents across three student subject pools at distinct educational institutions perform on the same survey experiment about crisis bargaining between states. I argue that, due to selection biases inherent in university matriculation and the self-selection of students into experimental protocols, respondents across these subject pools will exhibit key demographic differences. I also examine whether respondents across these subject pools think similarly about international politics and respond comparably to experimental treatments. I find that, while there are significant demographic differences across subject pools, subjects across institutions respond similarly to experimental treatments—with the key exception of information regarding the regime type of a state. Furthermore, there is little evidence that these demographic differences impact conditional average treatment effects across subgroups. These findings carry critical implications for the use of student samples across political science and within international relations more specifically, particularly regarding the current replication crisis in the discipline.

We introduce a Bayesian approach to conduct inferential analyses on dyadic data while accounting for interdependencies between observations through a set of additive and multiplicative effects (AME). The AME model is built on a generalized linear modeling framework and is thus flexible enough to be applied to a variety of contexts. We contrast the AME model to two prominent approaches in the literature: the latent space model (LSM) and the exponential random graph model (ERGM). Relative to these approaches, we show that the AME approach is (a) to be easy to implement; (b) interpretable in a general linear model framework; (c) computationally straightforward; (d) not prone to degeneracy; (e) captures first-, second-, and third-order network dependencies; and (f) notably outperforms ERGMs and LSMs on a variety of metrics and in an out-of-sample context. In summary, AME offers a straightforward way to undertake nuanced, principled inferential network analysis for a wide range of social science questions.