Suchergebnisse

Multilevel regression and post-stratification (MRP) is a popular use of hierarchical models in political science. Multiple papers have suggested that relying on machine learning methods can provide substantially better performance than traditional approaches that use hierarchical models. However, these comparisons are often unfair to traditional techniques as they omit possibly important interactions or nonlinear effects. I show that complex ("deep") hierarchical models that include interactions can nearly match or outperform state-of-the-art machine learning methods. Combining multiple models into an ensemble can improve performance, although deep hierarchical models are themselves given considerable weight in these ensembles. The main limitation to using deep hierarchical models is speed. This paper derives new techniques to further accelerate estimation using variational approximations. I provide software that uses weakly informative priors and can estimate nonlinear effects using splines. This allows flexible and complex hierarchical models to be fit as quickly as many comparable machine learning techniques.

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.

Much of the data used in social science is aggregated into spatial units, even if the analysis itself does not explicitly incorporate that information. A key concern with such aggregation, however, is that changes in the units of aggregation themselves cause difficulty in comparing data gathered on the old boundaries and the new boundaries. Such changes present serious concerns to researchers who may exclude observations or cases due to a lack of comparable units or omit certain key variables. While geographers have long examined this problem and created methods of projecting data from one spatial unit into another, known as areal interpolation, it is telling that a recent article notes the difficulty for researchers in implementing even the most basic solutions without relying heavily on programming skills or proprietary software.

AbstractKernel regularized least squares (KRLS) is a popular method for flexibly estimating models that may have complex relationships between variables. However, its usefulness to many researchers is limited for two reasons. First, existing approaches are inflexible and do not allow KRLS to be combined with theoretically motivated extensions such as random effects, unregularized fixed effects, or non-Gaussian outcomes. Second, estimation is extremely computationally intensive for even modestly sized datasets. Our paper addresses both concerns by introducing generalized KRLS (gKRLS). We note that KRLS can be re-formulated as a hierarchical model thereby allowing easy inference and modular model construction where KRLS can be used alongside random effects, splines, and unregularized fixed effects. Computationally, we also implement random sketching to dramatically accelerate estimation while incurring a limited penalty in estimation quality. We demonstrate that gKRLS can be fit on datasets with tens of thousands of observations in under 1 min. Further, state-of-the-art techniques that require fitting the model over a dozen times (e.g., meta-learners) can be estimated quickly.

Most work on redistribution in democracies is anchored in long-standing unidimensional models, notably the seminal Meltzer-Richard-Romer model. When scholars venture outside the security of unidimensionality, many either abandon theoretical rigor or miss the full consequences of adding more dimensions (whether ideological or economic). There is now a substantial literature on redistributive politics in multidimensional policy spaces, but it tends to be very technical and frequently misinterpreted, if not ignored. This purpose of this article is to review this relatively new literature using simple graphical representations, focusing on the key assumptions, intuitions, and results. We show how issue bundling, issue salience, and the distribution of preferences can affect redistribution, and we discuss the role of political institutions in inducing particular outcomes. We also highlight the opportunities for dialogue between formal models and more constructivist approaches by exploring the effects of political entrepreneurs manipulating salience, institutions, and even identities.

Assembly dissolution is a key dimension of constitutional variation in parliamentary democracies. It conditions the timing of elections, influences electoral accountability, and shapes how politicians bargain about government formation, termination, and policy. Yet, despite the importance of dissolution rules, no measure exists that applies to the different actors involved in dissolutions and records the complexity of the rules sufficiently accurately to capture their substantive implications. This article develops, operationalizes, and tests a detailed index of parliamentary dissolution powers that generalizes to all relevant actors. We demonstrate the substantive utility of the index by examining how election timing powers modify electoral accountability.

AbstractA key feature of parliamentary democracy is government accountability vis‐à‐vis the legislature, but the important question of who speaks for the government—cabinet ministers or unelected bureaucrats, and the institutional underpinnings of this behavior—receives scant attention in the existing literature. We investigate this question using the case of Japan, and data on more than four million committee speeches spanning distinct electoral and legislative institutional environments. We document how a party‐strengthening electoral system reform in 1994 facilitated a dramatic shift in the nature of government accountability to parliamentary committees: speeches by ministers increased, speeches by bureaucrats decreased, and discursive accountability between ministers and opposition legislators increased. Subsequent legislative reforms expanding junior ministerial roles and placing explicit limits on bureaucratic participation further reinforced the effects. These findings shed new light on the institutional foundations of responsible party government in general as well as its progressive development in Japan.

Political parties and legislators use legislative debates to establish their reputation, challenge rivals, and engage in coalition management, among many other tasks. Yet, existing theories on parliamentary debates have abstracted away from the need for information and expertise, which are costly to acquire. Drawing on the "informational" perspective on legislative organization, we address this problem by arguing that party leaders use committees as training arenas for their backbenchers. They task their assigned members with acquiring specific expertise and then rely heavily on those members during the corresponding debates. We turn to the Portuguese legislature, from 2000 to 2015, to discuss how saliency, government dynamics, and party size affect the use of experts. We test this theory using a novel approach to classify speeches that leverages the texts of legislation as training data for a supervised approach.