Thomas Robinson

Abstract Principled methods for analyzing missing values, based chiefly on multiple imputation, have become increasingly popular yet can struggle to handle the kinds of large and complex data that are also becoming common. We propose an accurate, fast, and scalable approach to multiple imputation, which we call MIDAS (Multiple Imputation with Denoising Autoencoders). MIDAS employs a class of unsupervised neural networks known as denoising autoencoders, which are designed to reduce dimensionality by corrupting and attempting to reconstruct a subset of data. We repurpose denoising autoencoders for multiple imputation by treating missing values as an additional portion of corrupted data and drawing imputations from a model trained to minimize the reconstruction error on the originally observed portion. Systematic tests on simulated as well as real social science data, together with an applied example involving a large-scale electoral survey, illustrate MIDAS’s accuracy and efficiency across a range of settings. We provide open-source software for implementing MIDAS.

As the results were finalized, LeDuc’s law of referenda—that voters increasingly favor the status quo as referendum campaigns wear on—had clearly failed to prevent Britain’s dramatic exit from the European Union (EU). Despite apocalyptic warnings from those who supported the status quo, 51.9 percent of voters chose to endorse “Brexit.” Instant reactions presented the result as a shock, but Harold D. Clarke, Matthew Goodwin, and Paul Whiteley’s book analyzing the factors that led up to this monumental decision makes it clear that a “leave” vote was far more likely than many expected. The book’s most substantial contribution to the nascent Brexit literature is a series of models developed to analyze the factors behind voters’ support and opposition to the EU. The empirical evidence presented by the authors is wide-ranging and addresses the key issues leading up to the referendum: both direct and indirect attitudes toward the EU (Chapter 4), support for the United Kingdom Independence Party (Chapters 5 and 6), and the drivers of vote choice in the referendum itself (Chapter 7).

How does the public want a COVID-19 vaccine to be allocated? We conducted a conjoint experiment asking 15,536 adults in 13 countries to evaluate 248,576 profiles of potential vaccine recipients who varied randomly on five attributes. Our sample includes diverse countries from all continents. The results suggest that in addition to giving priority to health workers and to those at high risk, the public favors giving priority to a broad range of key workers and to those with lower income. These preferences are similar across respondents of different education levels, incomes, and political ideologies, as well as across most surveyed countries. The public favored COVID-19 vaccines being allocated solely via government programs but were highly polarized in some developed countries on whether taking a vaccine should be mandatory. There is a consensus among the public on many aspects of COVID-19 vaccination, which needs to be taken into account when developing and communicating rollout strategies.

Conjoint experiments are fast becoming one of the dominant experimental methods within the social sciences. Despite recent efforts to model heterogeneity within this type of experiment, the relationship between the conjoint design and lower-level causal estimands is underdeveloped. In this article, we clarify how conjoint heterogeneity can be construed as a set of nested, causal parameters that correspond to the levels of the conjoint design. We then use this framework to propose a new estimation strategy, using machine learning, that better allows researchers to evaluate treatment effect heterogeneity. We also provide novel tools for classifying and analyzing heterogeneity postestimation using partitioning algorithms. Replicating two conjoint experiments, we demonstrate our theoretical argument and show how this method helps estimate and detect substantive patterns of heterogeneity. To accompany this article, we provide new a R package, cjbart, that allows researchers to model heterogeneity in their experimental conjoint data.