The output of predictive models is routinely recalibrated by reconciling low-level predictions with known quantities defined at higher levels of aggregation. For example, models predicting vote probabilities at the individual level in U.S. elections can be adjusted so that their aggregation matches the observed vote totals in each county, thus producing better-calibrated predictions. In this research note, we provide theoretical grounding for one of the most commonly used recalibration strategies, known colloquially as the “logit shift.” Typically cast as a heuristic adjustment strategy (whereby a constant correction on the logit scale is found, such that aggregated predictions match target totals), we show that the logit shift offers a fast and accurate approximation to a principled, but computationally impractical adjustment strategy: computing the posterior prediction probabilities, conditional on the observed totals. After deriving analytical bounds on the quality of the approximation, we illustrate its accuracy using Monte Carlo simulations. We also discuss scenarios in which the logit shift is less effective at recalibrating predictions: when the target totals are defined only for highly heterogeneous populations, and when the original predictions correctly capture the mean of true individual probabilities, but fail to capture the shape of their distribution.

A common approach when studying the quality of representation involves comparing the latent preferences of voters and legislators, commonly obtained by fitting an item response theory (IRT) model to a common set of stimuli. Despite being exposed to the same stimuli, voters and legislators may not share a common understanding of how these stimuli map onto their latent preferences, leading to differential item functioning (DIF) and incomparability of estimates. We explore the presence of DIF and incomparability of latent preferences obtained through IRT models by reanalyzing an influential survey dataset, where survey respondents expressed their preferences on roll call votes that U.S. legislators had previously voted on. To do so, we propose defining a Dirichlet process prior over item response functions in standard IRT models. In contrast to typical multistep approaches to detecting DIF, our strategy allows researchers to fit a single model, automatically identifying incomparable subgroups with different mappings from latent traits onto observed responses. We find that although there is a group of voters whose estimated positions can be safely compared to those of legislators, a sizeable share of surveyed voters understand stimuli in fundamentally different ways. Ignoring these issues can lead to incorrect conclusions about the quality of representation.

Supervised topic classification requires labeled data. This often becomes a bottleneck as high-quality labeled data are expensive to acquire. To overcome the data scarcity problem, scholars have recently proposed to use cross-domain topic classification to take advantage of preexisting labeled datasets. Cross-domain topic classification only requires limited annotation in the target domain to verify its cross-domain accuracy. In this letter, we propose supervised topic classification with pretrained language models as an alternative. We show that language models fine-tuned with 70% of the small annotated dataset in the target corpus could outperform models trained using large cross-domain datasets by 27% and that models fine-tuned with 10% of the annotated dataset could already outperform the cross-domain classifiers. Our models are competitive in terms of training time and inference time. Researchers interested in supervised learning with limited labeled data should find our results useful. Our code and data are publicly available.1

A critical barrier to generating cumulative knowledge in political science and related disciplines is the inability of researchers to observe the results from the full set of research designs that scholars have conceptualized, implemented, and analyzed. For a variety of reasons, studies that produce null findings are especially likely to be unobserved, creating biases in publicly accessible research. While several approaches have been suggested to overcome this problem, none have yet proven adequate. We call for the establishment of a new discipline-wide norm in which scholars post short “null results reports” online that summarize their research designs, findings, and interpretations. To address the inevitable incentive problems that earlier proposals for reform were unable to overcome, we argue that decentralized research communities can spur the broader disciplinary norm change that would bring advantage to scientific advance. To facilitate our contribution, we offer a template for these reports that incorporates evaluation of the possible explanations for the null findings, including statistical power, measurement strategy, implementation issues, spillover/contamination, and flaws in theoretical priors. We illustrate the template’s utility with two experimental studies focused on the naturalization of immigrants in the United States and attitudes toward Syrian refugees in Jordan.

We introduce hierarchically regularized entropy balancing as an extension to entropy balancing, a reweighting method that adjusts weights for control group units to achieve covariate balance in observational studies with binary treatments. Our proposed extension expands the feature space by including higher-order terms (such as squared and cubic terms and interactions) of covariates and then achieves approximate balance on the expanded features using ridge penalties with a hierarchical structure. Compared with entropy balancing, this extension relaxes model dependency and improves the robustness of causal estimates while avoiding optimization failure or highly concentrated weights. It prevents specification searches by minimizing user discretion in selecting features to balance on and is also computationally more efficient than kernel balancing, a kernel-based covariate balancing method. We demonstrate its performance through simulations and an empirical example. We develop an open-source R package, hbal, to facilitate implementation.

Understanding the gaps and connections across existing theories and findings is a perennial challenge in scientific research. Systematically reviewing scholarship is especially challenging for researchers who may lack domain expertise, including junior scholars or those exploring new substantive territory. Conversely, senior scholars may rely on long-standing assumptions and social networks that exclude new research. In both cases, ad hoc literature reviews hinder accumulation of knowledge. Scholars are rarely systematic in selecting relevant prior work or then identifying patterns across their sample. To encourage systematic, replicable, and transparent methods for assessing literature, we propose an accessible network-based framework for reviewing scholarship. In our method, we consider a literature as a network of recurring concepts (nodes) and theorized relationships among them (edges). Network statistics and visualization allow researchers to see patterns and offer reproducible characterizations of assertions about the major themes in existing literature. Critically, our approach is systematic and powerful but also low cost; it requires researchers to enter relationships they observe in prior studies into a simple spreadsheet—a task accessible to new and experienced researchers alike. Our open-source R package enables researchers to leverage powerful network analysis while minimizing software-specific knowledge. We demonstrate this approach by reviewing redistricting literature.

Ensuring descriptive representation of racial minorities without packing minorities too heavily into districts is a perpetual difficulty, especially in states lacking voter file race data. One advance since the 2010 redistricting cycle is the advent of Bayesian Improved Surname Geocoding (BISG), which greatly improves upon previous ecological inference methods in identifying voter race. In this article, we test the viability of employing BISG to redistricting under two posterior allocation methods for race assignment: plurality versus probabilistic. We validate these methods through 10,000 redistricting simulations of North Carolina and Georgia’s congressional districts and compare BISG estimates to actual voter file racial data. We find that probabilistic summing of the BISG posteriors significantly reduces error rates at the precinct and district level relative to plurality racial assignment, and therefore should be the preferred method when using BISG for redistricting. Our results suggest that BISG can aid in the construction of majority-minority districts during the redistricting process.

Contagion across various types of connections is a central process in the study of many political phenomena (e.g., democratization, civil conflict, and voter turnout). Over the last decade, the methodological literature addressing the challenges in causally identifying contagion in networks has exploded. In one of the foundational works in this literature, Shalizi and Thomas (2011, Sociological Methods and Research 40, 211–239.) propose a permutation test for contagion in longitudinal network data that is not confounded by selection (e.g., homophily). We illustrate the properties of this test via simulation. We assess its statistical power under various conditions of the data, including the nature of the contagion, the structure of the network through which contagion occurs, and the number of time periods included in the data. We then apply this test to an example domain that is commonly considered in the context of observational research on contagion—the international spread of democracy. We find evidence of international contagion of democracy. We conclude with a discussion of the practical applicability of the Shalizi and Thomas test to the study of contagion in political networks.