Search

Elected officials have privileged roles in public communication. In contrast to national politicians, whose posting content is more likely to be closely scrutinized by a robust ecosystem of nationally focused media outlets, sub-national politicians are more likely to openly disseminate harmful content with limited media scrutiny. In this paper, we analyze the factors that explain the online visibility of over 6.5K unique state legislators in the US and how their visibility might be impacted by posting low-credibility or uncivil content. We conducted a study of posting on Twitter and Facebook (FB) during 2020-21 to analyze how legislators engage with users on these platforms. The results indicate that distributing content with low-credibility information attracts greater attention from users on FB and Twitter for Republicans. Conversely, posting content that is considered uncivil on Twitter receives less attention. A noticeable scarcity of posts containing uncivil content was observed on FB, which may be attributed to the different communication patterns of legislators on these platforms. In most cases, the effect is more pronounced among the most ideologically extreme legislators. Our research explores the influence exerted by state legislators on online political conversations, with Twitter and FB serving as case studies. Furthermore, it sheds light on the differences in the conduct of political actors on these platforms. This study contributes to a better understanding of the role that political figures play in shaping online political discourse., Comment: Accepted for publication in the International AAAI Conference on Web and Social Media (ICWSM) 2025

The significance and influence of US Supreme Court majority opinions derive in large part from opinions' roles as precedents for future opinions. A growing body of literature seeks to understand what drives the use of opinions as precedents through the study of Supreme Court case citation patterns. We raise two limitations of existing work on Supreme Court citations. First, dyadic citations are typically aggregated to the case level before they are analyzed. Second, citations are treated as if they arise independently. We present a methodology for studying citations between Supreme Court opinions at the dyadic level, as a network, that overcomes these limitations. This methodology -- the citation exponential random graph model, for which we provide user-friendly software -- enables researchers to account for the effects of case characteristics and complex forms of network dependence in citation formation. We then analyze a network that includes all Supreme Court cases decided between 1950 and 2015. We find evidence for dependence processes, including reciprocity, transitivity, and popularity. The dependence effects are as substantively and statistically significant as the effects of exogenous covariates, indicating that models of Supreme Court citation should incorporate both the effects of case characteristics and the structure of past citations., Comment: 34 pages, 10 figures

Information, ideas, and diseases, or more generally, contagions, spread over space and time through individual transmissions via social networks, as well as through external sources. A detailed picture of any diffusion process can be achieved only when both a good network structure and individual diffusion pathways are obtained. The advent of rich social, media and locational data allows us to study and model this diffusion process in more detail than previously possible. Nevertheless, how information, ideas or diseases are propagated through the network as an overall process is difficult to trace. This propagation is continuous over space and time, where individual transmissions occur at different rates via complex, latent connections. To tackle this challenge, a probabilistic spatiotemporal algorithm for network diffusion (STAND) is developed based on the survival model in this research. Both time and spatial distance are used as explanatory variables to simulate the diffusion process over two different network structures. The aim is to provide a more detailed measure of how different contagions are transmitted through various networks where nodes are geographic places at a large scale.

Decades of research has found that democratic dyads rarely exhibit violent tendencies, making the democratic peace arguably the principal finding of Peace Science. However, the democratic peace rests upon a dyadic understanding of conflict. Conflict rarely reflects a purely dyadic phenomena---even if a conflict is not multi-party, multiple states may be engaged in distinct disputes with the same enemy. We postulate a network theory of conflict that treats the democratic peace as a function of the competing interests of mixed-regime dyads and the strategic inefficiencies of fighting with enemies' enemies. Specifically, we find that a state's decision to engage in conflict with a target state is conditioned by the other states in which the target state is in conflict. When accounting for this network effect, we are unable to find support for the democratic peace. This suggests that the major finding of three decades worth of conflict research is spurious., Comment: 38 Pages, 5 Figures

We introduce the hyperedge event model (HEM)---a generative model for events that can be represented as directed edges with one sender and one or more receivers or one receiver and one or more senders. We integrate a dynamic version of the exponential random graph model (ERGM) of edge structure with a survival model for event timing to jointly understand who interacts with whom, and when. The HEM offers three innovations with respect to the literature---first, it extends a growing class of dynamic network models to model hyperedges. The current state-of-the-art approach to dealing with hyperedges is to inappropriately break them into separate edges/events. Second, our model involves a novel receiver selection distribution that is based on established edge formation models, but assures non-empty receiver lists. Third, the HEM integrates separate, but interacting, equations governing edge formation and event timing. We use the HEM to analyze emails sent among department managers in Montgomery County government in North Carolina. Our application demonstrates that the model is effective at predicting and explaining time-stamped network data involving edges with multiple receivers. We present an out-of-sample prediction experiment to illustrate how researchers can select between different specifications of the model.

We study models of weighted exponential random graphs in the large network limit. These models have recently been proposed to model weighted network data arising from a host of applications including socio-econometric data such as migration flows and neuroscience. Analogous to fundamental results derived for standard (unweighted) exponential random graph models in the work of Chatterjee and Diaconis, we derive limiting results for the structure of these models as the number of nodes goes to infinity. Our results are applicable for a wide variety of base measures including measures with unbounded support. We also derive sufficient conditions for continuity of functionals in the specification of the model including conditions on nodal covariates. Finally we include a number of open problems to spur further understanding of this model especially in the context of applications., Comment: 27 pages

With the growth of interest in network data across fields, the Exponential Random Graph Model (ERGM) has emerged as the leading approach to the statistical analysis of network data. ERGM parameter estimation requires the approximation of an intractable normalizing constant. Simulation methods represent the state-of-the-art approach to approximating the normalizing constant, leading to estimation by Monte Carlo maximum likelihood (MCMLE). MCMLE is accurate when a large sample of networks is used to approximate the normalizing constant. However, MCMLE is computationally expensive, and may be prohibitively so if the size of the network is on the order of 1,000 nodes (i.e., one million potential ties) or greater. When the network is large, one option is maximum pseudolikelihood estimation (MPLE). The standard MPLE is simple and fast, but generally underestimates standard errors. We show that a resampling method---the parametric bootstrap---results in accurate coverage probabilities for confidence intervals. We find that bootstrapped MPLE can be run in 1/5th the time of MCMLE. We study the relative performance of MCMLE and MPLE with simulation studies, and illustrate the two different approaches by applying them to a network of bills introduced in the United State Senate.

Researchers interested in statistically modeling network data have a well-established and quickly growing set of approaches from which to choose. Several of these methods have been regularly applied in research on political networks, while others have yet to permeate the field. Here, we review the most prominent methods of inferential network analysis---both for cross-sectionally and longitudinally observed networks including (temporal) exponential random graph models, latent space models, the quadratic assignment procedure, and stochastic actor oriented models. For each method, we summarize its analytic form, identify prominent published applications in political science and discuss computational considerations. We conclude with a set of guidelines for selecting a method for a given application.

The large majority of inferences drawn in empirical political research follow from model-based associations (e.g. regression). Here, we articulate the benefits of predictive modeling as a complement to this approach. Predictive models aim to specify a probabilistic model that provides a good fit to testing data that were not used to estimate the model's parameters. Our goals are threefold. First, we review the central benefits of this under-utilized approach from a perspective uncommon in the existing literature: we focus on how predictive modeling can be used to complement and augment standard associational analyses. Second, we advance the state of the literature by laying out a simple set of benchmark predictive criteria. Third, we illustrate our approach through a detailed application to the prediction of interstate conflict.

How should a network experiment be designed to achieve high statistical power? Ex- perimental treatments on networks may spread. Randomizing assignment of treatment to nodes enhances learning about the counterfactual causal effects of a social network experiment and also requires new methodology (ex. Aronow and Samii 2017a; Bow- ers et al. 2013; Toulis and Kao 2013). In this paper we show that the way in which a treatment propagates across a social network affects the statistical power of an ex- perimental design. As such, prior information regarding treatment propagation should be incorporated into the experimental design. Our findings justify reconsideration of standard practice in circumstances where units are presumed to be independent even in simple experiments: information about treatment effects is not maximized when we assign half the units to treatment and half to control. We also present an exam- ple in which statistical power depends on the extent to which the network degree of nodes is correlated with treatment assignment probability. We recommend that re- searchers think carefully about the underlying treatment propagation model motivat- ing their study in designing an experiment on a network.

Previous research indicates that race, ethnicity, and gender influence legislative behavior in important ways. The bulk of this research, however, focuses on the way these characteristics shape an individual legislator's behavior, making it less clear how they account for relationships between legislators. We study the cosponsorship process in order to understand the race and gender based dynamics underlying the relational component of representation. Using a temporal exponential random graph model, we examine the U.S. House cosponsorship network from 1981 through 2004. We find that Black and Latino members of Congress are at a comparative disadvantage as a result of race-based assortative mixing in the cosponsorship process, yet this disadvantage is mitigated by the electoral pressures that all members face. Members representing districts with significant racial and ethnic minority populations are more likely to support their minority colleagues. We also find that women members do not appear to face a similar disadvantage as a result of their minority status. We argue that these race and gender dynamics in the cosponsorship network are the result of both the inherent tendency towards intra-group homophily in social networks and the electoral connection, which is manifested here as members supporting minority colleagues to broaden their own electoral base of support among minority constituencies.

Dyadic research designs concern data that comprises interactions among actors. Dyadic approaches unambiguously constitute the most frequent designs employed in the empirical study of international politics, but what do such designs cary with them in terms of theoretical claims and statistical problems? These two issues are closely intertwined. When testing hypotheses empirically, the statistical model must be a careful operationalization of the theory being tested. Given that the theoretical and statistical cannot be separated, we discuss dyadic research designs from these two perspectives; highlighting model misspecification, erroneous assumptions about independence of events, artificial levels of analysis, and the incoherent treatment of multilateral/multiparty events on the theoretical side and difficult-to-escape challenges to valid inference on the statistical side.

Scholars, policymakers, and research sponsors have long sought to understand the conditions under which scientific research is used in the policymaking process. Recent research has identified a resource that can be used to trace the use of science across time and many policy domains. US federal agencies are mandated by executive order to justify all economically significant regulations by regulatory impact analyses (RIAs), in which they present evidence of the scientific underpinnings and consequences of the proposed rule. To gain new insight into when and how regulators invoke science in their policy justifications, we ask: does the political attention and controversy surrounding a regulation affect the extent to which science is utilized in RIAs? We examine scientific citation activity in all 101 economically significant RIAs from 2008-2012 and evaluate the effects of attention -- from the public, policy elites and the media -- on the degree of science use in RIAs. Our main finding is that regulators draw more heavily on scientific research when justifying rules subject to a high degree of attention from outside actors. These findings suggest that scientific research plays an important role in the justification of regulations, especially those that are highly salient to the public and other policy actors.

In most domains of network analysis researchers consider networks that arise in nature with weighted edges. Such networks are routinely dichotomized in the interest of using available methods for statistical inference with networks. The generalized exponential random graph model (GERGM) is a recently proposed method used to simulate and model the edges of a weighted graph. The GERGM specifies a joint distribution for an exponential family of graphs with continuous-valued edge weights. However, current estimation algorithms for the GERGM only allow inference on a restricted family of model specifications. To address this issue, we develop a Metropolis--Hastings method that can be used to estimate any GERGM specification, thereby significantly extending the family of weighted graphs that can be modeled with the GERGM. We show that new flexible model specifications are capable of avoiding likelihood degeneracy and efficiently capturing network structure in applications where such models were not previously available. We demonstrate the utility of this new class of GERGMs through application to two real network data sets, and we further assess the effectiveness of our proposed methodology by simulating non-degenerate model specifications from the well-studied two-stars model. A working R version of the GERGM code is available in the supplement and will be incorporated in the gergm CRAN package., Comment: 33 pages, 6 figures. To appear in Social Networks

Across the sciences, the statistical analysis of networks is central to the production of knowledge on relational phenomena. Because of their ability to model the structural generation of networks, exponential random graph models are a ubiquitous means of analysis. However, they are limited by an inability to model networks with valued edges. We solve this problem by introducing a class of generalized exponential random graph models capable of modeling networks whose edges are valued, thus greatly expanding the scope of networks applied researchers can subject to statistical analysis.