Although communication with audiences abroad is important for movements based in regions with complicated geopolitical contexts, it has received limited research attention. To address this gap, we collected and analyzed English-language tweets posted during the Russia-Ukraine conflict of 2014, focusing on understanding the content and practices associated with external-facing communication on social media. Based on the political alignment of tweets and users, determined by hashtag projection, we divided the data into discourses associated with the opposing pro-Ukrainian and pro-separatist movements, along with a neutral discourse. Through topic modeling and qualitative coding, we identified 21 diverse themes across the three discourses, suggesting that users could benefit from more targeted ways of engaging with social media content that enable focused participation. We observed that the pro-Ukrainian movement struggled to sustain long-term participation and compete with inauthentic content. Our findings could be useful for investigating external-facing communication of other online social movements, which have become increasingly prevalent.

Abstract Charts used for persuasion can easily veer into being outright misleading when, for instance, cherry‐picked data is paired with a deceptive caption, as is commonly encountered on social media. The rise of interactive time‐series data explorers for hotly debated topics makes such framing easy to produce and spread. Post‐hoc interventions like fact‐checking often arrive too late and suffer from persistence of belief. Prior work suggests that guardrails, in the form of contextual comparison lines embedded directly into charts, can reduce these effects. We propose and evaluate a practical set of guardrail sampling strategies for implementing such contextual lines in real systems. In a preregistered mixed‐design study with two real‐world scenarios (COVID‐19 and Stocks), participants viewed persuasive charts with different sets of guardrails and reported trust, estimated rank in the dataset, expressed their perceived completeness of context, as well as subjective preference for different tasks. Across scenarios, guardrails improved trust, accuracy of performance judgments, and perceived completeness of context compared to the control. Taken together, the study offers practical guardrail sampling methods, evidence of their contextual benefits, and insights into participants' preferences.

State sponsored information operations, or SSIOs, are a growing problem across many of the information spaces we inhabit online. These instances of coordinated misinformation and propaganda have been perpetrated by over 80 state actors in the last decade, and have been used to exert influence on digital media consumption habits, discussions of contentious issues, and even national elections. Concern over the power that SSIOs wield is only growing as the proliferation of automated tools and services is making it easier than ever to launch large-scale manipulation campaigns. But what role do such automated agents play within the broader operations that they are deployed in? Are they even successful at making an impact in information spaces online? In this work, we address both of these questions through the use of a sequence-based clustering method and advanced linear modeling. Using these methods, we investigate the relationship between agent automation, role, and network characteristics and how much success those agent's achieve over the course of their lifetimes. We find that automated agents perform worse across every success metric compared to human agents, and that they play a smaller, supporting role to the primarily human SSIO workforce. What's more, we find that the extent to which agent's engage in amplifying- or producing-centric roles is by far the biggest determinant of how successful they will be, highlighting the importance of social-roles in the analysis of automated agents.

Detecting context shifts in human experience is critical for applications in cognitive modeling, human-AI interaction, and adaptive neurotechnology. However, formalizing and identifying these shifts in real-world settings remains challenging due to annotation inconsistencies, data sparsity, and the multimodal nature of human perception.

In January 2014, what started as peaceful protests in Kyiv, Ukraine escalated into an ousting of the Ukrainian president, then annexation of Crimea, and a protracted war between the new Ukrainian government and Russian-backed separatists. Meanwhile, Ukrainian citizens documented their experiences on Twitter, leading to the creation of multitudes of conversation spaces, or online publics, that evolved alongside the conflict they were discussing. We collected tweets in those conversations from December 2013 to August 2015. In this work, we use a novel, LDA-based topic tracking tool to analyze how the conversations evolved during the crisis in Ukraine, and the factors associated with this volatility. As existing tools often do not support researchers in pinpointing when online publics shift from one conversation to another, we discuss how our tool can be used to extract cleaner data from longitudinal crises. Finally, we discuss how our discursive analysis of Ukraine in 2014 provides insight on the current conflict in Ukraine.

Official crisis communication on social media is critical to crisis response, as it provides the public with accurate and timely information regarding risks and advised protective actions during crises. A polarized environment can complicate effective crisis communication, as different officials may have different priorities in their communication, making it more challenging for the public to trust information pertaining to risks and protective actions on social media. In this work, we analyze the COVID-19 crisis communication leading up to the 2020 U.S. presidential / congressional election. We focus on what is prioritized in crisis communication by politicians and officials of government agencies - official sources most likely impacted by the polarized environment of contentious national elections. We find their consistency and discrepancy in the communication of primary and secondary risks, as well as corresponding protective actions. In addition, on most topics pertaining to risks and protective actions, the communication by officials of government agencies is predictive of that by politicians. This complicates prior findings that politicians can affect the implementation efficacy of bureaucracy, offering some hope for impartial bureaucracy even in a polarized environment.

Individuals who have sustained a Spinal Cord Injury (SCI) undergo abrupt changes in their functional abilities, impacting all aspects of their lives and imposing a life-long reliance on assistive tools and support from others. This paper aims to understand individuals’ support-seeking behavior in social media as they adjust to their “new normal”—life with reduced mobility and sensation. To understand their online support-seeking behavior, we conducted content analysis on 960 post-threads from SCI-specific subreddit groups. We found that individuals seek informational and emotional support regardless of injury level and time elapsed since injury. Additionally, individuals seek and receive online informational support concerning assistive logistics, motor-functionality, newly acquired self-care, and daily living activities. Similarly, individuals seek emotional support for motivation, and creating new self-identity. Finally, we discuss how social media support dynamics might facilitate reconstructing self-identity, adopting assistive technology, and improving relationships to help adjust to the “new normal.”

Attempting to make sense of a phenomenon or crisis, social media users often share data visualizations and interpretations that can be erroneous or misleading. Prior work has studied how data visualizations can mislead, but do misleading visualizations reach a broad social media audience? And if so, do users amplify or challenge misleading interpretations? To answer these questions, we conducted a mixed-methods analysis of the public’s engagement with data visualization posts about COVID-19 on Twitter. Compared to posts with accurate visual insights, our results show that posts with misleading visualizations garner more replies in which the audiences point out nuanced fallacies and caveats in data interpretations. Based on the results of our thematic analysis of engagement, we identify and discuss important opportunities and limitations to effectively leveraging crowdsourced assessments to address data-driven misinformation.

During crisis situations, observations of other people’s behaviors often play an essential role in a person’s decision-making. For example, a person might evacuate before a hurricane only if everyone else in the neighborhood does so. Conversely, a person might stay if no one else is leaving. Such observations are called social cues. Social cues are important for understanding people’s response to crises, so recognizing them can help inform the decisions of government officials and emergency responders. In this paper, we propose the first NLP task to categorize social cues in social media posts during crisis situations. We introduce a manually annotated dataset of 6,000 tweets, labeled with respect to eight social cue categories. We also present experimental results of several classification models, which show that some types of social cues can be recognized reasonably well, but overall this task is challenging for NLP systems. We further present error analyses to identify specific types of mistakes and promising directions for future research on this task.

Data visualizations can empower an audience to make informed decisions. At the same time, deceptive representations of data can lead to inaccurate interpretations while still providing an illusion of data-driven insights. Existing research on misleading visualizations primarily focuses on examples of charts and techniques previously reported to be deceptive. These approaches do not necessarily describe how charts mislead the general population in practice. We instead present an analysis of data visualizations found in a real-world discourse of a significant global event—Twitter posts with visualizations related to the COVID-19 pandemic. Our work shows that, contrary to conventional wisdom, violations of visualization design guidelines are not the dominant way people mislead with charts. Specifically, they do not disproportionately lead to reasoning errors in posters’ arguments. Through a series of examples, we present common reasoning errors and discuss how even faithfully plotted data visualizations can be used to support misinformation.

Unlike most other forms of coordinated, inauthentic behavior occurring online, the goals of state-sponsored information operations, or SSIOs, are often complex and multifaceted. These goals range from flooding conversations with a certain narrative, to increasing the public's engagement with news sources of questionable quality, to stoking tensions between ideologically opposed groups to weaken public trust. The prevailing theoretical framework for understanding SSIOs is to treat them as a social botnet: a behaviorally homogeneous cluster of coordinated activity. However, the social bot framework is both at odds with some of the behaviors observed in early SSIOs and more broadly with the wide swathe of goals these operations set out to accomplish. To examine the fit of the social bot framework in the SSIO context, we develop a novel bag-of-words based method for clustering and describing user activity traces. Applying this method to a comprehensive repository of SSIOs conducted on Twitter over the last decade, we find that SSIOs violate both the core assumption of the social bot framework, and how it is operationalized in practical work. Instead, we find that SSIOs exhibit a clear division of labor and propose cooperative work with social roles as a more effective theoretical framework for understanding SSIOs. Through applying this framework, we find that the roles that SSIO agents take on have become more stable and simple over time, which holds substantial implications for developing methods for detection of these operations in the wild.

As social media has become more present in people's day-to-day lives, many turn to these platforms in natural disasters to keep abreast of the ever-evolving crisis situation. Facing the increasing amount of crisis-related information available on the social media platforms, users tend to focus on and reach out to authoritative sources---individuals or organizations that provide authoritative and credible crisis-related information due to their official status or position. As they provide high-quality information, response from the authoritative sources can be especially valuable to social media users directly affected in natural disasters. In this study, we aim to extend the reach of credible information during crisis, and direct the attention of authoritative users to the affected users who need their help. Specifically, we investigate what factors differentiate the tweets by regular users that receive responses from authoritative accounts from those that do not. We find that regular users' popularity and official accounts' level of busyness do not seem to affect the likelihood of tweets receiving a response. We thus explore the linguistic features of the tweets' content. Topic modeling and sentiment analysis results suggest that these linguistic aspects of the tweets may affect the response rate from authoritative sources. Our findings suggest crisis-related policy implications, as well as design implications for social media platforms where such exchanges take place, which can potentially increase the reach of credible information in a crisis and help those affected obtain the safety-critical information they need.

The National Science Foundation's Large Facilities are major, multi-user research facilities that operate and manage sophisticated and diverse research instruments and platforms (e.g., large telescopes, interferometers, distributed sensor arrays) that serve a variety of scientific disciplines, from astronomy and physics to geology and biology and beyond. Large Facilities are increasingly dependent on advanced cyberinfrastructure (i.e., computing, data, and software systems; networking; and associated human capital) to enable the broad delivery and analysis of facility-generated data. These cyberinfrastructure tools enable scientists and the public to gain new insights into fundamental questions about the structure and history of the universe, the world we live in today, and how our environment may change in the coming decades. This paper describes a pilot project that aims to develop a model for a Cyberinfrastructure Center of Excellence (CI CoE) that facilitates community building and knowledge sharing and that disseminates and applies best practices and innovative solutions for facility CI.

This work propels social media research beyond the single post as the unit of analysis toward fuller treatment of interaction by making the construct of the conversation analytically available. We offer a method for constructing @reply conversations in Twitter to apprehend social media conversational features at scale. We apply this method to the high-tempo, high-stakes environment of 2012's Hurricane Sandy, with its high volume of online talk by affected locals and distinct disaster-stage phasing by which to consider interactional difference. We investigate the temporality of conversations; the relationality of who speaks to whom; the number and kind of conversationalists; and how content affects temporal features. The analysis reveals that, during the height of the emergency, people expand conversations both in number and kind of conversational partners-just as their information search intensifies. This expansion contributes to longer, slower-paced conversations in the high-emergency period, suggesting reliance on online relationships during times of greatest uncertainty.

OpenStreetMap (OSM) is the most widely used volunteer geographic information system. Although it is increasingly relied upon during humanitarian response as the most up-to-date, accurate, or accessible map of affected areas, the behavior of the mappers who contribute to it is not well understood. In this paper, we explore the work practices and interactions of volunteer mappers operating in the high-tempo, high-volume context of disasters. To do this, we built upon and expanded prior network analysis techniques to select high-value portions of the vast OSM data for further qualitative analysis. We then performed detailed content analysis of the identified activity and, where possible, conducted interviews with the participants. This research allowed the identification of seven distinct mapping practices that can be classified according to dimensions of time, space, and interpersonal interaction. Our work represents a baseline for future research about how OSM crisis mapping practices have evolved over time.

Natural disasters are associated with breakdown of existing structures, but they also result in creation of new social ties in the process of self-organization and problem solving by those affected. In highly-distributed setting of social media, collaborative arrangements must depend on the aspects of work that facilitate (or not) the creation of a shared information space-such as an explicit shared site of work and visible, legible record of the activity. In my dissertation I investigate what organizational structures emerge through problem solving in the context of more or less explicit shared site of work and more or less visible record of activity.

Hurricane Sandy wrought $6 billion in damage, took 162 lives, and displaced 776,000 people after hitting the US Eastern seaboard on October 29, 2012. Because of its massive impact, the hurricane also spurred a flurry of social media activity, both by the population immediately affected and by the globally convergent crowd. In this paper we explore how retweeting activity by the geographically vulnerable differs (if at all) from that of the general Twitter population. We investigate whether they spread information differently, including what and whose content they chose to propagate. We investigate whether the Twitter-based relationships are preexisting or if they are newly formed because of the disaster, and if so if they persist. We find that the people in the path of the disaster favor in their retweeting locally-created tweets and those with locally-actionable information. They also form denser networks of information propagation during disaster than before or after.