In this dissertation, I investigate the socioeconomic causes of consequences of ethnic conflict, and evaluate interventions that can reduce social animosity and misperceptions about outgroups. In particular, I focus on conflict between Hindus and Muslims in India. In recent years, online misinformation has emerged as a major contributor to misperceptions and animosity towards Muslims in India. In Chapter 1, I investigate if we can inoculate people against misinformation and mitigate its impact on people’s beliefs, attitudes, and behavior? We conduct a large field experiment in India with an intervention providing weekly digests containing a compilation of fact-checks of viral misinformation. In these digests, we also incorporate narrative explainers to give details and context of issues that are politically salient and consistent target of false stories. Specifically, we address misperceptions about Muslims increasingly fuelled by online misinformation. We find that familiarity with fact-checks increases people’s ability to correctly identify misinformation by eleven percentage points. However, belief in true news also decreases by four percentage points. We estimate a structural model to disentangle the two mechanisms of impact—truth discernment, which is the ability to correctly distinguish between false and true news; and skepticism, which changes the overall credulity for both false and true news. The impact is driven by an increase in both truth discernment and skepticism. Whereas skepticism increases immediately, it takes several weeks to become better at discerning truth. Finally, our intervention reduces misperceptions about Muslims, as well as leads to changes in policy attitudes and behavior. Treated individuals are less likely to support discriminatory policies and are more likely to pay for efforts to counter the harassment of inter-faith couples. In Chapter 2, I investigate the economic impacts of conflict and social animus by estimating the causal impact of ethnic violence on economic growth in India. For causal identification, I use shift-share instruments to isolate exogenous national shocks to violence from endogenous local shocks. On average, a riot reduces state GDP growth rate by 0.14 percentage points. To investigate mechanism, I estimate the dynamics of impact using the synthetic control method and compare it to theoretical predictions from a shock to social capital versus physical capital. This shows that the negative impact of violence is likely driven by a negative shock to social capital from higher animosity and discrimination among communities exposed to violence. This impact of violence on growth creates a vicious cycle when one also considers the effect in the opposite direction – lower growth leading to more violence. The multiplier due to this vicious cycle magnifies the impact of external growth shocks by 40 percent in equilibrium. Overall, the results highlight the importance of strong institutions to manage conflict for the long-term prosperity of societies. In Chapter 3, I investigate the historical origins of ethnic violence in India by comparing violence in regions that were directly ruled by British, versus those that were indirectly ruled through native kings who had significant autonomy. I find that regions that are directly ruled have more violence in post-independence period. I then use direct British rule as an instrument for ethnic violence to estimate the impact of violence and residential segregation.

Social media is replacing many of traditional media’s purposes, information gathering is one of them. In the digital culture today we use our phones, the internet, and social media in our daily lives, especially younger generations, and those generations could be relying their information gathering on social media. Therefore, the purpose of this thesis is to get a better understanding of which role social media plays in the information, news, and political gathering of Generation Z. The aim is also to investigate how Generation Z views potential threats, personalization, and fact-checking on social media. Those questions were answered through a semi-structured interview with 10 interviewees from different parts of Sweden that were within Generation Z. Generation Z has a beginning of birth year starting at 1997 to 2012 and were chosen because that generation is the first generation to be born and grow up with access to portable digital technology and internet. The results through the thematic analysis made six themes which were social media use, experience of misinformation, source dependency, physical vs digital newspaper perceived trustworthiness, fact-checking preferences, and personalization. The interviewees indicated that social media is their main source of information, news, and politics, a few used traditional media now and then but mostly social media platforms. The perceived ability of the interviewees was that they knew the importance of following and gathering information, news, and politics through trustworthy sources on the platforms. If the interviewees were interested in a subject they saw a post about or if it was considered important they would fact-check it, otherwise not. The interviewees knew more or less that they had seen misinformation, they had experienced personalization and had different opinions about it.

Desinformation har genom alla tider varit en del av samhället och har genom historien haft avgörandet betydelse för händelser som utspelat sig. Idag är desinformation ett ytterst aktuellt ämne som genom nyhetsflödets ökande tempo med digitaliseringen, samt AIs förmåga att producera manipulerade bilder är det ett område som är viktigt att studera. Syftet med undersökningen var att undersöka om det fanns någon korrelation mellan den svenska befolkningens känsla av inkludering kopplad till deras attityd mot desinformation. Vidare undersöktes det om det fanns något samband kopplade till intersektionella faktorer som kön och ålder. Undersökningen hade en kvantitativ ansats genom insamlandet av enkäter, som sedan sammanställdes. Resultatet visade att inget samband mellan känslan av inkludering och attityd till desinformation, samt mellan desinformation kopplat till kön eller ålder kunde fastställas. Slutsatsen som kan dras är att bristen på signifikanta resultat kan tillskrivas antingen en missad korrelation som inte fångades i denna studie eller det faktum att desinformation är en så genomgripande samhällsfråga att den ses negativt av de flesta individer, vilket gör det svårt att hitta signifikanta skillnader mellan olika grupper. En separat undersökning fann dock visst samband mellan konspiratoriska övertygelser och individer inom den politiska högern, vilket tyder på ett potentiellt samband mellan politiska ideologier och attityder till desinformation. Ytterligare forskning behövs för att utforska denna aspekt. / Disinformation has always been a part of society and throughout history has been decisive for events that unfolded. Today, disinformation is an extremely current topic that, through the increased pace of the news flow with digitization, and AI's ability to produce manipulated images is the most current area to study. The purpose of the survey was to investigate whether there was any correlation between the Swedish population's sense of inclusion linked to their attitude towards desinformation. Furthermore, the paper explored whether there was any relationship linked to intersectional factors such as gender and age. The survey was a quantitative survey through the collection of questionnaires, which were then compiled. The result showed that no relationship between the feeling of inclusion and attitude to disinformation, as well as between disinformation linked to gender or age could be established. The conclusion that can be drawn is that the lack of significant results can be attributed either to a missed correlation that was not captured in this study or to the fact that disinformation is such a pervasive societal issue that it is viewed negatively by most individuals, making it difficult to find significant differences between different groups. However, a separate survey found some association between conspiratorial beliefs and individuals on the political right, suggesting a potential link between political ideologies and attitudes toward disinformation. Further research is needed to explore this aspect.

Recent years have witnessed a slow but persistent erosion of the democratic governing of the U.S. Political scientists have identified several severe threats against the American democracy, including the spread of misinformation, the impact of negative partisanship, and the lack of political belonging for marginalized groups. While research on these threats abounds at the macro level, what remains under researched are the conduct of ordinary people as they navigate these issues and exercise civic rights.From the micro-analytic perspective of Conversation Analysis and Membership Categorization Analysis, this dissertation examines the practices that ordinary citizens and U.S. Members of Congress (MOCs) use to navigate these issues. 50 recordings of town hall meetings were collected from social media pages of the MOCs’ offices from 24 US States with the majority being House Representatives and several Senators. The dissertation addresses, on the one hand, how citizens navigate the following issues: introducing misinformation as part of one’s town hall contribution, (re)producing negative partisanship, and (re)producing exclusionary stances towards minoritized groups; and on the other, how MOCs respond to these issues. First, citizens assert epistemic superiority or appeal to rationality when bringing in factually unfounded information to town hall contributions. MOCs may choose to endorse, sidestep, or refute the misinformation when responding. Secondly, citizens present irreconcilable alternatives, presuppose a zero-sum game, or ascribe categories of threats to the opposing party. MOCs can choose to neutralize or upgrade the negative partisanship and even initiate it on their own. Finally, citizens assign categories of immorality against minoritized groups or establish their own religious superiority for maintaining exclusionary stances. MOCs may disalign with such stances by offering contesting categories, redirecting the focus of discussion, or disaffiliating with the citizens’ exclusionary stance via a telling. With the perspective of “motivated looking,” this dissertation is a continued effort in critically informed EMCA research and can be used to strengthen research on participatory democracy by its inherently emic approach. More importantly, findings from this dissertation can be adapted by organizations, teachers, and individuals to hold or facilitate more productive conversations around civic topics.

No description available.

No description available.

While conventional wisdom suggests political ads are often misleading, this is the first known study to quantify the prevalence of inaccuracies in political advertising. This study also examines how and explains why the U.S. news media provide coverage of political advertising in the manner that they do. A multi-method research design includes a content analysis of the television ads from the 2008 presidential election, secondary data analysis of the National Annenberg Election Survey 2008, semantic network analysis of press coverage of political television ads from the 2008 election, as well as in-depth interviews with scholars, practitioners, journalists and lawyers having expertise in the issues surrounding political advertising. Of all the English-language paid political ads that aired on television during the 2008 general election, just under 30% contained at least one inaccuracy based upon the ratings of FactCheck.org and PolitiFact.com. This figure, however, is likely a gross under representation of the prevalence of inaccuracies in political ads from 2008 because most of the ads (70%) were never evaluated by these purportedly elite fact-checkers. Among ads assessed, however, more than three out of four of the evaluated claims had some degree of inaccuracy. Furthermore, ads containing at least one inaccuracy aired twice as often on television as the ads that were never evaluated. To the degree inaccurate ads air on television more frequently, then, there is cause for concern particularly given the broadcasters' mandate to serve the public interest. Moreover, while political interest supporters were one of the groups most likely to have inaccuracies in their ads, nearly half of their spending was in the last weeks of the election contributing to most of their ads going without evaluation. Thus, in a post-Citizens United world, attempts by fact-checkers to review the onslaught of PAC ads during the final weeks of the 2012 election (and the final weeks of future elections) will be crucial in combating inaccuracies. This study also extends the work of Geer (2006) who offered an organized review of negativity in political advertising. Rather than finding support for the hypothesis that negative attack ads are more accurate than advocacy ads, the evidence challenges Geer's defense of negativity. Among the ads evaluated by the fact-checkers, inaccuracies were significantly more likely to be present in attack rather than either advocacy or contrast ads. While Geer may have demonstrated that negative ads offer more substantive evidence, simply because evidence is presented does not mean the evidence is accurate. In the more provocative ads of 2008 designed to gain attention, inaccuracies were rife. Moreover, rather than the mainstream news media fixation on political ad negativity, the evidence in the forthcoming pages suggests attention is more warranted concerning the accuracy of the claims within the ads regardless of the ad's tone. A first step toward a theory of strategic misinformation is also offered by demonstrating that it is possible to predict which political ads were more likely to draw an inaccurate rating from the fact-checkers. Holding all other variables constant, it was attack ads that had the highest odds of being evaluated as inaccurate with contrast ads also having a high likelihood. These predictions also confirmed that as the campaign progressed, the odds of an ad being rated inaccurate declined which was a function of ads not being evaluated. Furthermore, it was revealed that a loss of momentum or a decline in public perceptions of candidate characteristics increased the odds of candidates drawing inaccurate ratings in their attack ads. In extending understanding of how news media cover candidate campaigns when political advertising is referenced, a plurality of media outlets from the over two dozen in the study were characterized foremost by their focus on campaign strategy rather than fact-checking. One cluster, however, emerged as AdWatchers - those committed to using political ads to scrutinize the accuracy of what candidates and their surrogates were claiming. Nonetheless, the economic realities of adwatching are that there is a so called "chilling effect" because it is expensive, time-consuming, and divisive. Furthermore, the dearth of watchdog ad reporting enables broadcast stations to continue airing ads that may be false while preserving their ability to claim ignorance about the content when faced with regulatory compliance issues. Thus, the political ads most likely to air are the ones with inaccuracies. Chances are the ads will go unscrutinized by the mainstream news media while television stations profit from their proliferation. / Mass Media and Communication

The changes in the course of communication changed the world from different perspectives. Public participation on social media means the generation, diffusion, and exposure to a tremendous amount of user-generated content without supervision. This four-essay dissertation analyzes information diffusion through social media and its opportunities and challenges through management systems engineering and data analytics. First, we evaluate how information can be shared to reach maximum exposure for the case on online petitions. We use system dynamics modeling and propose policies for campaign managers to schedule the reminders they send to have the highest number of petition signatures. We find that sending reminders is more effective in the case of increasing the signature rate. In the second essay, we investigate how people build trust/ mistrust in science during an emergency. We use data analytics methods on more than 700,000 tweets containing keywords of Hydroxychloroquine and chloroquine, two candidate medicines, to prevent and cure patients infected with COVID-19. We show that people's opinions are concentrated in the case of polarity and spread out in the case of subjectivity. Also, they tend to share subjective tweets than objective ones. In the third essay, building on the same dataset as essay two, we study the changes in science communication during the coronavirus pandemic. We used topic modeling and clustered the tweets into seven different groups. Our analysis suggests that a highly scientific and health-related subject can become political in the case of an emergency. We found that the groups of medical information and research and study have fewer tweets than the political one. Fourth, we investigated fake news diffusion as one of the main challenges of user-generated content. We built a system dynamics model and analyzed the effects of competition and correction in combating fake news. We show that correction of misinformation and competition in fake news needs a high percentage of participation to be effective enough to deal with fake news. / Doctor of Philosophy / The prevalence of social media among people has changed information diffusion in several ways. This change caused the emergence of a variety of opportunities and challenges. We discuss instances of these in this dissertation in four main essays. In the first essay, we study online social and political campaigns. Considering the main goal of campaign managers is to gain the highest reach and signatures, we generate a model to show the effects of sending reminders after the initial announcement and its schedule on the final total number of signatures. We found that the best policy for online petition success is sending reminders when people are increasingly signing it rather than when people lose interest in it. In the second essay, we investigated how people build trust/ mistrust in scientific information in emergency cases. We used public tweets about two candidate medicines to prevent and treat patients infected with COVID-19 and analyzed them. Our results suggest that people trust and retweet the information that is based on emotions and judgments more than the one containing facts. We evaluated the science communication during the mentioned emergency by further investigating the same dataset in the third essay. We clustered all the tweets based on the words they used into seven different groups and labeled each of them. Then, we focused on three groups of medical, research and study, and political. Our analysis suggests that although the subject is a health-related scientific one, the number of tweets in the political group is greater than the other clusters. In the fourth essay, we analyzed the fake news diffusion through social media and the effects of correction and competition on it. In this context, correction means the reaction to misinformation that states its falsity or provides counter facts based on truth. We created a model and simulated it for the competition considering novelty as one influential factor of sharing. The results of this study reveal that active participation in correction and competition is needed to combat fake news effectively.

The dependence on Natural Language Processing (NLP) systems has grown significantly in the last decade. Recent advances in deep learning have enabled language models to generate high-quality text at the same level as human-written text. If this growth continues, it can potentially lead to increased misinformation, which is a significant challenge. Although claim verification techniques exist, they lack proper explainability. Numerical scores such as Attention and Lime and visualization techniques such as saliency heat maps are insufficient because they require specialized knowledge. It is inaccessible and challenging for the nonexpert to understand black-box NLP systems. We propose a novel approach called, ExClaim for explainable claim verification using NLP rationalization. We demonstrate that our approach can predict a verdict for the claim but also justify and rationalize its output as a natural language explanation (NLE). We extensively evaluate the system using statistical and Explainable AI (XAI) metrics to ensure the outcomes are valid, verified, and trustworthy to help reinforce the human-AI trust. We propose a new subfield in XAI called Rational AI (RAI) to improve research progress on rationalization and NLE-based explainability techniques. Ensuring that claim verification systems are assured and explainable is a step towards trustworthy AI systems and ultimately helps mitigate misinformation. / Master of Science / The dependence on Natural Language Processing (NLP) systems has grown significantly in the last decade. Recent advances in deep learning have enabled text generation models to generate high-quality text that is at the same level as human-written text. If this growth continues, it can potentially lead to increased misinformation, which is a major societal challenge. Although claim verification techniques exist, they lack proper explainability. It is difficult for the average user to understand the model's decision-making process. Numerical scores and visualization techniques exist to provide explainability, but they are insufficient because they require specialized domain knowledge. This makes it inaccessible and challenging for the nonexpert to understand black-box NLP systems. We propose a novel approach called, ExClaim for explainable claim verification using NLP rationalization. We demonstrate that our approach can predict a verdict for the claim but also justify and rationalize its output as a natural language explanation (NLE). We extensively evaluate the system using statistical and Explainable AI (XAI) metrics to ensure the outcomes are valid, verified, and trustworthy to help reinforce the human-AI trust. We propose a new subfield in XAI called Rational AI (RAI) to improve research progress on rationalization and NLE-based explainability techniques. Ensuring that claim verification systems are assured and explainable is a step towards trustworthy AI systems and ultimately helps mitigate misinformation.

This study examines how young people, in particular adolescents in the ages 18 to 20 years old, use Tiktok to access news. A qualitative interview survey is conducted and analyzed based on the uses-and-gratifications theory and the perception of News-Finds-Me. The respondents taking part in the interviews explain how they use Tiktok to access news, discuss the apps algorithms and how they function on Tiktok. They also talk about in what ways they are influenced by Tiktok and their view on Tiktok and misinformation. Among the respondents, it becomes clear that young people in the ages 18 to 20 years old use Tiktok as their first hand choice as a news source and that they use it every day - for many hours a day. These young adults still find Tiktok as a necessary social media to use to stay updated with world news and what is going on in society and to stay in tune with popular trends. The findings show that these young adults also use Tiktok to stay in touch with relatives and close and distant friends. It also appears that the young adults see Tiktok as a platform to waste time on when bored and that the negative side of using Tiktok for news is that you get addicted to the app, because of its addictive design and algorithms that suggest what you want to see on the app.