No description available.

No description available.

No description available.

Recent election campaigns generated extensive attention for their creative use of Internet, from President Obama’s 2008 Facebook tactic of allowing Facebook friends to share their support with each other to tools that allow supporters to mobilize and influence offline. This thesis asks whether Internet access alone can influence an individual to vote, within the context of American Presidential elections. First, I replicate similar literature by conducting a series of Linear Probability Models that indicate that Internet does have a significant impact on an individual’s decision to vote. However, one major issue that previous studies fail to address is the likelihood of endogeneity between self-reported Internet access and voting behavior. To address this, I introduce a measure of Internet Service Providers available in a given Congressional District as an instrumental variable. Once instrumented, it appears that Internet is largely insignificant. However, there is a key exception in 2008, where Internet access is significant. I argue that this is due to the developments in social media technology that revolutionized the ways in which candidates engaged with voters and voters engaged with one another.

More and more, elections around the world seem to be won or lost on the basis of the candidates’ personal qualities rather than their policies. Despite its prevalence and consequences, we still know very little about what explains such candidate-centered voting, particularly in new democratic contexts. I argue that variation in candidate-centered voting is largely a function of political sophistication: voters with higher levels of political sophistication are better able to process information relating to policy and performance, which tends to be more cognitively demanding than information relating to candidate’s personalities. To test this argument, I estimate models of vote choice and electoral utility using survey data from the 2002 presidential election in Brazil. The results largely support my contention that political sophistication conditions the weight of candidate considerations relative to policy and performance considerations. / text

<p>Unprecedented amount of data has been collected in diverse fields such as social network, infectious disease and political science in this information explosive era. The high dimensional, complex and heterogeneous data imposes tremendous challenges on traditional statistical models. Bayesian nonparametric methods address these challenges by providing models that can fit the data with growing complexity. In this thesis, we design novel Bayesian nonparametric models on dataset from three different fields, hyperspectral images analysis, infectious disease and voting behaviors. </p><p>First, we consider analysis of noisy and incomplete hyperspectral imagery, with the objective of removing the noise and inferring the missing data. The noise statistics may be wavelength-dependent, and the fraction of data missing (at random) may be substantial, including potentially entire bands, offering the potential to significantly reduce the quantity of data that need be measured. We achieve this objective by employing Bayesian dictionary learning model, considering two distinct means of imposing sparse dictionary usage and drawing the dictionary elements from a Gaussian process prior, imposing structure on the wavelength dependence of the dictionary elements.</p><p>Second, a Bayesian statistical model is developed for analysis of the time-evolving properties of infectious disease, with a particular focus on viruses. The model employs a latent semi-Markovian state process, and the state-transition statistics are driven by three terms: ($i$) a general time-evolving trend of the overall population, ($ii$) a semi-periodic term that accounts for effects caused by the days of the week, and ($iii$) a regression term that relates the probability of infection to covariates (here, specifically, to the Google Flu Trends data).</p><p>Third, extensive information on 3 million randomly sampled United States citizens is used to construct a statistical model of constituent preferences for each U.S. congressional district. This model is linked to the legislative voting record of the legislator from each district, yielding an integrated model for constituency data, legislative roll-call votes, and the text of the legislation. The model is used to examine the extent to which legislators' voting records are aligned with constituent preferences, and the implications of that alignment (or lack thereof) on subsequent election outcomes. The analysis is based on a Bayesian nonparametric formalism, with fast inference via a stochastic variational Bayesian analysis.</p> / Dissertation

This study has two purposes: first, to test the validity of the personal attribute model in explaining judicial voting behavior outside its original cultural context; second, to explain the variation in justice's voting behavior in the Canadian Supreme Court. For the most part, the result arrived in this study supports the validity of the model in cross-cultural analysis. The result of multiple regression analysis shows that four variables, region, judicial experience prior to appointment, political party of appointing Prime Minister, and tenure account for 60 percent of the variations in justice's voting behavior. This result, hence, provides an empirical finding to the development of the personal attribute model in explaining justices' voting behavior.

The purpose of this study is to examine two major psychological determinants of the vote in presidential elections - candidate image and party orientation. The central thesis of this study is that candidate image, as measured here, has been a greater determinant of electoral choice in the majority of presidential elections since 1952 than has party orientation. One of the vices as well as virtues of a democratic society is that the people often get what they want. This is especially true in the case of electing our leaders. Political scientists have often concentrated their efforts on attempting to ascertain why people vote as they do. Studies have been conducted focusing on the behavior of voters in making that important decision-who should govern?

A dissertação de mestrado analisa os artigos publicados a respeito do Comportamento Eleitoral no Brasil em periódicos científicos com impacto nacional. Para tal, discute os primeiros trabalhos publicados, datados do fim dos anos 50, até os dias atuais, em revistas nacionais de Qualis A1 e A2. O trabalho discute a evolução das teorias de explicação do voto na ciência política brasileira e tem como argumento principal que os estudos sobre o comportamento eleitoral foram cruciais para a construção da autonomia acadêmica da disciplina e a introdução e difusão do paradigma comportamentalista, ajudando a construir uma Ciência Política no Brasil que se preocupa com a análise empírica e a quantificação. Entretanto, as revistas que servem de plataforma para as publicações, os pesquisadores e seus vínculos institucionais mostram que há uma concentração dessa linha de pesquisa e da produção em instituições de pesquisa da região Sudeste. Também há uma concentração moderada no tipo de abordagem utilizada, pois predomina a explicação psicológica e, em segundo lugar, estudos de geografia eleitoral. Ao analisarmos o desenvolvimento teórico dos estudos publicados no formato artigo em revista, defendemos que a próxima etapa da construção da autonomia acadêmica da disciplina, pelo menos no que se refere à subárea de estudos de comportamento eleitoral, é a descentralização dos meios de produção científica – revistas, editoras, etc. – e a própria produção científica, assim como o aumento do pluralismo teórico e metodológico no estudo desse tema. / The master's thesis analyzes the articles published about the Voting Behavior in Brazil in scientific journals with national impact. As such, it discusses the first published works, dating from the late 50s until the present day, in national journals Qualis A1 and A2. The paper discusses the evolution of theories of vote explanation in Brazilian political science and its main argument is that the studies on voting behavior were crucial to the construction of the academic autonomy of discipline and the introduction and dissemination of the behaviorist paradigm, helping to build a political science in Brazil that cares about the empirical analysis and quantification. However, the magazines that serve as a platform for publications, researchers and their institutional links show that there is a concentration of such research and the production line at research institutions in the Southeast. There is also a moderate concentration in the type of approach used, as predominates the psychological explanation and, secondly, electoral geography studies. By analyzing the theoretical development of the studies published in article magazine format, it is argued that the next stage of construction of the academic autonomy of the subject, at least in regard to the subfield of electoral behavior studies is the decentralization of scientific production means - magazines, publishing houses, etc. - And the scientific production itself, as well as the increase of the theoretical and methodological pluralism in the study of this subject.

ABSTRACT The definition of economic voting by Lewis-Beck indicates constituency can cast a ballot for the benefit of the ruling party while they definitely confirm economic performance by past; vice versa. The study also expresses Western democratic countries periodically hold the Election of Central Administration Leading Cadre and the Legislative Election; furthermore, it shows in response to economic voting during constituency go into deciding by ballot at the meantime. Seeing that documents of the research in Taiwan of voting behavior among constituency much focus on national identification, political party and candidate characteristics; the other way round, less focus on economic performance or economic governance ability. It is unworthy truth for Taiwan Economy to be no longer wonderingly splendid manifestation. Therefore, this study longs for by way of observing materials and data in poll centers behind the President Election in 1996, 2000 as well as the Legislative Election in 1998, 2001. Consequently, it searches out whether economic voting manifestation for Taiwan constituency; it awaits to comprehend if the Economy is one of main variable while voting by constituency in Taiwan? Moreover, it attempts to observing constituency to be part of which social background characteristics. On the basis of the above-mentioned analysis, this study comes at something: I. Economic evaluation is indeed an influential factor; due to unique historical culture background, the crucial point in voting still inferiors to identification of political party and unity-independence position. Nevertheless, candidates among latterly national elections propose unexceptionable policy to lure constituency; newspapers and mass media describe prosperity fluctuation by a wide margin. The effect of economic issues is more and more significant in future election. It may observe continually. II. The study detects low-education-level constituency easily possess economic voting behavior than high-education-level constituency, but it is not fit in with foreign relevant economic voting theory - it is worthy of probing into variations between Taiwan and foreign countries. Over and above, owning-occupation constituency easily possesses economic voting behavior than non-owning-occupation constituency. It hopes adding the gender and age factors in future research. Key Words: Economic Evaluation, Voting Behavior, the Legislative Election, the President Election, Multinomial Logit Model