Nuclear energy is a source of carbon free power. With many countries striving to make deep carbon cuts in their energy sectors, nuclear energy could be a large part of the solution. One of the main obstacles standing in the way of the use of nuclear energy is the issue of used nuclear fuel disposal. According to the NEI, the U.S. creates about 2000 metric tons of used nuclear fuel per year and has generated around 76,000 metric tons of used nuclear fuel over the last 4 decades. While there are technical problems that need to be solved, it is primarily the public and political opposition to the disposal of used nuclear fuel that stands in the way of progress in this area. This work addresses this issue through Multi-Criteria Decision Analysis (MCDA). To make a decision among ten different fuel cycles, we have brought together five stakeholders: Nuclear Scientists and Engineers, Environmental Scientists, Economists, Political Scientists, and The General Public. Using Multi-Attribute Value Theory (MAVT), we have been able to develop decision models for each stakeholder as well as a model that combines them all and came to the conclusion that of the ten fuel cycles considered, the best decision is to continue to use On Site Dry Cask Storage. This decision is made with small sample sizes but the methodology could be applied at much larger scales and can potentially be used to choose a fuel cycle that encounters much less political and social opposition to its implementation.

There has been much literature about the economic effects of terrorism in democratic countries, but this literature often considers democracy to be a binary variable. This paper sought to explore how the effects might differ depending on the strength of a democracy. In the end, I found that the numbers of attacks and the effects of those attacks do not follow a linear path. The results for autocracies and anocracies require further analysis, but democracies have revealed interesting results. It seems that democracies as a whole have more terrorist attacks, but, within this group, the more democratic a country is the fewer attacks are carried out.

This dissertation builds on existing scholarship in political science and political sociology to explore the influence of interest groups in legislative action networks. The primary theoretical insight is that as the number of interest group affiliations between two members of Congress increases, so does the frequency with which they forge other sorts of social ties necessary to advance the interests of their interest group constituencies. In particular, the analysis looks at interest group donation strategies, legislative co-sponsorships, and roll-call votes during the 111th Congress (2009-2010). The analysis uses social network analysis methods to create network models of 19 different policy domains, as well as an aggregate model, for both the House and Senate. Legislator ideology, state, committee assignments, and experience have a generally significant impact on the number of interest group affiliations shared by each pair of legislators, whereas gender, race/ethnicity, office location and occupational history do not. The results show that interest groups do have consistent impact over co-sponsorships in the House, but somewhat more mixed influence in the Senate. In some instances, groups in the policy domain encourage policy change, and in other instances, status quo protection. The theory did not anticipate the latter effect, though it does make sense in context of other research findings. For roll-call votes, interest groups have a significant influence over some House policy domains but not many Senate policy domains. The increased polarization of the Senate, necessity of minority party discipline to maximize their leverage through use of the filibuster, and staggered nature of Senate elections makes interest group influence tougher to muster in the upper chamber of Congress.

For almost 40 years, scholars have sought to determine how elections affect the economy. Recently, certain studies have focused on the effect of political uncertainty on the economy. This paper focuses specifically on the effect of political uncertainty on business investment. We use 30 years of data from the U.S. states to show that policy uncertainty leads to significant declines in business fixed investment, sometimes referred to as “irreversible investment.” Moreover, we find that the magnitude of the decline in investment depends on the level of policy uncertainty. These results support predictions for “Electoral Investment Theory” and the existence of reverse political business cycles more generally.

Flows of fluids and solid particles through microchannels have a very wide range of applications in biological and medical science and engineering. Understanding the mechanism of microflows will help to improve the development of the devices and systems in those applications. The aim of this study is to develop a sophisticated simulation and analysis technique for the study of fluid-particle flow through microchannels. This work involves construction of mathematical models, development of analytical methods and numerical algorithms, and numerical investigation and analysis. / The study consists of three parts. The first part of the research focuses on the transient flow of an incompressible Newtonian fluid through a micro-annual with a slip boundary. The flow of the fluid is governed by the continuity equation and the Navier-Stokes equations, and is driven by the pressure field with a timevarying pressure gradient. By using the Fourier series expansion in time and Bessel functions in space, an exact solution is derived for the velocity field. The velocity solution is then used to obtain the exact solutions for the flow rate and the stress field. Based on the exact solutions, the influence of the slip parameter on the flow behaviour is then investigated. / The second part of the research focuses on the particle-fluid flow in microchannels. The transport of fluid in the vessel is governed by the continuity equation and the transient Navier-Stokes equations, while the motion of the particles is governed by Newton’s laws. The particle-wall and particle-particle interactions are modelled by the interacting forces, while the particle-fluid interaction is described by the fluid drag force. A numerical scheme based on the finite element method and the Arbitary Lagrangian-Eulerian method is developed to simulate the motion of the particles and the fluid flow in the vessels. The influence of boundary slip on the velocity field in the fluid is also investigated numerically. / Based on the work in the second part, the third part of the research focuses onthe control of the movement of particles in the fluid by applying an external magneticfield to the system. Maxwell’s equations are used to model the magnetic fieldgenerated by the external magnetic source, and a finite element based numericalscheme is developed to solve the underlying boundary value problem for the magneticflux density generated. From the computed flux density and magnetic vectorpotential, the magnetic forces acting on the particles are determined. These magneticforces together with the drag force and the particle-particle interacting forcesdominate the behaviour of the particle motion. A numerical scheme, similar to thatfor the second part of the research, is then developed to study the fluid-particle flowin microchannels under magnetic forces, followed by a numerical investigation onthe influence of the magnetic forces on the particle flow behaviour.

Over the past few decades, falling voter turnout rates have induced governments to adopt compulsory voting laws, in order to mitigate issues such as the socioeconomic voter gap and to bring a broader spectrum of voters into the fold. This paper presents evidence that the introduction of mandatory voting laws increases voter turnout rates by 13 points within a particular country through an entity- and time-fixed effect panel model. Moreover, it includes a discussion of the implications of adopting mandatory voting policies within the United States, finding that compelling citizens to vote would have increased participation rates to over 90 percent in the past four presidential elections.

This behavioral study examines the users’ engagement on social networking[ sites (SNS) in electoral races for public office in relation to their act of voting. This study was based on the premise that when certain criteria are met then SNS can be used as a predictive tool. The initial technique used was observations of the tools/functions on SNSs such as the “Like” button, favorable comments, retweets, friends/followers. Another technique used was surveys administered to individuals at political rallies, political debates, and college campuses to further analyze if their online engagement in politics translates to their physical participation. A quantitative case study analysis approach was used to analyze data gathered. Results of the data collected suggest that there is a positive relationship between online engagement and physical participation. The conclusion drawn from the findings suggests that SNS can be used as a predictive tool when examining electoral races for public offices.

Submitted by JÚLIO HEBER SILVA (julioheber@yahoo.com.br) on 2016-12-08T17:49:42Z No. of bitstreams: 2 Dissertação - Daniel Ferreira Gonçalves - 2016.pdf: 6039771 bytes, checksum: 9dd76bacc79e08f02f565c3a309abedd (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Jaqueline Silva (jtas29@gmail.com) on 2016-12-13T18:06:03Z (GMT) No. of bitstreams: 2 Dissertação - Daniel Ferreira Gonçalves - 2016.pdf: 6039771 bytes, checksum: 9dd76bacc79e08f02f565c3a309abedd (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2016-12-13T18:06:03Z (GMT). No. of bitstreams: 2 Dissertação - Daniel Ferreira Gonçalves - 2016.pdf: 6039771 bytes, checksum: 9dd76bacc79e08f02f565c3a309abedd (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2016-03-11 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Studies focused on modeling and mechanical system behavior prediction are relevant to the avoidance of excessive deflections and structural failures. Most of these systems are discretized by finite element models which are constituted by a large number of degrees of freedom, making it quite expensive computationally. In this context, the application of models reduction methods becomes particularly prominent because it provides a greater saving in time, keeping the solution quality. This work aims to study methods of reducing models applied to nonlinear dynamical systems. Among the methods aimed at reducing models, there is the Guyan method, Improved Reduction System (IRS), Iterative Improved Reduction System (IIRS) System Equivalent Reduction Expansion Process (SEREP), Component Mode Synthesis (CMS) and method of modal projection. In order to verify the efficiency of the models reduction methods when applied to nonlinear systems is presented in this paper a study where the system responses are analyzed by checking the computational cost and quality of the reduced solution of the problem for each of the mentioned methods. / Estudos voltados à modelagem e previsão do comportamento de sistemas mecânicos são relevantes na coibição de deflexões excessivas e falhas estruturais. A maioria destes sistemas são discretizados por elementos finitos cujos modelos são constituídos por elevado número de graus de liberdade tornando-se bastante oneroso computacionalmente. Neste contexto, a aplicação de métodos de redução de modelos ganha destaque especial pois proporciona maior economia de tempo, mantendo a qualidade de solução. Este trabalho tem por objetivo o estudo de métodos de redução de modelos aplicados a sistemas dinâmicos não lineares. Dentre os métodos voltados à redução de modelos, destaca-se o método de Guyan, Improved Reduction System (IRS), Iterative Improved Reduction System (IIRS), System Equivalent Reduction Expansion Process (SEREP), Component Mode Synthesis (CMS) e método da Base Modal. Com o objetivo de verificar a eficiência dos métodos de redução de modelos quando aplicados em sistemas não lineares, é apresentado neste trabalho um estudo onde são analisadas as respostas do sistema, verificando o custo computacional e a qualidade da solução do problema reduzido para cada um dos métodos citados.

Over forty years ago, Hanna Pitkin expressed concern that social scientists were failing to give concepts the attention which they needed (Pitkin 1972, 277). This thesis takes up the same theme, asking how the concept of responsiveness is treated by political scientists. The goal to reveal confusion that surrounds widely used concepts such as responsiveness. The analysis offered in this thesis has significance for the discipline of political science in three ways. First, it highlights confusion surrounding the concept of responsiveness itself. Responsiveness is a widely utilized concept employed throughout the social sciences; however, as this thesis shows, there is wide disagreement in how the concept is understood. This confusion is fueled by the frequent failure of scholars to critically analyze the concept and the assumptions which have been attached to the understanding of responsiveness. Second, by analyzing the related concepts of representation and democracy, this thesis suggests that there is a lack of attention to concepts which are employed for research that extends beyond responsiveness. Many other concepts that are foundational to our discipline warrant increased scrutiny. Third and finally, the thesis highlights the danger of ignoring the ideological commitments of political scientists, commitments that can shape in hidden but consequential ways how we study the world around us.

This research analyzes county-level support for the 2002 Constitutional Amendments for Voluntary Universal Prekindergarten Education (VPK) and Florida’s Amendment to Reduce Class Size (CSA). Three regression models are constructed for each dependent variable (support for VPK and support for CSA): a bivariate model with political party, a full model with all theoretically identified variables, and a best model with just the independent variables that have the most explanatory power. A variety of socioeconomic, demographic and political independent variables are tested. Four independent variables had a statistically significant positive relationship with support for both VPK and CSA: Democratic registration, Hispanic population, higher education, and population density.