<p> Current economic development policy is unevenly effective across Appalachia. At the root of this policy is the annual determination of County Economic Status, a metric intended to measure performance in relation to the rest of the nation. This designation, calculated from income, unemployment and poverty data, is deeply flawed. Understanding how public investment and economic accessibility interact can yield improved results when implementing development policies and funding initiatives aimed at the region. Failure to recognize the assets and challenges unique to the region, coupled with an imperfect understanding of the regional interactions between local economies, have led to mis-targeted programs and unsatisfactory results. </p><p> A comprehensive examination of how public and private facilities interact to improve the overall well-being of a region in terms of the spatial patterns of accessibility and investment can increase understanding of the role of public investment in Appalachia, and improve program targeting. An economic accessibility model is generated and evaluated at the community level, and evaluated in conjunction with public investment practices, to uncover important information about the effectiveness of development efforts and the evolution of economic regions.</p>

<p> In higher education and in engineering education in particular, changing majors is generally considered a negative event - or at least an event with negative consequences. An emergent field of study within engineering education revolves around understanding the factors and processes driving student changes of major. Of key importance to further the field of change of major research is a grasp of large scale phenomena occurring throughout multiple systems, knowledge of previous attempts at describing such issues, and the adoption of metrics to probe them effectively. The problem posed is exacerbated by the drive in higher education institutions and among state legislatures to understand and reduce time-to-degree and student attrition. With these factors in mind, insights into large-scale processes that affect student progression are essential to evaluating the success or failure of programs. </p><p> The goals of this work include describing the current educational research on switchers, identifying core concepts and stumbling blocks in my treatment of switchers, and using the Multiple Institutional Database for Investigating Engineering Longitudinal Development (MIDFIELD) to explore how those who change majors perform as a function of large-scale academic pathways within and without the engineering context. To accomplish these goals, it was first necessary to delve into a recent history of the treatment of switchers within the literature and categorize their approach. While three categories of papers exist in the literature concerning change of major, all three may or may not be applicable to a given database of students or even a single institution. Furthermore, while the term has been coined in the literature, no portable metric for discussing large-scale navigational flexibility exists in engineering education. What such a metric would look like will be discussed as well as the delimitations involved. </p><p> The results and subsequent discussion will include a description of changes of major, how they may or may not have a deleterious effect on one's academic pathway, the special context of changes of major in the pathways of students within first-year engineering programs students labeled as undecided, an exploration of curricular flexibility by the construction of a novel metric, and proposed future work.</p>

<p> Efforts are currently underway in the US federal context to improve and strengthen evaluation practice and increase the use of evaluation results to inform policies and programs. However, these efforts remain unrealized, due partly to the lack of a comprehensive theoretical framework that views evaluation and related organizational processes and institutions as part of a larger system. Early intuitive theoretical taxonomies of evaluation policy suffer from the lack of connection to specific examples and instances, and are missing clear classification criteria that would allow practical application. To generate a grounded taxonomy of evaluation policy, this study surveyed members of the American Evaluation Association in 2009, asking them to generate examples of evaluation policy, and then to sort and rate these suggested policies. Results are analyzed using the concept mapping method of Trochim (1989), which first translates aggregate sorting decisions into conceptual &ldquo;distances&rdquo; on a two-dimensional dot map, then uses hierarchical cluster analysis to generate groupings of ideas. These groupings become the foundation for categories in a theoretical taxonomy. Findings reveal several different dimensions by which participants grouped evaluation policies, including the dimensions of &ldquo;value&rdquo; and &ldquo;policy mechanism.&rdquo; A values-by-mechanisms taxonomy and instructions for its use in an evaluation policy inventory process are proposed.</p>

<p> Military personnel and first responders operate in complex operational environments, and must be able to perform under physical, psychological, and emotional stress. Research suggests that resiliency assuages stress and improves the performance of military personnel and first responders. However, there are no studies examining the effects of resiliency on military first responders in training. The purpose of this research was to determine whether the dispositional hardiness traits of commitment, control and challenge displayed by Marine aircraft rescue and firefighting (ARFF) specialist trainees correlated to success in classroom performance, success during practical exercises, higher graduation rates. The theoretical foundation for this ex post facto quantitative study was psychological and organizational resiliency, as represented by Kobasa's hardiness theory. The convenience sample consisted of 60 Marine ARFF specialists trainees using self-report surveys during 2013. Independent samples t tests and hierarchical regression analyses revealed no statistical significance between higher hardiness levels and academic and practical application performance, although physical injury and other factors not measured by the hardiness construct were found to impact graduation rates negatively. The implications for positive social change include expanding organizational conceptions of resilience to measure dispositional factors not assessed by hardiness. This study may also offer insights into improving Marine Corps and first responder selection, training, and educational programs, as well as their performance and quality of life.</p>

<p> Several researchers have found a link between inequality across groups within a country and armed conflict. However, this research has focused almost exclusively on inequalities in income and political access, forsaking other key societal dimensions such as education. In this article, I examine the link between educational inequality across groups within a country and armed conflict. I use data from the Worldwide Inequality Database on Education to create measures of educational inequality by gender, wealth, and country region for developing countries across the world. I combine these measures with data on income and political inequality in order to better establish the unique association between educational inequality and armed conflict. Overall, I find a positive association between educational inequality and the likelihood that a group within a given country and year will be involved in armed conflict. This association is particularly strong across gender groups and country regions. These findings have implications for conflict prediction, preventative diplomacy efforts, and domestic and international organizations focused on educational and human development.</p>

<p> This paper explores the relationship between a state's use of voter-approved debt and its credit ratings. The variation in credit ratings from 1973 - 2008 across the 50 US states is explained as a function of states' use of voter-approved debt while controlling for confounding variables. The analysis attempts to estimate the effect of issuing voter-approved debt on credit ratings relative to the effect of issuing legislature-approved debt using a panel dataset constructed from three data sources: the National Conference of State Legislature's Ballot Measure Database, the US Census Bureau's Survey of Government Finances and Standard &amp; Poor's credit ratings. While prior literature has focused on the effect of voter approval requirements on measures of state credit health, this paper investigates the use of voter-approved debt by relying on a variable that measures the share of voter-approved debt issued by a state, in a given year and over time. Ordered probit models controlling for state and year fixed effects, as well as state demographics, finances, economic performance and financial institutions are used to explore the relationship between the use of voter-approved debt and a state's credit rating. The paper finds a statistically significant negative relationship between a state's use of voter-approved debt and its credit ratings. The results show that issuing 60% of state debt using voter-approval (the average for states that issue voter-approved debt in a given year) is related to a 0.71 lower state credit rating on a scale from 1-7 (BBB=1, AAA=7).</p>

<p> International technology diffusion reflects global R&amp;D production and collaboration that increasingly accompany other forms of international activity such as trade and foreign direct investment. This thesis studies country-level market flows of disembodied technology or intangibles trade. The main conceptual premise is that operations of MNC subsidiaries have a substantial effect on these market-based flows, consistent with public goods aspects of industrial knowledge and with theories on MNC R&amp;D strategies. Extensive previous country-level work relating FDI and technology flows focuses largely on knowledge spillovers (benefits from involuntary, uncompensated knowledge flows). Further, this study considers simultaneously two types of MNC subsidiaries (foreign owned subsidiaries and overseas subsidiaries of domestic MNCs) to acknowledge the likely role of two-way FDI (measured by MNC activities) in intangibles trade. In turn, the influence of these subsidiary groupings on intangibles trade reflects varied motives of the underlying R&amp;D investments. The predicted effects of MNC operations on intangibles trade result in hypotheses that are tested with published aggregate statistics from the U.S. balance of payments on total U.S. exports and imports in R&amp;D services as the dependent variables. </p><p> Theoretically, the thesis introduces the concept of reverse knowledge transfer from international business research to the study of bilateral intangibles trade. More generally, the thesis contributes to the literature by integrating macro and micro perspectives useful to understand the direction and nature of disembodied technology flows. In particular, the conceptual approach is consistent with macro trade models (two-way trade and two-way FDI from new trade theory), international business research, knowledge-based and transaction costs theories of MNCs (internalization of knowledge production and transfer), and innovation theory (knowledge seeking/exploiting). Consistent with these theoretical considerations, the empirical implementation considers panel countries as both exporters/importers of intangibles and host/home countries of R&amp;D-performing MNC subsidiaries. In turn, estimated equations use panel econometrics to relate observed heterogeneity in the geographic structure of bilateral trade with the geographic distribution of MNC operations for the two types or groupings of MNC subsidiaries. </p><p> The main conceptual premise of this study was supported by the empirical findings. In the aggregate, U.S. MNCs and foreign MNCs with U.S.-located subsidiaries appear to engage in knowledge seeking R&amp;D investments that influence transactions captured in balance of payment statistics. At the same time, the hypotheses regarding the effect of value added operations were not sustained statistically, failing to support knowledge exploiting as conceptualized here. </p><p> International transactions in intangibles in the form of services trade have yet to be integrated in the mainstream S&amp;T policy literature. The analysis of aggregate R&amp;D services trade pursued in this study may complement research on industrial knowledge flows based on other S&amp;T indicators (or levels of aggregation) thus potentially allowing monitoring and analysis of international technology diffusion earlier in the innovation cycle (e.g., before or apart from patenting), and suggests the potential of non-spillover flows as targets of international S&amp;T policy tools, perhaps in conjunction with trade and investment policy frameworks. The study also discusses the need for enhanced and integrated domestic and international statistics on R&amp;D and related intangibles to support future research and the design or modification of policy tools to monitor and facilitate cross-border flows of industrial knowledge.</p>

<p> The backcountry skiing population has increased significantly over the past ten years. Current sales of backcountry ski and rescue gear outpace more traditional alpine, telemark, and snowboarding equipment. As higher numbers of people head into avalanche terrain more fatal accidents are occurring, human triggered avalanches being the primary cause of death. In the late 1980s there were an average of 11 avalanche related deaths per year in the United States. That number has risen to an average of 30 over the past five years. Over that time avalanche specific rescue gear has been quick to adapt to the changing needs of backcountry skiers. However, over that same period of time theoretical approaches to risk management have been slow to adjust to the community's changing needs. Consequently, this study examined the preemptive decisions and the risk perceptions skiers held as they prepared to enter Tuckerman Ravine. Results show that 69% of the survey population did not have formal avalanche training. Despite significant technological advances 56% of the survey population did not have everyone in their group carry a beacon, shovel and probe. Fifty-one percent of the survey population strongly disagreed with frequent beacon practice. However, a skier was more likely to practice with their beacon if they had taken a formal avalanche course. Women were less likely to feel as though the trip were a failure if goals were not met. Additionally, the underrepresentation of women in this study (<i>n</i>=4) required an assessment of broader gender disparities within the backcountry skiing community. Quantitative and qualitative results show a lack of female leadership among recreational backcountry skiers. This leadership gap limits mentoring opportunities and decreases accessibility to the community for novice female backcountry skiers. The lack of gender diversification has significant affects on the decision-making process. This study concludes that dual-process cognitive methods and multilayered risk management protocols can increase margins of safety and encourage diverse perspectives within the larger backcountry skiing community. This paper suggests that because backcountry skiing involves the interconnection between complex adaptive systems and complex physical systems skiers should adopt a systems thinking approach to their decision-making processes.</p><p> <i>Keywords:</i> Systems thinking, avalanches, dual-process cognition, complexity, preemptive decision making, risk management, female backcountry leadership, risk perception, human triggered avalanches.</p>

<p> The transition to late-capitalism in the U.S. has generated extensive societal change. This paper examines the intersection of three of these changes: the transition to a consumer-oriented economic and societal model, the increase of economic insecurity experienced by individuals and households and the heightened emphasis on a short-term orientation in individual decision-making. A review of literature from the fields of Sociology, Economics and Psychology describes differing understandings of how individuals react to the heightened economic insecurity that households experience under late-capitalism. Within mainstream Economics', theoretical and empirical work suggests that individuals respond to insecurity by reducing spending and maximizing long-term financial well-being. However, recent work by theorists such as Zygmunt Bauman and Richard Sennett suggest that the proliferation of risk that occurs under late-capitalism weakens individuals' ability to anticipate future risks and leaves them vulnerable to society's emphasis on short-term thinking. </p><p> In response to these two interpretations, I analyze differences in the effect of household employment insecurity on consumer behavior over time. Part one of the analysis is a descriptive evaluation using six years (1980-2005) of General Social Survey data and compares levels of perceived job insecurity across demographic and occupational groups, while controlling for occupation-specific unemployment levels. Results show that perceived insecurity increased during this time period, regardless of individual characteristics however, members of traditionally 'secure' groups had lower initial levels of perceived insecurity and sharper growth such that, by 2005, levels of job insecurity are not well predicted by demographic or occupational characteristics. </p><p> The second part of this study analyzes data from BLS' Consumer Expenditure Survey. A two-level regression model for repeated cross-sectional data decomposes the effect of occupational characteristics, the unemployment rate and time on households' spending in nine expenditure categories. Models include interaction effects to assess changes in the effect of occupational characteristics over time. Results suggest that occupational characteristics weakened as predictors of spending behavior during this time series, especially in 'long-term oriented' categories (e.g. housing, education). However, the presence of an earner in a high insecurity occupation is significantly associated with increased spending on 'short-term oriented' good and services.</p>

<p> Locating hydrocarbon reservoirs has become more challenging with smaller, deeper or shallower targets in complicated environments. Controlled-source electromagnetics (CSEM), is a geophysical electromagnetic method used to detect and derisk hydrocarbon reservoirs in marine settings, but it is limited by the size of the target, low-spatial resolution, and depth of the reservoir. To reduce the impact of complicated settings and improve the detecting capabilities of CSEM, I apply synthetic aperture to CSEM responses, which virtually increases the length and width of the CSEM source by combining the responses from multiple individual sources. Applying a weight to each source steers or focuses the synthetic aperture source array in the inline and crossline directions. To evaluate the benefits of a 2D source distribution, I test steered synthetic aperture on 3D diffusive fields and view the changes with a new visualization technique. Then I apply 2D steered synthetic aperture to 3D noisy synthetic CSEM fields, which increases the detectability of the reservoir significantly. With more general weighting, I develop an optimization method to find the optimal weights for synthetic aperture arrays that adapts to the information in the CSEM data. The application of optimally weighted synthetic aperture to noisy, simulated electromagnetic fields reduces the presence of noise, increases detectability, and better defines the lateral extent of the target. I then modify the optimization method to include a term that minimizes the variance of random, independent noise. With the application of the modified optimization method, the weighted synthetic aperture responses amplifies the anomaly from the reservoir, lowers the noise floor, and reduces noise streaks in noisy CSEM responses from sources offset kilometers from the receivers. Even with changes to the location of the reservoir and perturbations to the physical properties, synthetic aperture is still able to highlight targets correctly, which allows use of the method in locations where the subsurface models are built from only estimates. In addition to the technical work in this thesis, I explore the interface between science, government, and society by examining the controversy over hydraulic fracturing and by suggesting a process to aid the debate and possibly other future controversies.</p>