Theses and Dissertations from UMD

Permanent URI for this communityhttp://hdl.handle.net/1903/2

New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM

More information is available at Theses and Dissertations at University of Maryland Libraries.

Browse

Subject: search.filters.subject.statistics

Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Toward a Theory of Risk Information Processing: The Mediating Effects of Reaction Time, Clarity, Affect, and Vividness
    (2011) Skubisz, Christine; Turner, Monique M; Communication; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    This project examined the variables that mediate the relationship between the exogenous variables numerical presentation and numeracy and the endogenous variables risk perception and risk related decisions. Previous research suggested that numerical format and numeracy influence outcomes. The question that remained unanswered was why? The goal of this project was to peer into the proverbial black box to critically examine information processing at work. To examine possible mediating variables, two theoretical models that have emerged in the risk perception literature were tested. The first is an evolutionary theory proposing that over time, individuals have developed an augmented ability to process frequency information. Thus, frequency information should be clearer and people should be faster at forming risk perceptions with information in this format. According to this model, processing speed and evidence clarity mediate the relationship between evidence format and risk perception. A second framework, the affective processing theory, argues that frequency information is more vivid and people derive more affect from information in this format. Therefore, according to this model, affect and vividness mediate the relationship between presentation format and risk perception. In addition to these two perspectives, a third theory was proposed and tested. The integrated theory of risk information processing predicted that reaction time, clarity, affect, and vividness would all influence risk perception. Two experiments were conducted to test the predictions of these three theories. Overall, some support for an integrated model was found. Results indicated that the mediating variables reaction time, clarity, affect, and vividness had direct effects on risk perception. In addition, risk perception had a strong influence on risk related decisions. In Study 2, objective numeracy had a direct effect on reaction time, such that people with high numeracy spent more time forming risk evaluations. Furthermore, people with a preference for numerical information evaluated numerical evidence as clearer and more vivid than people who preferred to receive evidence in nonnumerical formats. Both theoretical and applied implications of these results are discussed and recommendations for future research are provided.
  • Thumbnail Image
    Item
    Scattering from chaotic cavities: Exploring the random coupling model in the time and frequency domains
    (2009) Hart, James Aamodt; Ott, Edward; Antonsen, Thomas M; Physics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Scattering waves off resonant structures, with the waves coupling into and out of the structure at a finite number of locations (`ports'), is an extremely common problem both in theory and in real-world applications. In practice, solving for the scattering properties of a particular complex structure is extremely difficult and, in real-world applications, often impractical. In particular, if the wavelength of the incident wave is short compared to the structure size, and the dynamics of the ray trajectories within the scattering region are chaotic, the scattering properties of the cavity will be extremely sensitive to small perturbations. Thus, mathematical models have been developed which attempt to determine the statistical, rather than specific, properties of such systems. One such model is the Random Coupling Model. The Random Coupling Model was developed primarily in the frequency domain. In the first part of this dissertation, we explore the implications of the Random Coupling Model in the time domain, with emphasis on the time-domain behavior of the power radiated from a single-port lossless cavity after the cavity has been excited by a short initial external pulse. In particular, we find that for times much larger than the cavity's Heisenberg time (the inverse of the average spacing between cavity resonant frequencies), the power from a single cavity decays as a power law in time, following the decay rate of the ensemble average, but eventually transitions into an exponential decay as a single mode in the cavity dominates the decay. We find that this transition from power-law to exponential decay depends only on the shape of the incident pulse and a normalized time. In the second part of this dissertation, we extend the Random Coupling Model to include a broader range of situations. Previously, the Random Coupling Model applied only to ensembles of scattering data obtained over a sufficiently large spread in frequency or sufficiently different ensemble of configurations. We find that by using the Poisson Kernel, it is possible to obtain meaningful results applicable to situations which vary much less radically in configuration and frequency. We find that it is possible to obtain universal statistics by redefining the radiation impedance parameter of the previously developed Random Coupling Model to include the average effects of certain classical trajectories within the resonant structure. We test these results numerically and find good agreement between theory and simulation.
  • Thumbnail Image
    Item
    Integrating Statistics and Visualization to Improve Exploratory Social Network Analysis
    (2008-08-21) Perer, Adam; Shneiderman, Ben; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Social network analysis is emerging as a key technique to understanding social, cultural and economic phenomena. However, social network analysis is inherently complex since analysts must understand every individual's attributes as well as relationships between individuals. There are many statistical algorithms which reveal nodes that occupy key social positions and form cohesive social groups. However, it is difficult to find outliers and patterns in strictly quantitative output. In these situations, information visualizations can enable users to make sense of their data, but typical network visualizations are often hard to interpret because of overlapping nodes and tangled edges. My first contribution improves the process of exploratory social network analysis. I have designed and implemented a novel social network analysis tool, SocialAction (http://www.cs.umd.edu/hcil/socialaction) , that integrates both statistics and visualizations to enable users to quickly derive the benefits of both. Statistics are used to detect important individuals, relationships, and clusters. Instead of tabular display of numbers, the results are integrated with a network visualization in which users can easily and dynamically filter nodes and edges. The visualizations simplify the statistical results, facilitating sensemaking and discovery of features such as distributions, patterns, trends, gaps and outliers. The statistics simplify the comprehension of a sometimes chaotic visualization, allowing users to focus on statistically significant nodes and edges. SocialAction was also designed to help analysts explore non-social networks, such as citation, communication, financial and biological networks. My second contribution extends lessons learned from SocialAction and provides designs guidelines for interactive techniques to improve exploratory data analysis. A taxonomy of seven interactive techniques are augmented with computed attributes from statistics and data mining to improve information visualization exploration. Furthermore, systematic yet flexible design goals are provided to help guide domain experts through complex analysis over days, weeks and months. My third contribution demonstrates the effectiveness of long term case studies with domain experts to measure creative activities of information visualization users. Evaluating information visualization tools is problematic because controlled studies may not effectively represent the workflow of analysts. Discoveries occur over weeks and months, and exploratory tasks may be poorly defined. To capture authentic insights, I designed an evaluation methodology that used structured and replicated long-term case studies. The methodology was implemented on unique domain experts that demonstrated the effectiveness of integrating statistics and visualization.
  • Thumbnail Image
    Item
    Life's Rich Pattern: The Role of Statistics and Probability in Nineteenth Century Argumentation for Theories of Evolution, Variation, and Heredity
    (2006-04-26) Wynn, James; Fahnestock, Jeanne; English Language and Literature; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Though modern philosophers of science recognize the inappropriateness of the reduction of all scientific investigations to mathematics, mathematics and science share a long history with one another during which mathematics has been employed as a major component of scientific argumentation. Over the last twenty years, rhetoricians have done substantial work studying the role of argumentation in science (Bazerman 1988; Gross 1990, 2002; Myers 1990; Fahnestock 1999); however, despite the importance of mathematics in making scientific arguments, little effort has been made to understand the role mathematics has played in making these arguments. This dissertation represents a move to resolve this shortcoming by investigating the role of mathematics in arguments in evolutionary biology from the middle of the nineteenth to the beginning of the twentieth century. In the first part of the nineteenth century, the mass collection and mathematical assessment of data for scientific purposes provides the context for understanding some of the rhetorical choices of an important group of natural philosophers and biologists who developed arguments in the second half of the century about the nature of variation, evolution, and heredity. In the works of Charles Darwin, Gregor Mendel, Francis Galton, and Karl Pearson, arguments from probability and statistics play important roles as support for their arguments and as a source of invention for their claims. This investigation of the rhetorical situations of these four biologists, their arguments, and the role of the principles, operations, and formulae of probability and statistics supports the position that mathematization had a major impact on the nature of scientific evidence in the nineteenth century. What it also suggests is that, though mathematized arguments may have had a great deal of credibility within the scientific community in general, factors such as the stature of the rhetor and of their biological theory within their specific discourse communities played an equally important role in the persuasiveness of their arguments.