Dissertation Defense Announcements

While women have attained higher-level roles and greater representation in higher education administrator roles, the concept of the higher, the fewer (Nidiffer, 2002) represents the “gendered prestige hierarchies” (Allan, 2011, p. 58) that limit women’s representation at more prestigious institutions while bolstering representation at institutions of greater access. In student affairs, women represent a majority of all professionals in the field, though representation in senior student affairs officer roles has not increased significantly since the 1980s (Blackhurst, 2000), with women more likely to advance at institutions with fewer than 1,000 students (Rickard, 1985a). While extensive research exists that illustrates the disparities for women in higher education and academic affairs, little comparable research exists for women in student affairs.

This qualitative study sought to understand how women in senior-level student affairs positions have navigated their experiences and career paths in order to advance to their current roles. Using critical and post-structural feminist perspectives of power, the researcher utilized a critical phenomenological approach to consider how nine women in senior-level student affairs positions in large, public institutions understood their career advancement. Participants completed two semi-structured interviews. The resulting themes include an encompassing theme of genderization, as well as four embedded themes: the person on the path, achieving through and with others, conflicting messages of competence and value, and overtasked and alone. Implications of the study require that policy, practice, and research all examine the ways that genderization perpetuates masculine-centered norms and power dynamics that penalize women for deviating from those norms. For current or aspiring student affairs leaders, organizational constructs must be dismantled in order to disrupt internalized genderization by women.

One of the questions in algebraic groups is about the asymptotic behavior of the probability of return of a random walk, which closely related on the growth rate of a group. Upper-triangular matrices form a group. Solvable groups have an exponential growth rate and it was shown that the asymptotic behavior of the probability of return on these groups has a fractional-exponential decal. The results in the paper by Molchanov and others, are different from the previous finding. They showed that in the case of solvable groups of upper-triangular 2x2 matrices the return probability of the Brownian motions has a polynomial decay. In this dissertation, we extended this research to the case of solvable groups of upper-triangular 3x3 matrices. The elements in the 3x3 matrices that define a Brownian motion on these groups contain integrals of geometric Brownian motions. These integrals have an important role in Asian and Asian-Basket options. We proved some properties of these integrals and showed that certain cases of geometric Asian-basket call options with two assets have a higher risk that the same type of put options. Which implies that some trading strategies might benefit from a reevaluation using a new risk assessment of geometric Asian-Basket.

This dissertation comprises a series of authored papers, delineated by Chapter, which include advanced manufacturing, both techniques and processes, throughout the design processes for the future development of high quality reflective optics. The dissertation includes a novel kinematic mount design used for manufacturing and metrology of a freeform optic, an experimental study on additively manufactured silicon carbide for optical applications, and a new design methodology for higher efficiency lightweight mirrors considering additive manufacturing as the main process chain. Freeform optics, additive manufacturing, and silicon carbide mirrors are disruptive technologies independently. The work described in this dissertation merges these disruptive technologies into a systematic framework that has the potential to revolutionize both the manufacturing process chain and the mechanical design of lightweight mirrors. The combination of the three papers of this dissertation lays foundational work in reflective optics for overcoming manufacturing challenges, and for advancing mechanical design in consideration of advanced manufacturing. The result is a significant advancement in the state of the art for the creation of silicon carbide, additively manufactured, high efficiency, freeform reflective optics.

The increasing performance of feature extraction and regression modeling in various domains raises the hope for machine and deep learning to assist clinicians in numerous healthcare applications. However, the complex and multimodal nature of the problems and the scarce resource of high-quality labeled data in this domain introduces several challenges and limitations. These challenges, along with lack of interpretability, undermines the generalizability and usability of many state-of-the-art machine learning models.

This dissertation focuses on using multimodal sources of data for regression modeling in healthcare applications. The argument is that domain knowledge describes the nature of each modality's relationship with the target function. This relationship can characterize the appropriate level of representation and an efficient integration method. We define a framework with two heterogeneous modalities, one modality provides more local features, while another contains higher-level global information. We demonstrate the framework's applicability for multiple healthcare regression tasks.

In this framework, we propose two approaches for increasing the performance in the absence of large-scale data: leveraging the abstraction of the modality representations based on domain knowledge, and a tree-structure convolutional neural network for integrating the information from the heterogeneous modalities. This framework is discussed in more detail for two different cases of "Alzheimer's disease progression prediction" and "radiation therapy treatment planning." The former predicts a scalar target variable, while the latter approximates a two-dimensional one. The first application's performance is compared with the previous submissions for the same dataset; it outperforms the best-reported results.