Dissertation Defense Announcements

In this study, the performance of the Lambda Coefficient of Rater-Mediated Agreement was evaluated with other chance-corrected agreement coefficients. Lambda is grounded in rater-mediated assessment theory and was developed as an alternative to Kappa (Cohen, 1960) and other chance-corrected agreement coefficients. Lambda has two variations, a general form that is calculated similarly to how most chance-corrected agreement coefficients are calculated, such as Kappa (Lambert et al., 2021). The general form of Lambda is referred to as Lambda-1. Lambda-2 differs from Lambda-1 in the calculation of the proportion of expected chance agreement. Lambda-2 uses known population proportions when available and applies those proportions in the calculation of expected chance agreement. In total, six coefficients were calculated using generated data by varying the amount and location of agreement and disagreement between ratings across two-, three-, and four-point rating scales. The exact agreement specifications ranged from 75% to 95% across 135 planned data conditions. The simulations adjusted prevalence indices according to exact agreement specifications (Xie, 2013). Results demonstrated the robustness of Lambda-1 and Lambda-2 to data conditions that are problematic for other coefficients. Both variations of Lambda produced benchmark agreement results that maintained meaning that may be diminished by other coefficients.

Recent technological advances have facilitated the use of mobile robots for a wide range of coverage applications such as inspection and mapping of infrastructure, precision agriculture, and disaster management. With the proliferation of these tasks comes an increasing need for autonomous systems to efficiently gather data pertinent for analyzing the state of the environment. The dissertation answers the following fundamental question: How should resource-constrained robots traverse the environment to collect data from all the relevant features? These features of interest can be represented as points, lines or curves, and areas. This dissertation unifies simultaneous coverage of all three types of features into a novel generalized coverage framework, develops algorithms for efficient coverage using multiple mobile robots, and validates them in experiments.

The dissertation comprehensively studies the line coverage problem, i.e., coverage of one-dimensional features, which lays the foundation of the generalized coverage problem. We develop algorithms to transform point and area features into linear features and use line coverage algorithms to solve generalized coverage efficiently. The algorithms substantially improve the state of the art while incorporating battery life constraints, nonholonomic constraints for robots that cannot take turns in place, and multiple home locations for large-scale environments.

Organic-inorganic hybrids may offer material properties not available from their inorganic components. However, they are typically less stable and disordered. A group of highly ordered II-VI based hybrid structures has been shown to possess various unusual properties and potential applications. As a prototype, β-ZnTe(en)0.5 can be viewed as a superlattice with alternating layers of two-monolayer thick (110) ZnTe and single-molecule length ethylenediamine. In contrast to all the known inorganic superlattices where interfacial diffusion is inevitable, we demonstrate in this thesis that β-ZnTe(en)0.5 exhibits an unusually high degree of crystallinity, as is evidenced by < 25′′ XRD rocking curve linewidth and < 1 cm-1 Raman linewidth, which are comparable to many high-quality binaries. Besides manifesting in the macroscopic scale crystallinity characterization, it also shows an exceptionally low level of microscopic scale defects, as suggested by the observed linear dependence of PL intensity on the excitation density over 6 orders of magnitude, which has not been possible even for the very high-quality CdTe and GaAs.
β-ZnTe(en)0.5’s highly-ordered crystallinity enables a systematic investigation of its vibrational property. We apply the orthogonal polarization and the angle-resolved polarization Raman techniques to study β-ZnTe(en)0.5’s vibrational modes. A set of orthogonal polarizations are used to decouple the vibration modes according to their symmetries. A mode-by-mode analysis allows unambiguous assignment for the Raman-active modes. A few exceptions and additional features are discussed. With the assignment, we demonstrated that the Raman tensor could be estimated from both the orthogonal technique and the angle-resolved technique. The two independent measurements yield consistent estimations. In addition, it has been shown that a combination of the two techniques enables unambiguous determination of the crystal orientations.
A distinction among the hybrid materials is its unprecedented ambient long-term stability over 15 years, which is still limited by extrinsic mechanisms but is already the longest documented hybrid semiconductor. In this work, we used Raman spectroscopy to investigate its degradation in air and a protected condition and framed the factors contributing to its long-term stability into (1) intrinsic effect such as large formation energy and large activation barrier in excess of the formation energy; (2) extrinsic factors, including surface or edge effect, where degradation can initiate through processes such as oxidation, and the structural defects, which may provide more accessible paths for degradation. Based on this approach, we estimate the room-temperature lifetime of β-ZnTe(en)0.5 in a protected environment can achieve 1.9x10^8 years, while in the ambient air, its lifetime is on the order of 10^1 years.

Teacher evaluations are routinely conducted across the United States for licensure and professional development supports. However, there is limited research on the interrater reliability of these evaluation assessment systems, despite federal recommendations (Graham et al., 2012). This research explores the systematic approach to interrater reliability utilized by the Early Educator Support (EES) Office in North Carolina. The EES Office supports the Birth-through-Kindergarten (B-K) teacher licensure of over 900 early educators in both private and public sectors. The evaluators employed undergo extensive trainings and hold a B-K license themselves. As part of the training, the evaluators undergo an interrater reliability activity that requires them to rate ten fictitious profiles, using the North Carolina Teacher Evaluation Process (NCTEP) Rubric. This research aims to understand the evaluator response process. In this study, Many Facets Rasch Models are used to understand evaluator patterns of strictness, leniency and potential bias based on the race of teacher profile. Additionally, two of the models are compared to understand the extent that these rater response patterns are exhibited in their real caseloads of actual early educators. In conclusion, the group of evaluators do show evidence of strictness, leniency, and bias, however it is mostly exhibited by a small number of individual evaluators. It is possible to use the results to inform the professional growth of these evaluators, so that all early educators served by the EES Office receive valid, fair, and reliable teacher evaluations. Furthermore, it depicts a systematic approach to interrater reliability that could be used by other evaluation systems across the country.