Graduate College

Permanent URI for this community

https://udspace.udel.edu/handle/19716/5009

The doctoral dissertation and master's thesis reflects the scholarly research in a graduate program as required for the completion of the degree at the University of Delaware.

Browse

Browsing Graduate College by Title

Now showing 1 - 20 of 5416
  • Thumbnail Image
    Item
    100 Hz 512x512 SLEDS system design
    (University of Delaware, 2014) Nabha, Kassem
    Infrared (IR) detectors applications are widely used across fields from scientific and military to medical and industrial. IR can detect information that human eyes cannot perceive allowing the advancement in science and technology. ☐ Due to the importance of the infrared detectors in today's world, it is necessary to accurately test and characterize these detectors with a frame of reference related to the application. IR projection systems are a great way to characterize the detectors because they can be used as high accuracy reference. ☐ Described in this work is a 512x512 super-latticed light emitting diode system (SLEDS) operating at a frame rate of 100Hz. This system has been fully developed, tested, and corrected for non-uniformity (NUC). Further work will be done to achieve higher frame rates and two frequencies of emission (colors) instead of one.
  • Thumbnail Image
    Item
    210Polonium and 210lead radionucidles in the Delaware and Chesapeake estuarine and coastal regions
    (University of Delaware, 2013) Marsan, David
    One of the primary objectives of this thesis is to present an integrated study of the 210Pb and 210Po radioactive tracers, and assess their use as tracers for particle and trace element export from different estuarine and coastal environments. In order to achieve this main objective, the thesis was split into three sections, each enhancing the understanding of the natural radionuclides 210Pb and 210Po in the estuarine and costal system. Chapter 1 is dedicated to testing the sampling and analytical methods of 210Po and 210Pb extraction from estuarine and coastal waters, specifically comparing the two most widely used scavenging methods, Fe(OH)3 and CoRAPDC. The chapter describes experiments conducted on about 100 samples collected from the Delaware estuary, Chesapeake estuary, Delaware intertidal marsh and an offshore continental slope site. Data in the chapter clarifies the accuracy and reliability of each method and includes suggestions to enhance them. Other details include results of calculations, error propagation, spike calibration, plating efficiency and the MnO2 scavenging method. Chapter 2 presents a synthesis of the estuarine and coastal biogeochemistry of 210Po and 210Pb in the Delaware and Chesapeake estuaries. A single box model is presented using steady state equations to determine residence times of the radionuclides. This chapter presents five highlights: 1) How 210Pb and 210Po dissolved and particulate data can revel key biogeochemical processes and rates in estuaries; 2) Are regional differences in estuaries dominated by a single or compilation of biogeochemical process; 3) Do subRoxic bottom waters affect the distribution of 210Po and 210Pb; 4) Can disequilibria between parent (210Pb) and grandRdaughter (210Po) be used to identify and quantify principle processes; 5) Will a simple mass balance model result in reliable net scavenging residence times for the Delaware and Chesapeake estuaries. Chapter 3 will advance the simple single boxRmodel from chapter 2 to a more complex twoRlayer model. The model will include evaluations of the fate of not only 210Po and 210Pb in the Delaware and Chesapeake estuaries but also the trace elements Fe, Cu, Cd, Ni, Zn, Pb, Cr, Co and Mn. Residences times, presented as halfRlives along with rates and partition coefficients will be identified for the water column including at the sediment water interface. The thesis will revisit the major conclusions obtained from the work presented along with suggestions for future work. Appendix sections include supporting hydrographic data and a compilation of salt marsh trace metal results conducted in Graz, Austria. Trace metal work included a suite of 26 elements measured in a core, two species of mussels, Spartina)alterniflora marsh plant and the sea surface microlayer (SML) from an intertidal Delaware salt marsh.
  • Thumbnail Image
    Item
    3D permeability characterization of fibrous media
    (University of Delaware, 2010) Okonkwo, Kenneth
    In Liquid Composite Molding (LCM) processes, a liquid resin is forced to flow through dry fibrous preform, usually fabrics, to impregnate it and create the composite part in net or near-net shape. The principal advantage of LCM processes is their capacity to produce high fiber volume fraction and high quality parts under low pressure at low cost. The main conditions for successful manufacturing are complete filling of the mold and perfect impregnation of the reinforcement material. If these conditions are not met the structural properties of the finished part are significantly impaired by defects like voids. The mold filling depends on the permeability of the fibrous media. Permeability is an intrinsic property of fiber reinforcement, which includes all interactions between fibers and fluid and characterizes the ease of flow through the medium. The complete prediction of second-order permeability tensor is critical to understanding and prediction of flow in the resin transfer molding process of thick composites or where the flow process is three dimensional In this thesis a new approach for characterizing the three dimensional permeability tensor of fabrics used as reinforcement in liquid injection molding processes from a single experiment is presented and validated. In this approach, a liquid is injected into a preform placed in a mold containing 192 electrical resistance flow sensors radially embedded in the top and the bottom platens of the mold. The proposed method uses an optimization routine in which the permeabilities in a 3D flow simulation of the identical mold is updated continuously until the error between the simulation arrival times at all the 192 sensor locations and the experimental arrival time is minimum. The optimization routine systematically changes the values of the components of the permeability tensor using golden search method until the best match is obtained. The validation and sensitivity of this method is explored and it has been shown that this technique is promising for permeability characterization. The approach is shown to be valid for reinforcements with anisotropic and isotropic nature The advantage of this approach is that it can be used to obtain permeability values from a single experiment; there is no need to scale the circular injection inlet, and it is not limited to principal permeability values. The sensors utilized are unobtrusive to the flow unlike say optic fibers embedded in the fabric that interfere with the flow of the test fluid. The electrical resistance sensors used in this approach are embedded in mold platens instead and which flush with the surface. The method can be used to help predict and understand resin flow behavior during liquid molding of advanced composite materials.
  • Thumbnail Image
    Item
    A 3D photonic sensor integrated tissue model for strain sensing
    (University of Delaware, 2019) Geiger, Sarah J.
    The study of wound healing and wound healing therapies is motivated by the need to prevent the formation of thick scar tissue in pathologically healing wounds and tissues that rely on their elasticity and modulus to perform their function, such as cardiac and vocal fold tissues. The development of in vitro platforms that can detect cell-induced strain in mimics of healing wounds has expanded our understanding of the mechanical, chemical, and physical cues that drive wound healing. However, these platforms are limited in their resolution, dimensionality, and ability to gather information about changes in strain throughout thick, opaque tissue models. In this work we describe the development of flexible, deterministically buckled 3D photonic device arrays that are designed and fabricated to meet the specific spatial, temporal, and strain resolution requirements needed for the detection of cell-induced strains in a millimeters-thick tissue model. ☐ A polymer or silicone-clad Ge23Sb7S70 chalcogenide glass resonant cavity array is selected for this application, as high-quality chalcogenide glasses devices can be deposited at low temperatures onto flexible and cytocompatible substrates. However, the reliability of these and other highly sensitive chalcogenide glass devices is affected by their aging-induced structural relaxation. The refractive index shifts resulting from this relaxation are on the same order of magnitude as the index shifts used to small-scale strain with our device arrays. In order to overcome this limitation, we develop and demonstrate a high-precision refractometry technique that tracks small changes in the refractive index of Ge23Sb7S70 chalcogenide glass, down to 10-5 RIU. This technique allows us to both identify the aging mechanism in this glass with high accuracy and compare different index stabilization methods to optimize our device processing. ☐ The expected performance of these arrays was tested both through finite element modeling and a proof-of-concept in vitro experiment. In the modeling experiments, PDMS buckled geometries were deformed in cardiac graft tissue-like environments. From these experiments we showed that devices embedded in these materials could easily detect small, localized changes in stiffness theoretically caused by limited perfusion of growth factor throughout this model. In vitro, an SU-8 clad, symmetrically buckled device was exposed to a contracting collagen gel, and the device response as a result of this deformation was analyzed. ☐ These deterministically buckled arrays of polymer or silicone-clad chalcogenide glass resonant cavities demonstrate sensitivity to relevant strains in 3D cell culture platforms, excellent ease of use, and the potential for a wide range of applications. This technique can be used as a standalone, low cost, plug-and-play local strain gauge for use in soft material systems. Thus, this technique’s flexibility both in terms of its deformability and range of applications easily surpasses other methods of in vitro force or strain detection.
  • Thumbnail Image
    Item
    3D reconstruction from coded plenoptic sampling
    (University of Delaware, 2019) Zhou, Mingyuan
    The plenoptic function describes a scene in terms of light rays, it is a 7-dimensional function with spectral, directional, spatial, and temporal variation. Traditional plenoptic sampling is acquired either by employing a standard plenoptic camera or a camera array, and the spatial-angular sampling can be potentially used to model 3D surface. ☐ In this dissertation, I present three coded plenoptic sampling schemes, i.e., the rotational cross-slit (R-XSlit) plenoptic sampling, the wavelength coded plenoptic sampling, and the polarimetric plenoptic sampling. The additional coded sampling information, such as non-centric sampling, spectral sampling, and polarization sampling, are conducive to 3D reconstruction. Therefore, I also develop the corresponding 3D reconstruction framework for each of them. ☐ First, I introduce the R-XSlit plenoptic sampling scheme by exploiting a special noncentric camera called the crossed-slit or XSlit camera. An XSlit camera acquires rays that simultaneously pass through two oblique slits. I show that instead of translating the camera as in the pinhole case, we can effectively sample the 4D plenoptic sampling by rotating individual or both slits while keeping the camera fixed, which makes the plenoptic sampling coded in the spatial-angular domain. The theoretical analysis shows that it provides denser spatial-angular sampling, which is beneficial for scene reconstruction and rendering. I develop a volumetric reconstruction scheme for scene reconstruction. ☐ Second, I present two wavelength coded plenoptic sampling schemes in the visible and infrared spectrum respectively. I firstly design a compact system with lights and cameras arranged on concentric circles to acquire a concentric wavelength coded plenoptic sampling in the visible spectrum, the cameras on each ring capture images in a unique spectrum. I employ the Phong dichromatic model onto its plenoptic function for 3D reconstruction and spectral reflectance map estimation. Experiments show that our technique can achieve high accuracy and robustness in geometry recovery. Moreover, I present an infrared wavelength coded plenoptic sampling and develop a hybrid sensing framework to efficiently achieve pose estimation and face reconstruction by exploiting the captured reflected infrared rays from human eyes. ☐ Finally, I present a polarimetric plenoptic sampling framework for recovering 3D surfaces, the polarization of light is included in its plenoptic function. I employ a new analysis analogous to the optical flow to correlate the polarization radiance function with both surface normal and depth. The proposed framework effectively resolves the azimuth-zenith ambiguity by forming an over-determined system. Extensive experiments on both synthetic and real data demonstrate that the technique is capable of recovering extremely challenging glossy and textureless objects.
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
    95 GHz silicon germanium low noise amplifier as front-end receiver for sparse aperture millimeter wave imaging
    (University of Delaware, 2015) Wright, Andrew Alexander
    Demand for the ability to navigate in degraded visual environments (DVE) such as dust, smoke, and fog, has lead the development of millimeter wave (mm-wave) real-time imaging systems. Millimeter wavelength radiation has shown that the wavelengths are long enough to penetrate the obscurants while also allowing sufficient resolution. A low attenuation atmospheric window in the 95 GHz region has pushed for these systems to operate at these millimeter wavelength frequencies. Due to low signal levels at these frequencies, the system requires high gain in the front-end to boost the signals of mm-wave frequencies. This involves collecting the electromagnetic waves with a horn antenna and then amplifying the signal with a low noise amplifier (LNA) to maximize the signal to noise ratio (SNR). In photonics-based imaging systems, the mm-wave signal is then up-converted to optical domain, where it then propagates through optical fibers to an infrared camera for further processing. The horn, LNA and up-converter comprise a single module. A large distributed array of modules, around 200, are required for a real-time mm-wave imaging system capable of peering through DVE. As a result, pushing this technology to higher frequencies can be very costly, due to the high prices of individual high frequency components. Therefore, an alternative technology is required to keep the costs to a minimum. One approach to controlling costs of components operating at higher frequencies is to adopt an alternative amplifier technology. Conventionally, commercially available GaAs and InP LNAs are used to obtain high gain at the high frequencies, but at 95 GHz, each amplifier used to be thousands of dollars. Since then, costs for each amplifier decreased to $100. With each module requiring three or more amplifiers, costs become prohibitively high for many applications. Therefore, this thesis focuses on the development of a 95 GHz amplifier using silicon germanium (SiGe) technology to obtain the required high gain while maintaining low costs. To date, extensive efforts have been made in the development of SiGe amplifier technology and high gain was demonstrated at the W-band. However, existing amplifier technology does not meet the requirements of the mm-wave imager. In particular, major limitation is the 3-dB bandwidth of the gain curve. A distributed aperture system with a wide field of view and broadband response will experience a phenomenon known as fringe-washing if an off-axis signal arrives with significant delay between the receivers on the opposite ends of the longest baseline. Severe fringe-washing occurs when this delay of the projected baseline is comparable to the correlation time of the signal, i.e. the inverse bandwidth of the system. To mitigate fringe-washing, each module must limit the bandwidth of operation, which can be accomplished either with a filter or an amplifier. Since filters can be lossy, ideally a narrow-band amplifier is preferred. In this thesis, using the basic principle of amplifier design, an LNA is developed based on advanced SiGe-foundry processes to operate in the 95 GHz regime. The advantage of a custom SiGe amplifier is the ability to design it to meet the imager's specific demands, including gain, noise figure, bandwidth, and power consumption in a single low cost device. This thesis details such design, and includes the discussion of tradeoffs and limitations imposed by the commercial SiGe-foundry processes employed.
  • Thumbnail Image
    Item
    A BALANCING ACT: EXAMINING THE RELATIONSHIP BETWEEN SCHOOL LEADERSHIP AND POLITICAL TENSION IN EDUCATION DECISION MAKING
    Cahill, Amanda
    The United States has a long history of political tension around education at the federal, state, and local level. District and school leaders must balance students’ learning and social needs while working to address political tension barriers on education decisions. Political tension involves the feeling of strain or anxiety around topics aligned with political ideology. The strain and anxiety generally result in difficulty with efficiently moving forward with clear decision making, specifically in education. Education decisions are choices made by district and school leaders that affect students’ growth academically within the school setting. Examples of education decisions could include restructuring the administration team, endorsing a specific curriculum, or implementing an enrichment program. Researchers have written extensively about effective leadership approaches but there is a lack of literature on strategies for how school leaders, particularly at the district level, can address political tensions that create barriers to decision making. School leaders' abilities to address political tension would support efficacious decision making with a focus on positive outcomes for students.Overtime, as politics has become more contentious, districts’ decision making has come under attack. These attacks significantly disrupt the ways in which school districts make and implement reforms, school improvements, policy changes, and the like. District and school leaders lack strategies and resources to proactively address political tension barriers which results in sporadic and often poorly implemented or communicated decision making. These implementation and communication failures are then poorly received by the community. The constant politicization of school decisions has caused leaders to focus on responding to political tensions, rather than focusing on decision making that can positively impact students’ learning needs. Therefore, it is imperative that educational leaders are equipped and prepared to address political tensions that cause barriers so that they can remain focused on decision-making processes that result in positively supporting students’ learning and social needs. Education leaders need both the resources and the opportunity to apply decision-making strategies that simultaneously address political tensions and remove barriers in order to communicate more effectively and work harmoniously with their school community. In my educational leadership portfolio (ELP), I focus specifically on how political tensions on the national level are influencing the decision making of local district and school leaders. I evaluated the relationship between district and school leaders’ decision-making process and political tension barriers within a rural school district in the Mid-Atlantic region. I used a rural school district as a case study to explore strategies district and school leaders need to manage political tension barriers without compromising decision making aimed at supporting student growth academically, socially, and emotionally. In this ELP, I gathered data through a case study. The ELP proposes a decision-making framework that district and school leaders can proactively utilize when they predict, or encounter, political tensions that result in barriers that may impede their ability to make decisions.
  • Thumbnail Image
    Item
    A Bayesian Cue Integration approach to racial bias in pain assessment and treatment
    (University of Delaware, 2023) Goharzad, Azaadeh
    Patients' reports of subjective pain experience are at least nominally a primary diagnostic cue in assessment and treatment of pain. Despite this, there is low concordance between provider assessments and self-reported pain ratings, such that patient pain is regularly underestimated and undertreated. Such discrepancies in care are particularly stark for Black patients, who receive less adequate pain care compared to White patients. While attending to facial expressions of pain marginally improves concordance in patient-provider pain ratings, it is not clear that this would improve concordance for Black patients given previous work demonstrating blunted recognition of pain on Black faces. Moreover, it is unclear how self-reported pain information is integrated with facial expressions of pain, and whether this integration is similarly biased as a function of patient race. In the present paper we construct three models of pain assessment for both Black and White targets to examine how individuals use facial expression and self-reported pain cues in making holistic judgements of pain intensity as well as subsequent treatment decisions. Overall, we find that the Bayesian Cue Integration model (compared to Face Dominant and Self-Report Dominant models) best predict participant assessments of pain as well as treatment outcomes for both Black and White targets, suggesting that both facial expression and self-report are integrated in pain assessment. ☐ Keywords: Social cognition, pain, race, cue integration
  • Thumbnail Image
    Item
    A BOUNDING SURFACE PLASTICITY-BASED HYPERELASTIC CONSTITUTIVE MODEL FOR UNSATURATED GRANULAR SOILS
    Kadivar, Mehdi
    One-third or more of the earth’s surface is situated in arid or semi-arid regions where the potential evaporation exceeds the precipitation and soils exist in their unsaturated state. Unsaturated soils are also abundant in most parts of the world where there is seasonal groundwater table fluctuation. The variation in the degree of saturation gives rise to a gamut of variability in soil’s hydromechanical behavior. The co-existence of pore-air and water in the void spaces and their interaction with each other and the solid particles are the main reasons why such variability exists and why unsaturated soils are more complex than saturated or dry ones. A robust understanding of the hydromechanical properties of unsaturated soils is crucial for geotechnical engineers worldwide, as well as for those concerned with the interaction of structures with the ground. Some engineering problems associated with unsaturated soils include precipitation-induced shallow-depth landslides, settlement of soil in the vadose zone, drainage of roadway materials, and borehole stability. Proper understanding of unsaturated behavior requires considerations that go beyond those available for saturated soils. In pursuit of addressing this requirement, a plethora of research has been devoted to measuring, modeling, predicting, and interpreting unsaturated soil behavior. Many theories for characterizing the mechanical response, methodologies for laboratory testing, as well as equipment to determine the constitutive parameters have been developed. Instruments to study in-situ behavior of unsaturated soils have also been promoted and used in a few cases. The outcomes of past research include: the development and critical evaluation of various forms of the effective stress principle and its fundamental role in determining strength and deformation properties of unsaturated soils; identification of independent state variables; development of failure envelopes and yield surfaces; formulation of macromechanical and micromechanical constitutive relationships; and formulation of suction-induced stress as a component of the intergranular stress tensor. Despite the volume of work dedicated to the field of unsaturated soil mechanics, compared to two-phase (i.e., saturated) soils, relatively few advancements have been made in the development of characterization frameworks for unsaturated soils. In this doctoral research, a novel, 14-parameter, state-dependent bounding surface plasticity model that simulates the behavior of unsaturated granular soils is developed. In the development of this model, a critical state compatible hyperelastic formulation for saturated granular soils is selected as a base model and is enhanced and extended to predict unsaturated granular soil behavior. Accounting for deformation phenomena in unsaturated soils, the elastoplastic response has no purely elastic component. The hyperelasticity and assumption of no purely elastic deformation sets this model apart from existing ones. To handle the inherent hydro-mechanical coupling in unsaturated soils, a newer generation stress framework, consisting of the Bishop-type effective stress with a second stress variable, is used in conjunction with a soil-water characteristic curve function. Available unsaturated soil data for sands and silty sands were used to calibrate, validate, and assess the performance of the new model. Additional laboratory data, consisting of a suite of consolidated drained triaxial shear tests, were generated. The shear strength and volumetric behavior of a native mid-Atlantic transitional silty sand were investigated under varying values of matric suction, confining pressure, strain rate, and fines content. The experimental results are used to validate the predictive capabilities of the new bounding surface plasticity constitutive model for unsaturated granular soils. It is shown that with a set of parameter values, the model realistically simulates the main features that characterize the shear and volumetric behavior of unsaturated granular soils over a wide range of matric suction, density, and net confining pressure. In the literature, it was observed that multiple analytical expressions exist for effective stress, critical void ratio, and soil water characteristic curve. To see the effects that variations in these different analytical forms have on the model simulations of unsaturated soil behavior, a parametric investigation is performed using the constitutive model developed and the aforementioned in-house generated laboratory data. It is observed that, depending on the desired prediction accuracy, a variety of functions (with varying numbers of model parameters) could be implemented as part of the constitutive model.
  • Thumbnail Image
    Item
    A comparative evaluation of the microbiome effects on easy and hard keeper horses
    (University of Delaware, 2023) Johnson, Alexa C.B.
    Horses with different metabolic tendencies are anecdotally referred to as “easy” or “hard” keepers. Easy keepers (EK) tend to gain weight easily while hard keepers (HK) require extra feed to maintain body condition. Horses that do not struggle with maintaining a healthy weight are referred to as “medium keepers” (MK). The horse, as an obligate herbivore, relies on the gut microbiome to provide more than half of its energy requirements. Therefore, equine energetics and metabolism is greatly influenced by the gut microbiome. It is not yet known what causes a horse to be an EK, MK, or HK but, I hypothesize that the gut microbial structure and function play a vital role in equine metabolic tendencies. The dynamic interactions between the horse and its gut microbiome likely reflect individual capacities and genetics to harbor specific populations as well as host-specific abilities to utilize available nutrients. To test this central hypothesis regarding the microbial side of the conversation, these research projects focus on the bacterial and protozoal fractions. ☐ The first objective of this work was to develop a reliable and standardized tool for determining equine keeper status. The lack of a standardized method to identify equine keeper status requires reliance on the owner reported keeper statuses which is unreliable and irreproducible. The Equine Keeper Status Scale (EKSS) was developed and validated on data gathered from 240 horses. With EKSS assignments, incorrect keeper status assignments provided by owners was reduced by 60%. The EKSS was used in all further studies to identify EK, MK, and HK study cohorts. ☐ The second objective of the project was to compare bacterial composition (16S rRNA surveys) of EKSS statuses. 16S rRNA surveys of equine feces in an observational study (n=73) found differences in alpha and beta diversities and taxa abundances based on EKSS assignments. However, when a controlled cohort (n=12) was used, significances in alpha and beta diversities were lost, but unique bacterial cores and representative bacteria of each EKSS status were found. Determining the bacterial core of each EKSS status will aid during probiotic choice and probiotic development to target these key groups and improve EK and HK weight management strategies. ☐ The third objective was to compare protozoal composition (18S rRNA surveys) of EKSS statuses. As an extension to this objective, we sought to obtain reference sequences for uncharacterized protozoans to improve molecular methods to identify protozoans. Two previously unsequenced equine protozoan species (Blepharocorys valvata and Blepharoconus benbrooki) and two other equine protozoans (Tripalmaria dogieli, Cochliatoxum periachtum), were successfully single sorted and sequenced. After the addition of these sequences to the classifier, the protozoan (18S rRNA) profile of horses (n=39) in the Mid-Atlantic Region and EKSS statuses were evaluated. Thirty-five species level protozoans were identified in the Mid-Atlantic Region, and protozoal richness was lowest in HK horses compared to EK and MK animals (P = 0.05). Describing the commensal protozoal fraction is the first step towards ultimately understanding this population’s purpose during microbial metabolism and host health. ☐ The fourth objective was to determine if bacterial activities differed between EKSS statuses. Bacterial functionality between EKSS statuses was tested using PICRUSt to hypothesize fermentative differences between the gut microbiomes, and 48h in vitro challenge protocols. PICRUSt predictions were performed on the 16S rRNA surveys from the observational study (n=73) and found overall, 18 metabolism pathways were differentially abundant in EKSS statuses (P < 0.10); seven of which were significantly enriched in the EK representing both foregut metabolism (i.e. starch and lipid digestion), and hindgut metabolism (i.e. fiber and amino acid digestion). PICRUSt is inherently limited but these results indicate that the EK has an enhanced metabolic potential to breakdown feed and harvest energy compared to the MK and HK. ☐ In vitro experiments (n=12) demonstrated that the HK microbiome had the quickest and most drastic reaction to both carbohydrate and protein challenges. The slower metabolic response demonstrated by EK cultures may be a resiliency mechanism that these communities utilize to slow overall metabolism while increasing ATP production. These different bacterial responses suggest unique microbial strategies between keeper communities to maintain stability that may ultimately change metabolism patterns. ☐ In conclusion, the data collected from these studies supports our central hypothesis that the microbiome plays a pivotal role in equine keeper status. Our results suggest that bacterial activity and functionality are more influential towards equine keeper status than the bacterial composition. Results further suggest that the protozoan population is a significant contributor in equine keeper status and this population deserves further investigation.
  • Thumbnail Image
    Item
    A COMPARISON OF THREE EFFECT SIZE INDICES FOR COUNT-BASED OUTCOMES IN SINGLE-CASE DESIGN STUDIES
    Shrestha, Pragya
    In Single-Case Designs (SCD), the outcome variable most commonly involves some form of count data. However, statistical analyses and associated effect size (ES) calculations for count outcomes have only recently been proposed. Three recently proposed ES methods for count data are: Nonlinear Bayesian effect size (Rindskopf, 2014), Log Response Ratio effect size (Pustejovsky, 2018), and Bayesian Rate Ratio effect size (Natesan Batley, Shukla Mehta, & Hitchcock, 2021). Although all three methods calculate ES for count outcome data and can be used with an ABAB design, they use either different statistical modeling or a different estimation framework (Bayesian or frequentist), they may assume the presence or absence of autocorrelation, which is frequently present in SCD data and it is yet to examine how the ES and standard error estimates from these three ES indices are affected by overdispersion, a common occurrence in count data. These fundamental differences call for a closer examination and comparison of the methods and estimates obtained. The proposed dissertation aims to investigate the interpretability and understandability of the estimates produced as proposed by May (2004), examine if the three ES indices can be converted to a common metric to facilitate comparison of the ES estimates, document the benefits and challenges while implementing each method, and examine the performance of these ES methods under positive autocorrelation and overdispersion using Monte Carlo Simulation. Schmidt (2007), a published SCD study that examined the effect of Class-wide Function-related Interventions Teams (CW-FIT) on reducing the disruptive behavior of three first grade students using an ABAB design, was used to examine the interpretability and understandability of the estimates produced and whether the indices can be converted to a common metric. It consisted of 3 cases with 4 phases (ABAB) for each case. For the simulation study, 1000 datasets for each case were simulated using pre-specified data parameters (number of cases, number of data points within each phase of a case, and phase means) taken from Schmidt (2007) study and for various conditions of autocorrelation and overdispersion. A fully crossed factorial design with three autocorrelation (0.0, 0.2, 0.4) and four overdispersion (0.0001, 0.05, 0.1, 0.3) resulting in 12 simulation conditions for each case was used for the data generation purpose. All analyses were carried out in R software. Results indicate all three ES estimates are interpretable. LRR meets the understandability criteria, however both BRR and NLB require advance statistical knowledge to run the models. The three ES can be converted to a common metric because they are ratios of the mean count of the phases. Based on simulation, all the three methods produce almost unbiased estimates of the effect size under different data conditions, however the standard error is affected by autocorrelation and overdispersion. This dissertation can serve as a resource for other SCD researchers and applied practitioners to understand and interpret the different ES values from the LRR, NLB, and BRR methods and help them make better informed decision about which of the three ES indices to use in their own research study if there is presence of autocorrelation and overdispersion in their data.