62588 research outputs found
Sort by
Latiné Students\u27 Perceptions of Dual Enrollment Programs
No full textThe purpose of this study was to understand the characteristics or factors that influence Latiné students\u27 perceptions of access to dual enrollment (DE) programs. Utilizing a modified version of Perna\u27s (2006) socioecological model of college choice, the study sought to answer the following questions: How do student, district and state-level characteristics influence Latiné students\u27 perceptions of access to DE programs? How do perceptions of access differ across DE participation status? Results from within-case and cross-case analyses and analyses from a critical realist perspective found two predominant themes influencing Latiné students’ perceptions of DE accessibility: the DE awareness gap and a sense of belonging. Results also indicate that school districts can significantly influence Latiné students\u27 perceptions of DE access depending on how they market and communicate about DE programs and foster an inclusive school climate.The DE awareness gap, or a lack of knowledge, awareness, or misconceptions about DE programs contributing to perceptions of DE inaccessibility, was evident across student and district-level characteristics. At the student level, results indicated that parents and friends had limited knowledge of DE programs which narrowed their influence on Latiné students’ perceptions. At the district level, the DE awareness gap contributed to Latiné students\u27 lack of access to financial information and misperceptions about who should participate in DE programs. Results also revealed that DE participants had a higher sense of belonging than non-DE participants, suggesting that Latiné students who feel they belong in their schools are more likely to pursue advanced coursework.To increase Latiné students\u27 awareness of DE programs and improve their perceptions of DE accessibility, district leaders should articulate an inclusive, equitable vision for these programs and align their practice to that vision. They should also provide professional development to broaden the understanding of who should have access. Additionally, at the school level, educators need to evaluate how they market and communicate about DE programs, dispel misconceptions, and ensure that all students have an opportunity to participate. Overall, this study underscored the importance of advancing equitable access for Latiné students in DE programs.</p
Orthogonal Time Frequency Space (OTFS) Modulation in Underwater Acoustic Communications
No full textThis thesis introduces a heuristic Doppler compensation algorithm to mitigate severe intra-frame Doppler effects observed in high-mobility acoustic communication environments where Orthogonal Time Frequency Space (OTFS) Modulation is applied. Packets experiencing the severe Doppler issue are challenging to decode even though the signal-to-noise ratio(SNR) is high. In this work, these challenging packets are first detected by analyzing the Doppler shifts over all subcarriers in the delay-time domain. Then the quality of each subcarrier is evaluated for further subcarrier selection. Instead of estimating the channel based on all subcarriers, this algorithm focuses on a few high-quality subcarriers and reconstructs the signal based only on the selected subcarriers. The effectiveness of this algorithm is demonstrated and evaluated by comparing the reproduced results from one previous work and the improved results with the algorithm, and it is observed that this algorithm has a more pronounced improvement in the time domain than in the delay-Doppler domain, with more packets achieving error-free decoding and a notable reduction in the average bit error rates.</p
Adopting an Ecological Approach to Misinformation: Understanding the Broader Scope and Impacts of Misinformation on Online Communities
No full textMisinformation plays a significant role in people\u27s lives. While numerous interventions are designed to address misinformation and its impacts, these interventions primarily focus on addressing individual pieces of false and misleading content. This dissertation argues that such individualistic focus on misinformation de-emphasizes and draws attention away from the broader scope and impacts of misinformation. Instead, it advocates for conceiving of misinformation as a broad societal phenomenon that transcends any isolated, individual pieces of false or misleading content. This perspective to misinformation, in particular, emphasizes the crucial role of community-oriented mechanisms in the broader scope and impacts of misinformation, which are often under-looked in individualistic approaches to misinformation. Therefore, to study the broader scope of misinformation, and to account for the community-oriented mechanisms involved in this phenomenon, this dissertation takes an ecological approach to misinformation. Specifically, in an experimental setting, it demonstrates how false and misleading content, and community responses to such content, together contribute to the way misinformation influences perceptions about social norms within online communities, underscoring the way misinformation impacts online communities beyond misleading their members about any individual pieces of content. To examine the interplay between false and misleading content, and community responses in authentic online interactions in an observational setting, it then embraces the concept of framing from sociological research, and demonstrates how these elements and the interplay between them together contribute to the shifts in the way people come to understand the world\u27s events, again impacting online communities beyond misleading their individual members about any individual pieces of content. The approach taken in this dissertation serves as an example that can inform how future research might adopt an ecological approach to misinformation, and further expand knowledge about the broader scope and impacts of this phenomenon. This dissertation concludes by outlining the implications that it offers for researchers to study misinformation\u27s broader scope, for community moderators to expand their moderation practices beyond individual content moderation, and for social media designers to leverage the role of community members in addressing the broader impacts of misinformation on how their communities run and evolve. </p
Efficient Machine Learning Model for Progressive Damage Analysis in Fiber-Reinforced Composites
No full textFiber-reinforced composites (FRC) are extensively utilized in the automotive and civil engineering sectors owing to their superior strength-to-weight ratio, corrosion resistance, and design versatility. The progression of composite modeling has shifted from analytical techniques to the creation of computationally efficient methods; semi-analytical strategies like the Generalized Method of Cells (GMC) and High-Fidelity Generalized Method of Cells (HFGMC) provide a compromise between precision and efficiency, although examining multiple scales during each phase of damage evolution continues to result in heightened computational demands. Surrogate machine learning (ML) models have gained increasing attention because they can forecast nonlinear behaviors in significantly reduced time compared to analytical, semi-analytical, and computational methods.Our objective is to develop a deep learning-based surrogate model capable of computing the mechanical properties of FRCs while preserving damage-pattern information throughout progressive damage evolution. The surrogate model comprises two sequential blocks—first an LSTM, followed by a CNN—trained and validated separately. By leveraging deep learning, the model forecasts nonlinear behaviors such as stress–strain relationships and failure strengths. The cumulative loss and accuracy of our predictions indicate that the LSTM and CNN models achieve over 97% accuracy in forecasting the subsequent damage pattern and the compromised homogenized stiffness matrix. The surrogate model is about three times faster than the high-fidelity HFGMC model. The close agreement between the surrogate and HFGMC curves highlights the surrogate model\u27s ability to accurately approximate complex progressive damage behavior, providing valuable insights for post-processing and further analysis. In conclusion, the integrated LSTM-CNN surrogate model advances both computational efficiency and interpretability, representing a significant step toward reliable, transparent, and accelerated progressive damage analysis methodologies for composite material systems.</p
Insights into Gram Negative Bacteria: Model Membrane and Secreted Vesicles
No full textAntibiotic resistance is increasingly rising, specifically with an increased focused on gram- negative pathogens, due to their additional outer membrane (OM) structure providing additional protection. The OM is the first line of defense, which antibiotic interacts with, making it critical to access this interaction. However, the diverse composition of the OM, including phospholipids, lipopolysaccharides, and proteins, creates challenges for accurately mimicking it. Additionally, gram-negative bacteria secrete smaller portions of their membrane to form outer membrane vesicles (OMVs), which add further complexity in membrane analysis and provide parent bacteria protection from antimicrobial peptides (AMP). This doctoral dissertation aims to contribute two novel approaches to mimic the bacterial membrane and use as a platform to investigate AMP interaction. Alongside, it also report a novel single OMV heterogeneity analysis method using a fluorescence microscope, to determine size-based protein/toxin discrepancies and investigates the key membrane properties that regulate OMV – AMP binding. Since OMVs bud directly from the membrane, they contain key bacterial membrane components making them ideal for membrane mimics. To create a molecularly complete planar bacterial membrane mimic using OMVs, we present a novel method that integrates synthetic lipids via freeze-thaw treatment. We demonstrate that we can successfully retain proteins and nucleic acids in our bilayer, making it ideal for antibiotic testing. Lastly, we report an additional method using chaotropic agent- assisted supported lipid bilayer using synthetic lipids, such as phosphatidylethanolamine, phosphatidylglycerol, and cardiolipin, which are primarily found in bacterial membranes at physiologically relevant concentrations. Overall, these techniques for bacterial membrane mimics and OMV analysis deepen our understanding of the OM and pave way for new therapeutic strategies.</p
Dissecting the Korean Nationalist Allegory of Masculinity: A Comparative Study of Crash Landing on You (사랑의 불시착) and My Demon (마이 데몬)
No full textThis dissertation explores the ways in which the K-dramas interact with the Korean national allegory and either reinforce or reimagine strict gender roles and hegemonic masculinity in connection to the military and men embodying the ideals of the nation. Focusing on the 2019 drama Crash Landing on You (사랑의 불시착) and the 2023 drama My Demon (마이 데몬)– this study examines how the male leads in each drama are impacted by the nation’s role in constructing masculinity through expectations of gender performativity. Drawing on postcolonial and feminist theory, this analysis considers how the North Korean and South Korean national allegory is constructed and how said allegory impacts the depictions of ‘hard’ and ‘soft’ masculinity in both dramas. This analysis highlights the difficulty that Crash Landing on You and My Demon have in breaking away from the Korean nationalist allegory of masculinity and reimagining new ways to depict masculinity. </p
Interstitial/ "in-between" space in Architectural Form
No full text{"value":"Several architectural practitioners saw interstitial/ “in-between” space in Late Modernism (approximately 1950-1970 A.D.) as a significant type of space in architectural form that could
solve both the limitations of Modern Architecture and reverse the spatial dissolution (i.e. the
collapse of the spaces of ‘city’ and country’ and the collapse of ‘interior’ and ‘exterior’) of the
built environment. Late Modern architectural practitioners believed the use, or ‘insertion,’ of
interstitial/ “in-between” space into architectural form could solve not only the problems that
Modern architecture created but could also re-affirm the spaces (i.e., re-establish ‘city’ and
‘country’ and ‘interior’ and ‘exterior’) of the built environment. Architectural practitioners believed
this because they recognized interstitial/ “in-between” space could both ‘mediate’ and ‘activate’
spaces, both within and outside architectural form, simultaneously. Yet, the ability of interstitial/
“in-between” space to ‘mediate’ and ‘activate’ spaces in (and outside) of architectural form was
not a new phenomenon, but rather, was a phenomenon that was seen in both previous styles of
architecture and suggested in earlier written texts by architectural practitioners and philosophers.
This research argues that interstitial/ “in-between” space as a space in (and outside) of
architectural form has a history that stretches back much beyond Late Modernism – a history which
is little acknowledged by architectural historians today. Texts, plans, diagrams, and photographs
are examined simultaneously in this research to show that the use/ ‘insertion’ of interstitial/ “in-between” space in architectural form not only has a deep and lengthy history but that interstitial/
“in-between” space in architectural form should be discussed by architectural practitioners and
historians alike as the space that has significantly ‘shaped’ and influenced architectural form
throughout history because of its unique abilities.
","attr0":"abstract"
Genome evolution of Saccharomyces species and interspecific hybrids
No full text{"value":"Adaptive evolution is shaped by both chance and constraint, with mutations arising stochastically but their fate determined by selection. In this thesis, we use experimental evolution in Saccharomyces yeast to investigate how genomes adapt to novel environments and resolve genetic conflicts. We combine long-term evolution experiments, phenotyping, and whole-genome sequencing to identify general principles of adaptation and to uncover specific mutations associated with both environmental pressures and interspecific hybridization. In Chapter 1, we review best practices for identifying targets of selection in laboratory evolution experiments. We discuss how parallel evolution and statistical frameworks can be leveraged to distinguish adaptive mutations from neutral hitchhikers, and how these principles apply to complex genomes such as hybrids.In Chapter 2, we evolved 48 diploid S. cerevisiae populations for 4,000 generations in galactose and observed greater fitness gains than in glucose-evolved controls. Whole-genome sequencing of 96 clones revealed that most adaptive mutations occurred outside the canonical GAL pathway, including recurrent mutations, CNVs, and aneuploidies—particularly Chromosome VIII amplification. Reconstruction of evolved alleles in SEC23 and IRA1 confirmed their specific benefit in galactose, highlighting regulatory and metabolic adaptation beyond the expected pathway.Subsequent chapters focus on the evolution of interspecific yeast hybrids. We evolved 320 independent hybrid populations over 5,000 generations. Populations showed rapid initial fitness gains, followed by diverse evolutionary trajectories. Whole-genome sequencing identified recurrently mutated genes, including HSP104, UPC2, PTR2, and BSC1, marking them as key targets of selection. In particular, HSP104, a central regulator of proteostasis, displayed consistent and unique patterns of missense mutations across hybrid backgrounds, suggesting compensatory adaptation to proteotoxic stress.Our work provides a framework for identifying adaptive mutations and genetic incompatibilities and highlights the power of experimental evolution to uncover the mechanisms that shape genome evolution following hybridization.
","attr0":"abstract"
Reading and writing a new script: Young Black and Latinx learners’ experiences with counternarratives
No full textDespite the growing racial and ethnic diversity within the K–12 school system, curricula remain overwhelmingly Eurocentric and fail to reflect the identities of the learners they serve. At the same time, media reinforce stereotypes that shape societal perceptions and self-concept regarding race, further marginalizing racially minoritized youth. Together, these two powerful socialization tools, curricula and media, shape a dominant narrative that influences how these youth are seen and how they see themselves. In response to this dominant narrative, this research employed a qualitative case study design to examine the experiences of elementary-age Black, Latinx, and multiethnic learners who saw people like them centered, that is, featured prominently and consistently, in school curricula. In the context of five-week critical media literacy pedagogy (CMLP) unit on storytelling, nine learners from an urban elementary school learned to challenge conventional stories and were empowered to share their own. This study used the theoretical framework of counternarratives, informed by Critical Race Theory (Solórzano & Yosso, 2002), to analyze data from learner interviews, observations, and learner-created artifacts. The findings revealed that while learners did not frame their narratives as direct responses to white supremacy or dominant racial ideologies, their stories functioned as counternarratives by emphasizing joy, familial connections, and aspirations, within spaces where they are routinely misrepresented or erased. These narratives represent a form of everyday resistance: affirmations of identity that subtly disrupt the dominant narrative by reclaiming representational space. As learners shaped and shared their narratives, they demonstrated the ability to critique conventional narratives, reframe how they and their communities are represented, and position themselves as knowledge producers. These findings underscore the transformative potential of integrating counternarratives into elementary curricula through CMLP, illuminating how young learners can interrogate dominant narratives, construct self-affirming stories, and redefine whose voices matter in education. Implications call for systemic shifts in K–12 curriculum design and teacher preparation, emphasizing pedagogical frameworks that amplify diverse voices and validate the experiences of historically marginalized learners.</p
A Retrospective Study on Risk Factors for Cesarean Delivery in the United States.
No full text{"value":"Abstract Caesarean sections (CS) are used in one-third of child births in the United States. While many CS
are medically necessary, their growing use, especially when elected by expectant mothers, poses
important implications regarding maternal risks, strain on the healthcare delivery system, and
healthcare costs. In the current study, a retrospective cross-sectional study was conducted using
3.1 million birth records in 2023 obtained from Centers for Disease Control and Prevention (CDC),
to assess primary drivers’ that influence mode of delivery (vaginal vs caesarean). Risk factors for
CS were divided into three distinct groups (predisposing, enabling, and need-related factors) using
the Andersen Behavioral model. Descriptive statistics, bivariate analysis, and logistics regression
were employed in this study.
Results indicated that women who were 35 years of age or older, with history of breech-presenting
babies, or had pregnancy complications such as gestational diabetes, hypertension, or eclampsia
were associated with a higher risk for CS. Additionally, black women with Medicaid were more
likely to have a Cesarean delivery. Encouraging prenatal education, tackling racial and
socioeconomic differences, advancing collaborative decision making and reviewing clinical
guidelines to support deliveries through the vagina are some of the specific strategies this study
suggests minimizing the unnecessary use of CS. Reducing the financial incentives that promote
unnecessary CS procedures may be a viable option through the reform of insurance reimbursement
structures.
While the study is limited by the variables included in the birth certificate data, which possibly
lacks exhaustive clinical and personal context, this study provides significant insights into
individuals or organizations who aspire to promote maternal health outcomes and contribute to
unbiased, evidence-based childbirth procedures throughout the U.S healthcare system. ","attr0":"abstract"