Scholar Commons - Institutional Repository of the University of South Carolina
Not a member yet
34867 research outputs found
Sort by
Assessing Health and Academic Disparities in the Deep South: The Enduring Impact of Segregation and Redlining in Georgia
No full textBackground: Within the American South, marginalized Black populations continue to suffer from health disparities. A key resource that could aid in the prevention of health disparities such as teen pregnancy is the school nurse. School nurses are well positioned to assist in the development of independent health maintenance among student populations. However, school nurses remain understaffed and underfunded throughout most of the United States, which limits their effectiveness. It is possible that within southern states there are inequities in how school resources are allocated that have ties to segregation and redlining practices that systemically marginalized Black populations throughout the 1900s. The purpose of this project was to understand disparities in health and academic outcomes across the state of Georgia in the context of segregation/redlining to understand if adequate staffing of school nurses could be a path forward for promoting health/academic equity and prevention of teen pregnancy.
Methods: Secondary data were collected from the Georgia Department of Education (GADOE) and the Georgia Department of Public Health (GADPH) to analyze school nurse staffing, educational quality indicators, and teen pregnancy rates. In addition, data were collected from the University of Richmond’s Mapping Inequality to capture redlining/segregation in Atlanta, as well as the Social Vulnerability Index (SVI) from the Centers for Disease Control and Rural-Urban Continuum Codes (RUCC) from the United States Department of Agriculture. Descriptive statistics, linear regression, spatial autocorrelation, and Getis Ord Gi* mapping were used to analyze data.
Results: The analysis revealed a statistically significant relationship between school nurse staffing levels and academic outcomes across the state of Georgia. Geospatial analysis revealed that patterns of nurse staffing and academic outcomes may follow histories of segregation and redlining. Analysis of redlining in the city of Atlanta revealed that statistically significant clusters of teen pregnancy are found in redlined neighborhoods and these clusters follow statistically significant clusters of low school quality measures.
Discussion: This analysis indicates that the school nurse may be a pathway forward for policymakers and public health interventionists dedicated to breaking cycles of marginalization and systemic racism in marginalized communities
Heritage and Second Language Russian: Grammar Acquisition of Specificity
No full textGenerative theories of language acquisition have been traditionally applied to second language (L2) learners, while heritage language (HL) speakers’ linguistic intuitions are currently under-researched, especially in less commonly taught languages like Russian. Extending the predictions of generative L2 acquisition onto HL research, this dissertation investigates the processing of specificity by L2 and HL Russian users. Within the epistemic account of specificity, also known as speaker intent to refer (Fodor & Sag, 1982), the speaker assumes the existence of an entity in the actual world and can identify this entity. In Russian, specificity is expressed through various morpho-syntactic structures. Previous research on specificity in Russian has relied on offline methods focusing on metalinguistic knowledge (Cho & Slabakova, 2015); no research has applied online methods that explore real-time processing of [+/−specific] structures.
Addressing this research gap, the current study compares the processing of [+/−specific] structures in Russian by English-dominant L2 learners of Russian (N=46), HL speakers of Russian (N=39), and Russian monolingual controls (N=24). Experiment 1 focuses on the processing of 1) nominal and adjectival modifiers denoting possession, 2) accusative/genitive case alternations under negation, 3) the numerals odin ‘one’ and xot’ odin ‘at least one,’ and 4) the particles -to and -nibud’ on indefinite pronouns, all functioning as a direct object in a post-verbal position. Experiment 2 aims to tear apart the effect of syntactic position (sentence-initial vs. post-verbal) and function (a subject vs. a direct object) on the processing of 1) nominal and adjectival modifiers denoting possession, and 2) the particles -to and -nibud’. In both experiments, online data were collected using a self-paced reading task. As participants read one phrase at a time, each experimental item included personal pronouns referential to either [+specific] structures (a grammatical context) or [−specific] structures (an ungrammatical context). Reaction times to the personal pronoun and the phrase following it were measured, with the assumption that ungrammatical [−specific] contexts should cause processing delays.
Linear mixed models were run to investigate the effects of language speaker group, language proficiency, and type of structure in Experiment 1. Additionally, for Experiment 2, linear mixed models explored the effects of syntactic position and function. Results revealed that L1 Russian speakers were sensitive to [+/−specific] values of each morpho-syntactic structure (p \u3c .05). However, specificity manipulations did not impact L2 processing, regardless of the structure. HL speakers demonstrated some processing trends that matched native speaker intuitions, only in certain types of structures. However, these trends were never statistically significant. Results also found that the structure’s function did not significantly affect processing, whereas its syntactic position was significant, and a longer surface distance between the target structure and the referential pronoun caused more processing delays. Finally, the results did not reveal any significant effect of proficiency level on L2/HL linguistic processing
Functional Data Analysis on Life Expectancy and Healthcare Expenditure
No full textThis study analyzes the life expectancy for 237 countries from 1950-2023 and the healthcare expenditure for 50 countries from 1970-2022, and how life expectancy and healthcare expenditure relate to each other. Functional Principal Components Analysis was used to analyze the life expectancy and healthcare expenditure for each of the countries. Additionally, the regions of the countries were analyzed to identify any regional trends for the life expectancy data. Due to missing data and the structure of the healthcare data, multiple imputation methods and Principal Component Analysis techniques were explored for the healthcare data. Furthermore, a simulation study was conducted to assess the accuracy of these methods. Additionally, Canonical Correlation Analysis was used to understand the relationship between life expectancy and healthcare expenditure. To better understand the variability of the analysis and stability of the Principal Component loading and eigenfunctions, a boot-strapping simulation was conducted to see how repeatable the results are
Ideologies of Language, Race, and Time in U.S. Political Discourses
No full textThis dissertation follows how public discourses that emerged during the 2020- 2022 “Critical Race Theory Panic” played out in different contexts and at different scales in the United States, from conservative national media to South Carolina legislative action. While political discourse has become increasingly nationalized (Berry & Sobieraj, 2014), new legislation has largely taken place at the state level (Johnson, 2022). Drawing on over two years of fieldwork and methods of discourse analysis, I thus examine how nationally circulating discourses about race and racism were taken up by South Carolinians to create partisan alignments and distinctions as well as to further concrete legislative objectives. Focusing on the seven months in 2021 when critical race theory (CRT) received the most national media coverage, I first show how speakers on the conservative talk show Tucker Carlson Tonight discursively constructed a version of American history that suggested that racism had been solved and that CRT reintroduced racism into the public consciousness. Then, I look to the ways that the debate was taken up in the South Carolina Legislature and among politically active South Carolinians in 2022. Across these contexts, I analyze three discursive strategies that conservatives used to establish and rationalize their political position that CRT had infiltrated public life and must be banned from schools. In particular, I show how partisans, in arguing their positions, imagined historical narratives of racial progress and decline and drew on commonsense assumptions about how language works, or “semiotic ideologies” (Keane, 2018), to support these racial chronotopes (Koven, 2013; Rosa, 2016) that implicitly vii answered questions such as, when was America racist, where was America racist, when did America stop being racist, what did the Civil Rights Movement accomplish, and how has racism changed over time? Several important patterns emerged in my analysis that support an overarching observation: I found that while partisans on both the left and the right constructed nuanced language ideologies and chronotopic imaginaries to justify not only their political positions but also their ostensible worldviews, these ideologies were flexible and strategically applied, despite longstanding rhetoric that positions conservatism as backward-looking (Abramowitz, 2018; Robin, 2018; Stanley, 2024; Stern, 2019). As I demonstrate, a fiercely maintained language ideology in one discourse event could be utterly rejected in the next, and in one context, conservatives constructed themselves as future-focused—even linguistically and socially “progressive”—while in another they argue for a return to an idealized past. I ultimately argue that conservatives’ orientations to ideologies of language, race, and history were hardly reflective of a fixed set of beliefs about the world but discursive strategies that served their political and moral interests (Irvine, 1989)
Emotional Landscapes: St. Teresa of Avila and the Spiritual Journey
No full textThis paper examines the writings of St. Teresa of Avila (1515-1582) through the methodological lens of the history of emotions, arguing that her mystical accounts represent a crucial yet understudied contribution to our understanding of early modern emotional frameworks. Through close textual analysis of Teresa\u27s major works, this study demonstrates how her detailed articulations of spiritual ecstasy, divine love, and religious melancholy offer unique insights into both the religious and secular emotional landscapes of Counter-Reformation Spain. Teresa\u27s innovative emotional vocabulary, which combined traditional Catholic devotional language with embodied metaphors and spatial conceptualizations, allowed her to navigate the complex ecclesiastical politics of her era while simultaneously developing sophisticated taxonomies of affective states that challenged contemporary understandings of emotional experience. This research contributes to ongoing scholarly conversations about gendered dimensions of religious emotions, the relationship between embodiment and spirituality in mystical discourse, and the historical emergence of emotional introspection as a form of self-knowledge. By positioning Teresa\u27s work as a significant case study within the history of emotions, this paper illuminates how religious experience served as a crucial site for emotional innovation, negotiation, and expression in early modern European culture
In Their Own Words: How Standardization Shapes Student Learning Experiences
No full textThis qualitative action research study explores how students perceive and experience standardized and elective educational environments in a rural, Title I high school in the U.S. South. Through semi-structured interviews and focus group discussions with seven diverse participants, the study investigates student-defined learning objectives, motivation and engagement across different classes, and youth perceptions of valuable learning experiences. The research is grounded in constructivism, applying feminist theory and critical race theory as lenses to challenge power dynamics and promote equality in education.
Findings confirm the problem of practice that inspired the study by revealing a contrast between students’ experiences in standardized core classes versus more flexible electives. Participants consistently reported diminished engagement and increased anxiety in standardized learning environments, while expressing a strong desire for practical, real-world skill development. The study highlights the critical role of student autonomy, supportive learning environments, and the impact of standardization on student motivation and mental health. These insights underscore the need for a reevaluation of current educational approaches to better serve diverse student needs.
This research contributes to the ongoing conversation about standardized education by amplifying student voices and providing a platform for their counternarratives. The study’s implications extend beyond the local context, offering insights for educators, policymakers, and researchers seeking to create more equitable and effective learning environments. By documenting the problematic impacts of standardized education through students’ perspectives, this study paves the way for meaningful educational reform that honors both institutional requirements and student needs
Episode 93: Reconstruction at USC: Christian Anderson
No full textThe Reconstruction Era after America\u27s Civil War brought about big changes as the former Confederate states were readmitted to the United States. Education professor and director of the Museum of Education Christian Anderson studies the history of higher education and is leading efforts to remember the changes that took place at USC 150 years ago during Reconsruction.https://scholarcommons.sc.edu/rememberingthedays/1094/thumbnail.jp
Deep Learning Methods for Complex Survival Models in Electronic Health Record
No full textElectronic Health Record (EHR) data from millions of patients are now routinely collected across diverse healthcare institutions, offering unprecedented opportunities to enhance personalized medicine and improve healthcare quality. EHR data contain rich information, including patient demographics, diagnoses, laboratory test results, medication prescriptions, medical images, and clinical notes. However, analyzing EHR data accurately is challenging due to its large scale, high dimensionality, and inherent complexity. Traditional statistical models often rely on strong assumptions and lack the flexibility needed to accommodate complex, nonlinear relationships among variables. In this dissertation, we propose incorporating deep learning methods to improve risk prediction functions within different types of survival models, designed specifically to handle the high-dimensional, nonlinear associations present in EHR data. The three projects presented in this dissertation are motivated by statewide COVID-19 infection-related and vaccine data from South Carolina.
For patients vaccinated against COVID-19, some individuals may never experience a breakthrough infection, either due to effective vaccine-induced immunity or natural immunity. In the first project, we develop a predictive framework using a mixture cure model (MCM) to handle diseases with a potentially cured subpopulation. The MCM has two components: an incidence component, modeled using logistic regression, which estimates the likelihood of experiencing a breakthrough infection, and a latency component, modeled with Cox proportional hazards (Cox PH) regression, which estimates the timing of breakthrough events if they occur. Deep learning methods are applied to capture the nonlinear risk functions within both incidence and latency components, and the Expectation-Maximization (EM) algorithm is employed to estimate model parameters efficiently.
The second project focuses on using deep learning to predict the nonlinear risk function within the Generalized Odds Rate (GOR) model, an alternative to the PH model that flexibly accommodates non-proportional hazards. This flexibility makes the GOR model particularly suitable for dynamic health data where hazard ratios may vary over time. To further enhance model simplicity and account for unobserved heterogeneity, a gamma-distributed frailty term is incorporated into the GOR model. The EM algorithm is used here as well, aiding in the estimation of latent variables within the frailty-adjusted framework. In the third project, I extend the methodologies developed in the first two projects by combining the MCM and GOR approaches into a GOR mixture cure model. This combined model utilizes deep learning to predict risk functions for both the incidence and latency components, allowing for complex, nonlinear associations across different subpopulations. The EM algorithm is proposed to estimate the parameters of both components.
All three proposed methods are evaluated through extensive simulation studies, demonstrating their robustness and flexibility. To prevent overfitting, an early-stopping criterion is implemented to identify an optimal stopping point during model training. Finally, the methods are applied to statewide COVID-19 infection and vaccination data, highlighting their practical utility and potential impact on real-world healthcare applications
Episode 103: Full Circle: Jim Bowers returns to USC\u27s School of Law
No full textJim Bowers was among the second small cohort of Black students to desegregate the University of South Carolina in the early 1960s, and he would later become the first Black professor in the university\u27s School of Law. More than 50 years later, Bowers has returned to the law school with a substantial gift to improve the institution where he served as a trailblazer.https://scholarcommons.sc.edu/rememberingthedays/1104/thumbnail.jp
Heuristics and Cultural Competition: How the Status of Cultural Objects Are Formed and Status Influences Competition
No full textPart of the human experience is the creation and consumption of culture. Because of this, over the course of evolution humans adapted cognitive resources for the identification and development of cultural objects in social environments. These objects are so essential to human interactions that they are integrated into social structures and assumptions are made about individuals that possess them. For example, an individual’s cultural consumption habits are used to draw conclusions about their education, merit, or fundamental worth and status. Utilizing music as a cultural object, I argue that individuals use the concept of Status as a cognitive shortcut during interaction, compressing complex information about music into a single dimension. I also propose that Status’s function as a cognitive shortcut has consequences for the success of music genres in terms of gaining consumers and maintaining investment in its consumption and cultural relevance. This dissertation utilizes a series of 3 studies to test 2 theories and a theory expansion that explore how Status is tied to music genres, and how Status shapes the consumption of music genres by individuals of different races, ages, levels of education, and political ideology.
This dissertation also utilizes two datasets, the Survey for Public Participation in the Arts and the Cultural Objects – Music survey, a dataset that is part of a wider series of Cultural Objects datasets. This dissertation also utilizes the Hybrid Blau Space model, novel simulation method that joins together agent-based and ecological models. Results from these studies provide support for the argument that Status acts as a form of compressed information about music, and that differences in race, age, education, and political ideology influence how individuals perceive the status and value of music genres. Finally, support is found for the argument that Status increases the competitive success of music genres in terms of gaining consumers and maintaining investment