The National University System Repository exists to increase public access to research and other materials created by students and faculty of the affiliate institutions of National University System. Most items in the repository are open access, freely available to everyone.
Recent Submissions
Item Gender Differences in ADHD Diagnosis(2025-12)Attention Deficit Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder that impacts 3% of the population worldwide and approximately 11% of children. Most of the research on ADHD has been conducted on samples of predominantly males which has resulted in a lack of research and knowledge on the symptom presentation in females. The goal of this capstone is to address the gap in ADHD research on females, as well as analyze the differences in symptom presentation between males and females. This capstone looks at female ADHD diagnosis through the lens of attachment and family systems theories. A proposal for a group session for professionals that highlights the differences in symptoms between males and females is explained in chapter 3.Item Applications of Machine Learning Algorithms for Examining the Impact of COVID-19 on the Dropout Rate for High Schools in the State of Illinois(2025-12)High school dropout remains a persistent and pressing issue in the United States and globally. This quantitative, non-experimental study aimed to develop a predictive model for high school dropout rates in the state of Illinois before and during the COVID-19 pandemic, covering the years 2017–2019 and 2020–2022, respectively. Publicly available datasets from the Illinois State Board of Education were analyzed using multiple linear regression, random forest, and XGBoost models to assess the impact of the pandemic and to identify which school-level features most strongly predict dropout outcomes. The study applied the CRISP-DM framework and interpreted results through the lens of survival analysis theory to address the problem of academic attrition over time. Research questions and hypotheses were tested using the three predictive models. The analysis identified mobility rate, COVID-19 period, and low-income enrollment as the most influential predictors of high school dropout, with mobility rate emerging as the top signal across models. Model performance was evaluated using R², mean absolute error (MAE), and root mean squared error (RMSE). The XGBoost model offered the best balance of predictive accuracy and computational efficiency, making it the most effective and preferred model for this study. Recommendations for future research are grounded in the study’s predictive scope and methodological limitations. Proposed next steps include evaluating model performance over extended timeframes, incorporating post–COVID-19 data, and exploring additional demographic and school-level predictors using time-aware validation and stratified replication. These extensions will strengthen the generalizability and practical value of predictive modeling for dropout prevention in educational settings.Item Teachers’ Perceptions of the Effectiveness of Positive Behavior Intervention and Supports (PBIS) for Title 1 Schools: A Qualitative Phenomenological Study(2026-01)The disproportionate use of exclusionary discipline practices against African American students contributes significantly to the preschool-to-prison pipeline, highlighting the need for equitable alternatives. Positive Behavioral Interventions and Supports (PBIS) has been promoted as a framework for reducing exclusionary practices by emphasizing prevention, consistency, and equity. This qualitative phenomenological study explored teachers’ perceptions of PBIS implementation, its effectiveness in addressing disproportionate discipline, and the challenges that limit its fidelity. Semi-structured interviews with 8 preschool teachers from Title 1 schools were analyzed thematically using MAXQDA software and implementation science as a guiding framework. Findings revealed that teachers valued PBIS interventions and incentives as tools for prevention but noted inconsistent implementation across classrooms, insufficient resources and staffing, and a lack of culturally responsive professional development. Teachers also emphasized the ongoing disproportionate impact of exclusionary discipline on students of color, describing their roles as navigating bias, advocating for fairness and collaborating with colleagues to strengthen equity through PBIS. Implications for practice include the need for ongoing professional learning centered on equity, and collaborative engagement among educators. The study contributes to the literature by elevating teacher voices in understanding the complexities of PBIS implementation and by highlighting the conditions necessary for PBIS to fulfill its potential as a tool for equity in discipline and as a means of disrupting the preschool-to-prison pipeline.Item Exploring Perceptions and Experiences of Blockchain Implementation in Auditing: A Multiple Case Study(2025-10)Blockchain technologies are poised to reshape financial statement audits, yet authoritative guidance remains limited. The purpose of this qualitative multiple case study was to examine perceptions and experiences with established accounting standards regarding blockchain implementation in practice in the United States. Accounting theory and the extended technology acceptance model guided the study. Data were collected from 14 semi-structured interviews, an open-ended survey, and organizational documents from auditing firms. Reflexive thematic analysis and triangulation produced seven themes across three domains. Auditors reported limited real-world exposure, heavy reliance on legacy professional guidance, and implementation hurdles, including client education, tool limitations, and regulatory ambiguity. Participants anticipated efficiency gains from immutable, real-time ledgers, such as faster confirmations and more reliable provenance, and emphasized that transparent, codified standards are a prerequisite for mainstream adoption. Alignment with existing theoretical frameworks was uneven, often recognized conceptually but seldom operationalized, and current regulatory materials were viewed as helpful for framing risk but incomplete for method design. The study contributes practitioner-grounded evidence on blockchain’s auditability, highlighting immediate needs for targeted standards, competence development, and carefully scoped pilots on permissioned networks. These findings inform standard setters, firms, and educators seeking to integrate distributed-ledger evidence without compromising assurance quality.Item Balancing Data Accessibility and Privacy: Machine Learning Approach to PII Detection in Electronic Health Records(2026-01)This constructive research study examined the development of a scalable, context-aware machine learning (ML) framework for detecting personally identifiable information (PII) in unstructured electronic health records (EHRs). The research problem addressed the absence of reproducible, data-driven methods capable of balancing privacy preservation and data accessibility while maintaining compliance with legal frameworks such as the Health Insurance Portability and Accountability Act (HIPAA) and the General Data Protection Regulation (GDPR). The study focused on healthcare organizations and researchers who face challenges protecting sensitive health data while facilitating secure data sharing for clinical and analytical purposes. The study's purpose was to construct, implement, and evaluate a privacy-preserving artifact guided by the Cross-Industry Standard Process for Data Mining (CRISP-DM) framework. The research design integrated natural language processing (NLP) with unsupervised and hybrid ML algorithms, including term frequency–inverse document frequency (TF-IDF) vectorization, singular value decomposition (SVD), and density-based spatial clustering of applications with noise (DBSCAN). A transformer-based named entity recognition (NER) module utilizing Bidirectional Encoder Representations from Transformers (BERT) to validate clustering outputs. The research data were obtained from the Medical Information Mart for Intensive Care (MIMIC-III) database, a publicly available and de-identified dataset licensed through PhysioNet (Johnson et al., 2016).The experimental code and replication scripts are available at: https://github.com/NU-Academics/PII-Detection or Bert & Regular_Expression PII Detection - Colab. The model was trained and evaluated in Google Colab using BigQuery integration to ensure compliance with PhysioNet's data-use requirements. Empirical results showed that at a sample size of 5,000 records, the model achieved a precision of 0.955 and a recall of 0.466. When scaled to 10,000 records, precision remained high at 0.854, while recall improved to 0.580. Clustering validity indices confirmed coherent separation between PII-dense and non-PII clusters (silhouette coefficient ≈ 0.38–0.45; Davies–Bouldin Index ≈ 0.95–0.99). Approximately 61 percent of the records were labeled as noise, indicating that the model effectively isolated high-risk text regions while minimizing false positives. The study concluded that unsupervised NLP methods can reliably identify latent PII patterns within de-identified clinical narratives, achieving performance comparable to that of supervised models with lower computational costs. These findings demonstrate that scalable ML frameworks can reconcile the privacy–utility balance in EHR analytics. The research recommends incorporating hybrid explainable AI components, such as SHAP and LIME, to improve interpretability and extend future validation to institutionally governed datasets containing unredacted identifiers under Institutional Review Board (IRB) oversight.
Communities in National University System Repository
Select a community to browse its collections.