Digital Commons @ Florida Atlantic
Not a member yet
400 research outputs found
Sort by
DIGITAL BEGINNINGS: EXPLORING THE IMPACT OF SCREEN EXPOSURE ON INTERNALIZING BEHAVIORS DURING TODDLERHOOD
No full textThe present study examines the relationship between screen exposure behaviors and toddler internalizing behaviors, with sleep behaviors being observed as a moderator. The study emphasizes the need for continued research to adapt to the ever-changing nature of the daily use of screens during toddlerhood and early childhood. Fifty-seven mothers of two-year-old children participated in one-on-one parent-report only, interviewer-based quantitative interviews discussing their child’s internalizing and sleep behaviors. Participants were also asked to fill out a questionnaire about their child’s screen exposure behaviors. It was hypothesized that (1) higher levels of screen exposure behaviors will result in increased internalizing behaviors, (2) sleep behaviors will moderate the relationship between screen and internalizing behaviors, and (3) parent-child interactivity during screen viewing will influence the presence of internalizing behaviors. Screen behaviors did not significantly predict internalizing behavior, yet the analyses showed a trend towards increased internalizing behavior levels as the child’s screen behaviors increased. Findings also demonstrated that sleep is a moderator, showing that as sleep problems increase, it decreases the relationship between screen and internalizing behaviors. Lastly, interactivity levels and internalizing behaviors were not related. The current study demonstrates the complexities of the constantly developing nature of screens’ influence on early childhood
FROM AWARENESS TO ACTION: IMPROVING ADVANCED DIRECTIVES WITH THE FIVE WISHES IN COLLEGE AGED YOUNG ADULTS
No full textThis quality improvement project aimed to increase early education, initiation, documentation, and completion rates of advance directives (ADs), specifically the Five Wishes, among college-aged young adults (18–24 years old) in a concierge primary care setting. The project addressed the low engagement and completion rates of ADs in this population, which can result in inadequate end-of-life care planning. The initiative involved training all staff members through collaborative instruction, informational sessions, and personalized provider-patient discussions. A convenience sample of eligible patients was identified via chart review. Inclusion criteria included English-speaking young adults aged 18–24, regardless of comorbid conditions who were members of the practice.
Statistical analysis included descriptive methods using tables, charts, and graphs to visualize trends. Pre- and post-surveys using a six-question Likert scale assessed provider knowledge, comfort, and confidence. Aggregate mean values were compared to evaluate changes, and each question was analyzed individually. A dependent sample t-test was used to determine statistically significant differences between pre- and post-intervention scores. Additionally, Wilcoxon signed-rank tests were applied to assess non-parametric data, reporting test statistics, p-values, and normality p-values.
Implications for practice include improved provider communication strategies tailored to young adults, increased AD documentation rates, and enhanced trust in patient-provider relationships. The project supports a culture of proactive healthcare planning, empowering young adults to make informed decisions. It also equips providers with tools and training to initiate meaningful conversations, fostering continuity and personalization in care
HEARTMATH: A STRESS REDUCTION PROGRAM FOR HEALTHCARE PROVIDERS IN AN ADDICTION TREATMENT SETTING
No full textWork-related stress significantly impacts the well-being of healthcare providers, specifically nurses and nurse practitioners, in addiction treatment facilities. This Quality Improvement (QI) project implemented the HeartMath Quick Coherence Technique to reduce stress and enhance resilience among staff, thereby improving the well-being of healthcare providers and the quality of care. The project utilized a pre- and post-intervention design with five participants, two registered nurses, and three nurse practitioners. Participants received training on HeartMath techniques, the EmWave Pro sensor, and the InnerBalance smartphone application. Over a six-week period, participants practiced Quick Coherence meditation for five minutes, twice daily, to achieve heart rate variability (HRV) coherence. Outcome measures included the Perceived Stress Scale (PSS-10) and average HRV coherence rates. Comparing baseline data Week 1 to Week 6, participants demonstrated a 24.5% improvement in average HRV coherence. Additionally, pre- and post-intervention surveys revealed an 11% decrease in perceived stress scores. Results indicated that HeartMath meditation techniques are an effective, low-cost intervention for stress reduction. Organizational prioritization of such brief, scalable interventions is essential for fostering a healthy work environment and supporting staff well-being
PERSONS WITH ALZHEIMER’S DISEASE AND RELATED DEMENTIAS AND CAREGIVER BURNOUT: A QUALITY IMPROVEMENT PROJECT
No full textAlzheimer\u27s Disease and Related Dementias (ADRD) represented an escalating public health challenge, profoundly affecting individuals, families, and healthcare systems. As the number of individuals diagnosed with ADRD rises, the prevalence of caregiver burden and burnout among those providing care also increases. The purpose of this quality improvement project was to evaluate the existing assessment processes for caregiver burnout and determine the effectiveness of the Zarit Burden Interview (ZBI), a measurement which identifies caregivers’ levels of stress, at a memory disorder center in south Florida. Sixteen caregivers reported overall mild-to-moderate caregiver burden, with a mean ZBI score of 22.19 (SD = 15.06). While most item-level means were below 1.5, indicating infrequent burden experiences, several items demonstrated higher scores suggestive of more consistent strain (behavior management and perceived loss of autonomy). The outcomes of this quality improvement project contributed to the improved identification of caregiver stress/burnout, connection of resources, and strengthened protocols for comprehensive dementia care
CONFLICTING BASES FOR JUDGING ACTION AND ACTORS: THE ROLE OF INDIVIDUAL DIFFERENCES
No full textThis study examined how political orientation, locus of control, and action identification predict moral reasoning across six dilemmas contrasting consistency and compensation. Participants (N ≈ 245) completed measures of ideological beliefs, perceived control, and action construal. Reliability analyses indicated modest internal consistency for Action Identification (α = .53) and low reliability for Locus of Control (α = .32). The six moral scenarios showed very low reliability (α = .11), indicating they captured diverse judgment domains. Binary logistic regressions revealed that political orientation, particularly its social dimension—was the strongest and most context-dependent predictor of moral choice. Economic conservatism predicted compensatory preferences in college admissions, whereas social conservatism showed opposing effects across dilemmas. Action identification and locus of control demonstrated minimal predictive power. Findings suggest that moral judgment is context-dependent and shaped primarily by ideological rather than cognitive or control-related dispositions
ROBUST ANOMALY DETECTION UNDER DATA IMBALANCE, NOISE, AND LABEL SCARCITY
No full textIn today’s data-driven landscape, large volumes of data are generated continuously, often containing imperfections such as noise, missing data, or unreliable labeling. These real-world datasets are typically high-dimensional, sparsely labeled, and imbalanced, creating substantial challenges for both supervised and unsupervised learning. These challenges are especially prevalent in the task of anomaly detection, where instances belonging to the class of interest are rare and underrepresented compared to normal instances. This dissertation proposes and evaluates robust frameworks for anomaly detection that address these data challenges and improve model performance and robustness using real-world datasets, including credit card transactions and cognitive assessments. Supervised learning requires labeled data which can be costly, hard to produce, and prone to mislabeling. We propose a reconstruction error–based method to identify and correct mislabeled samples, thereby improving the quality of labeled data. To address imbalance and high dimensionality, we combine deep feature extraction using convolutional autoencoders, an unsupervised learning technique, with class rebalancing strategies to improve classification performance. Then, we examine how the order of preprocessing steps affects downstream ensemble learners. For unlabeled data, we propose a novel hybrid unsupervised framework that integrates convolutional autoencoders for representation learning with Isolation Forest for anomaly detection (CAE-IF). CAE–IF demonstrates robust performance on unlabeled, high-dimensional, and imbalanced data across cognitive and fraud detection domains, relative to common baselines such as Isolation Forest and Local Outlier Factor. In addition, we apply an instance-based iterative cleaning method that uses reconstruction error to remove likely outliers and improves representation quality for downstream detection without requiring manual annotation. The results demonstrate that our proposed approaches improve model robustness in various imperfect data conditions. Collectively, these contributions provide a practical and generalizable toolkit for anomaly detection, addressing the core challenges of class imbalance, label noise, and label scarcity across both supervised and unsupervised settings
PROMOTING TUTEES’ SELF-EFFICACY, TUTOR-TUTEE RELATIONSHIP, AND MATHEMATICAL KNOWLEDGE/PERFORMANCE
No full textThis action research study investigates how I, as a tutor, implemented affective instructional practices to support early adolescent tutees’ self-efficacy, tutor-tutee relationship, and mathematical knowledge/performance. Data were collected from multiple sources, including journal entries, video recordings, semi-structured interviews, report card evaluations, and Florida Assessment of Student Thinking (FAST) scores. These sources were used to gain comprehensive insight into how my affective instructional strategies were embedded within mathematics tutoring sessions with five early adolescent tutees between the ages of 10 and 12. A constant comparative analysis was employed to examine the data, with careful attention given to the four dimensions of trustworthiness—credibility, transferability, dependability, and confirmability—to ensure rigor and validity. Findings revealed that affective instructional practices supported tutees’ self-efficacy, mathematical performance, and the tutor-tutee relationship through three key approaches: 1) intentionally cultivating opportunities for competency—such as simplification, scaffolded questioning, repetition, prompting, success-oriented inquiry, encouraging explanations, and role-reversal—may support early adolescent tutees’ self-efficacy, while concurrently promoting tutees’ mathematic knowledge/performance, 2) incorporating games into tutoring sessions may support the development of the tutor-tutee relationship while simultaneously enhancing early adolescent tutees’ mathematic knowledge/performance, and 3) check-ins (i.e., emotional, emotional-academic, and academic check-ins) may assist in seamlessly transitioning between fostering the tutor-tutee relationship and supporting early adolescent tutees’ self-efficacy and mathematic knowledge/performance. This study contributes to the field by shining light onto a range of affective dimensions and instructional practices that have been less explored within tutoring contexts. The findings offer meaningful implications for tutors and educators, particularly for those who provide individualized or small group instruction and aim to create learning experiences that are both emotionally supportive and academically beneficial for tutees
NAMING ABILITIES IN AGING MONOLINGUALS AND BILINGUALS: LONGITUDINAL ANALYSES OF THE MULTILINGUAL NAMING TEST
Full text linkIndividuals who can speak more than one language may score lower than monolinguals on verbal naming tasks. The Multilingual Naming Task (MINT) is a 32-item assessment that is used at the Alzheimer’s Disease Research Center in the USA. This study looks at the impact of monolingualism and bilingualism on language abilities over three years in three clinical groups (cognitively normal, mild cognitive impairment, and dementia), using the MINT as a measure. The results show that bilingual individuals perform worse on the MINT than monolinguals at each time point, after controlling for demographic factors; however, there is no difference between language groups and decline of MINT scores. These results indicate bilingualism negatively affects naming performance on verbal tasks; however, since the MINT can be biased, that could also be the case here. Further research is needed to determine if bilingualism has an increased adverse impact on verbal task abilities with age and if the MINT is biased against bilingual individuals
MINERALIZATION, CHARACTERIZATION AND SELECTIVE DEGRADATION OF LIGNOCELLULOSIC BIOCOMPOSITE MATERIALS
No full textDeveloping novel functional materials, chemicals and biofuels from renewable plant sources has become a main area of focus within material science. The biosphere contains naturally occurring polymers that can be harvested to this end. Lignocellulosic materials are an important class of plant biomass consisting of three main components, cellulose, hemicellulose, and lignin. Wood has an inherent 3D hierarchical structure ranging from the nano- to bulk levels of organization, thereby achieving excellent mechanical properties while maintaining high porosity and low specific density. It carries significant promise for the production of chemicals, refinement of biofuels, and the development of novel functional materials. As lignocellulose is a complex anisotropic hierarchical structure differing distributions in functional chemicals or materials deposited within the scaffolding will result in a significant variance and impact in the properties and function of the final obtained composite. In this work, cell wall specific reinforcement was achieved through mineralization and confinement of nanocrystalline ferrihydrite (Fh) within the secondary wood cell wall. The distribution and mineralogy were characterized by an array of imaging and multiscale mechanical testing methods to accurately assess material property changes across all levels of organization. The functionality of the resulting biocomposites was assessed in a pilot study for the remediation of arsenate-contaminated groundwaters that proved effective under static and flow-through conditions. Chemical and enzymatic degradation treatments selectively targeting one of the three main wood polymers (cellulose, lignin) were utilized to investigate the degradation pathways and mechanical behavior of common chemical reactions used for biorefinery and chemical pulping. Overall, the studies detailed within this thesis contribute to a fundamental understanding of the structure-property-function relationship of chemically modified lignocellulose as a robust, renewable source of novel materials and biofuels