35180 research outputs found
Sort by
A SYSTEMATIC REVIEW OF EVIDENCE-BASED PRACTICES AND RESOURCES FOR INCREASING SCHOOL ATTENDANCE
Regular school attendance is crucial for academic success and students' overall well-being. However, school absenteeism remains a United States nationwide challenge, with a significant number of students missing several school days each year. School absenteeism, with its deep and far-reaching consequences on students and society, in terms of economic, social, and health impacts, continues to captivate the attention of scholars and education stakeholders in a bid to understand the underlying factors and develop interventions that may keep students on track for school completion and other positive outcomes. This systematic review identifies and evaluates research-based evidence to increase school attendance. A total of 41 studies were reviewed and synthesized. The results show that there exist various interventions that may improve school attendance. Strongest evidence was found for parental involvement, mentoring, and a combination of multiple interventions. Implications for practice include strategies for expending resources, as it is possible to have cost-efficient, effective programs
UNCERTAINTY-AWARE MACHINE LEARNING TECHNIQUES FOR SUSTAINABLE MALWARE DETECTION
Malware remains a significant global cybersecurity threat, with millions of new variants appearing rapidly, highlighting the urgent need for effective automated detection methods. Traditional malware detection systems face challenges in adapting to the rapidly evolving landscape, particularly due to concept drift, where malware characteristics change over time, leading to performance degradation. This thesis introduces an innovative framework for uncertainty-aware malware detection using Gaussian processes (GPs) to quantify prediction uncertainty and enhance the reliability of malware detection.Our approach integrates GPs into machine learning models to address the limitations of existing malware detection techniques. By providing a probabilistic measure of confidence in predictions, it enables selective abstention from highly uncertain classification decisions, reducing false positives and negatives. We applied this framework to two qualitatively different mobile malware datasets.Through comprehensive evaluations, we demonstrate that our GP-based models significantly outperform traditional methods in accuracy and adaptability to evolving malware inAndroid. Extensive ablation studies validated our hypothesis on the effectiveness of uncertainty quantification in improving the overall detection performance. Our results provide valuable insights into malware behavior dynamics and the importance of adaptive detection strategies.Overall, this research contributes significantly to the development of robust and adaptable mobile malware detection solutions by integrating uncertainty quantification into machine learning-based detection systems. This study underscores the potential for GPs to enhance long-term performance in real-world applications, addressing the critical need for reliable detection mechanisms in an increasingly complex cyberthreat landscape
FROM PARADOX TO EQUITY AN INTEGRATED ECOSOCIAL AND INTERSECTIONAL STUDY OF ALCOHOL-RELATED HARM
Background: Health inequities, shaped by socio-political determinants, challenge public health research. The Alcohol Harm Paradox (AHP) exemplifies such inequities, where groups with lower socioeconomic positions (SEP) exhibit higher alcohol-related harms despite similar or lower consumption compared to those with higher SEP. Research often focuses on the biomedical paradigm, sidelining broader socio-political influences.Objective: This study uses the All of Us research program and an integrated Ecosocial and Intersectionality framework to scrutinize the AHP. It aimed to elucidate how social identities (e.g., race, ethnicity, gender, sexuality, ability) and societal structures (e.g., healthcare access, neighborhood quality, socioeconomic position) contribute to alcohol-related harm inequities.Methods: Using the All of Us dataset, the study probed the relationship between SEP and alcohol-related harm, evaluating moderating effects of structural contributors and social identities linked with oppression.Hypotheses: It was anticipated that (1) groups with lower SEP will report heightened alcohol-related harms, (2) protective structural factors will mitigate these disparities, (3) compounded risks will manifest for oppressed social categories within groups with lower SEP, and (4) structural factors' influences on alcohol harms will differ across social categories in both SEP groups.Conclusion: This study offers insights into factors impacting alcohol-related harm, expanding from a biomedical view to a broader socio-political perspective. The study found that while socioeconomic position did not significantly predict alcohol-related harm, higher alcohol consumption, healthcare access, and gender were significant predictors, highlighting complex influences on alcohol-related harm within structural and social contexts
Do Visuals Matter? Analyzing the Readability of Tabulated Financial Data in the Annual Report
This study leverages information manipulation theory to investigate the association between decreased readability (obfuscation) of tabulated financial data within the annual report and deception. Using a sample of over 9,000 firm-year observations, I leverage artificial intelligence to automatically classify financial tables within the 10-K and a novel method of measuring visual readability to determine if accruals earnings management and real-activities management are significantly associated with the readability of tabulated financial data. Results demonstrate that firms with higher levels of accruals management and that meet or just beat prior year earnings tend to have tabulated financial data within the 10-K, and in particular the audited financial statements, that are less readable. Results also provide evidence that firms with higher levels of real-activities management tend to have tabulated data within the supplementary disclosures of the audited financial statements that are less readable. Furthermore, firms that restate their financial statements with the filing of an 8-K or that are subject to an investigation by the Securities and Exchange Commission also have audited financial statements that are less readable, and the dollar magnitude of such restatements for specific accounts varies proportionately with the increase in visual complexity for the primary financial statement tied to that account. Taken together, these findings provide evidence that firms which are manipulating earnings consistent with the process and outcome of earnings management tend to have tabulated financial data that appears obfuscated. Furthermore, this obfuscation appears to be prevalent in both the reviewed sections of the 10-K and the audited financial statements
EVALUATING TREATMENT OF STORMWATER CONTAMINANTS BY GREEN STORMWATER INFRASTRUCTURE USING CHEMICAL AND BIOLOGICAL TOOLS
Stormwater is a global issue occurring in developed landscapes where large areas of impervious surface disrupt key parts of the water cycle including infiltration, groundwater recharge, and evapotranspiration. Stormwater runoff is generated when rainfall falls onto impervious surfaces and is unable to infiltrate soils, leading to massive volumes of water to move across the landscape into local receiving waters. In addition to causing hydraulic issues like flooding, stormwater runoff is a major vector of pollutant transport into surface waters. Stormwater contaminants can be toxic to aquatic species, so am important goal of stormwater treatment is to protect aquatic organisms from this toxicity. Green Stormwater Infrastructure (GSI) consists of a suite of stormwater control measures which aim to restore bits of the landscape to predevelopment hydrology by capturing, slowing, and infiltrating stormwater back into native soils. There are many types of GSI technologies, including bioretention, bioswales, green roofs, and permeable pavements, among others. GSI technologies were initially developed and adopted to manage stormwater quantity, but because of their widespread implementation, researchers started studying the potential for GSI to improve stormwater quality by removing contaminants. Various forms of GSI have since been demonstrated to provide treatment of suspended solids, metals, nutrients, and organic contaminants. However, the mechanisms and effectiveness of contaminant removal by GSI are still not well-understood, especially for contaminants of emerging contaminants. The degree to which GSI can protect aquatic organisms from stormwater contaminants is also poorly understood. This dissertation contains manuscripts which evaluate the ability of GSI to treat stormwater contaminants. The first chapter investigates biochar and fungi as bioretention amendments for bacteria and PAH treatment. The second chapter investigates whether permeable pavements are capable of mitigating pollution from tire wear particles and tire additive chemicals. Using the GSI systems from Chapters 1 and 2 as case studies, the third chapter compares chemical and biological tools for evaluating the ability of GSI to remove polycyclic aromatic hydrocarbons (PAHs) and related compounds from stormwater
CONCEPT MAPPING AND REFLECTION PROMPTS ON CHEMISTRY LEARNING OUTCOMES
There is growing support for combining concept mapping activities with metacognitive strategies such as reflection prompt to improve student outcomes in difficult science concepts. Although concept maps and reflection prompts have been widely shown to be effective for learning, little is known about the effect of various concept map types combined with reflection prompts on student outcomes. We conducted two experiments to examine the comparative effectiveness of different types of concept map activities and reflection prompts. In experiment 1, we compared the effectiveness of two concept map types combined with reflection prompts (yes or no): (i) collaborative concept map, and (ii) individual concept map. Undergraduate students enrolled in an introductory general chemistry course were randomly assigned to the four conditions. Each concept map group entirely constructed the relationship between concept on the topic of gas. Results showed a main effect for concept map types, with individual concept map groups outperforming the collaborative concept map group. In addition, participants in collaborative concept map group and no reflection prompt group report lower cognitive load compared to other conditions. In experiment 2, we compared the effectiveness of three collaborative concept map formats (correction concept map, fill-in the blank concept map and self-generated concept map) combined with reflection prompts (yes vs no) on undergraduate students’ learning outcomes. Students (N = 480) were randomly assigned to one of the six conditions on the topic of quantum numbers. The dependent measures were tests of posttest and cognitive load score. The study found no interaction effect between the collaborative concept map formats and reflection prompts. There was no significant difference between the three collaborative concept map formats. However, collaborative correction map mean outperformed the means of collaborative scaffolded and self-generated maps. The reflection prompts provided no additional benefit over no reflection prompts because participants in both conditions were able to engage at interactive level of collaboration
Are plant breeders willing to change food systems?
Food systems is the term for the all-encompassing entity which includes the individuals involved in growing, transforming, and consuming food. From this perspective of course food is the essential core of the whole system. No food, no system. Every food item derives, directly or indirectly, from a plant. Most likely a domesticated plant, an organism that has been selected to fit and fulfill the desires of a particular group of people, a work of careful selection carried out initially by farmers and more recently by plant breeders. Wheat well exemplifies this relationship. A staple crop that has accompanied a significant amount of the human population for thousands of years, changing according to new latitudes, pathogens and tastes, today however wheat is shaped not so much by “us” but by the organized forces of the market. Refined flour is the focus of the American wheat industry, tasking breeders to develop varieties yielding high amounts of white refined flour satisfying the consumer’s demand for white bread. Breeders have the agency to shape crops, and consequently to enable cropping and food systems. Wheat has been bred to produce refined flour for over 150 years. A fact likely contributing to the current deficiency in dietary fiber experienced by the vast majority of the American population. Wheat breeders instead, could be working on developing varieties for whole grain use, overcoming technological and organoleptic concerns and limitations to its adoption. From an agronomic perspective, wheat also could be bred to overcome some of the negative externalities associated with its production, for example soil erosion. Hybridizing wheat with perennial wild relatives produces a perennial grain crop, combining potential soil health benefits and grain production. Breeders have contributed to change nutritional and agricultural landscapes in the past, will they now take the initiative to address present societal and environmental issues
RELATIONSHIPS BETWEEN PERINATAL MATERNAL CORTISOL, PSYCHOLOGICAL WELL-BEING, AND INFANT NEUROENDOCRINE STRESS AXIS ACTIVITY
The major question this body of research seeks to address was how perinatal maternal psychophysiological stress and military lifestyle experiences relate to infant hypothalamic-pituitary-adrenal (HPA) axis activity and temperament development. It is important to investigate these relationships because it could lead to a better understanding of later-life health effects as a result of detrimental conditions in a crucial developmental period. To address this question, an interdisciplinary approach was used spanning endocrinology, psychology and behavior, and sociology studies and utilizing both qualitative and quantitative methods for this research in a longitudinal manner with repeated measures.The research populations used in this research consisted of pregnant and postpartum women and their infants. Chapter two consisted of general population women recruited for in-person participation in the pacific northwest and only assessed maternal cortisol concentrations in relation to birth outcomes and internalizing symptoms. For chapter three and four this population was limited to spouses of active-duty military members (Chapter 3) and spouses and infants of active-duty Navy members (Chapter 4) but expanded to include the whole United States. Chapter 3 expanded on the research of chapter 2 by incorporating military lifestyle stressors as factors that influenced internalizing symptoms during the perinatal period. And chapter 4 incorporated methods from the studies from the two previous chapters and expanded the research to include analysis of impact on infant HPA axis and temperament development. Participant recruitment and psychological data collection was conducted online. Hair cortisol (HCC) was used as an integrated measure of cortisol concentration over a period of three months. Infant saliva was collected three times a day for two days to observe an averaged diurnal circadian cortisol response at two postpartum timepoints. Longitudinal data was analyzed using repeated measures MANOVA. Relationships between multiple factors and interactions were assessed using analysis of variance. Linear regression was used to identify simple relationships between variables and interactions.Key findings from this research include the identification of the relationship between unpredicted birth complications and maternal psychological distress as well as perinatal cortisol concentrations using HCC. The third chapter validated the military lifestyle questionnaire (MLQ) developed by the researcher and identified six potential factors relevant to military spouses and further associated those factors with depression and perceived stress scores during the perinatal period. The major finding from the fourth chapter was the identification of the link between perinatal maternal HCC and infant HPA axis activity and behavior. These findings support the need to monitor maternal psychological and biological stress throughout the perinatal period to ensure the healthy development of their children. This research also supports a growing body of research finding connections between maternal cortisol and infant HPA axis activity and behavior or temperament development. Additionally, this study expands on what is known about the stress factors specific to military populations and supports previous research identifying deployment stress as a major factor to maternal psychological distress. These findings indicate a need for future research into the associations between perinatal maternal cortisol and psychological distress. There is also further need to validate the MLQ scale with a larger sample. Lastly, further research should be conducted to investigate the associations between maternal cortisol concentrations and reports of infant behavior
ROOTS OF ROT IDENTIFYING THE MOLECULAR ACTORS AND COMMUNITIES UNDERLYING APHANOMYCES INFECTION IN TWO AGRICULTURALLY IMPORTANT LEGUME CROPS
Lentils (Lens culinaris) and alfalfa (Medicago sativa) are among multiple pulse crops susceptible to Aphanomyces root rot (ARR). Aphanomyces euteiches, an ancient and destructive oomycete is the filamentous pathogen responsible for ARR infection. As obligate hemibiotrophs, A. euteiches’ infection strategy includes secreting effectors, facilitating evasion and suppression of the plant immune system to complete its reproductive lifecycle, and generating hardy oospores viable in soils for up to a decade. To date, multiple effectors and their plant cell targets have been identified, though the broader molecular communities that these effector-target pairs operate within remain unknown. Using data generated during ARR infection of legume host roots, we conduct a systems level analysis, employing machine learning models to identify ARR resistance-associated features (RAFs) in lentil (LRFs), Medicago (MRFs), and A. euteiches (ARFs) before and during infection, to construct directional host↔pathogen gene antagonism networks (GANs). Supported by the associated functional and contextual analyses, these GANs identify specific ARF↔(LRF/MRF) interactions during infection, and associate functionality to the greater communities that they belong to. Furthermore, by constructing and analyzing these GANs, we identify LRF-1 and MRF-1 as potential A. euteiches-specific pattern recognition receptors (PRRs) that likely confer partial ARR resistance in lentil and Medicago. We also implicate a retrotransposon (RT) as potentially playing a key role during the infection of tolerant Medicago plant
Cooperative Extension as a Key Partner to Behavioral Health in Rural Communities
Since its beginning, the opioid crisis has been unrelenting and remains one of the most challenging public health problems to solve. This is especially true in rural communities which are disproportionately impacted. Although some initiatives to address overdose and poisoning deaths have found success, significant challenges remain (e.g., fentanyl, workforce shortages). To close this gap, it will take shared leadership and partnership with community-based organizations as key collaborators to strengthen health systems and related opioid outcomes. The present article proposes an expanded role for the nation's Cooperative Extension System in opioid prevention, treatment, and recovery through a local Health Extension Agent. Leveraging the extension system is especially important in rural and underserved communities. Building on key strengths of the extension network and guidance of national frameworks, this role would serve as an intermediary linking community members and programs, health care providers, and addiction prevention and treatment experts to facilitate access and uptake of evidence-based promotion, prevention, treatment, and recovery strategies. We propose core functions of a Health Extension Agent including (a) provision of training and technical assistance, (b) facilitation of community partnerships and shared resources, and (c) dissemination of evidence-based programs and/or approaches driven by community need. Given the key strengths of extension, this system is well poised as a key contributor to addressing the behavioral health continuum from promotion through recovery related to opioid misuse.Public Health Significance StatementRural communities are disproportionately impacted by opioid use and related consequences. Collaboration between health providers and community organizations is critical to a comprehensive approach addressing opioid use. This article describes how a Health Extension Agent approach can facilitate collaboration to support rural behavioral health