30921 research outputs found
Sort by
Culturally responsive leadership: Manitoba school principals’ perspectives on leading French immersion schools
No full textThis thesis explores the role of school leaders in facilitating culturally responsive learning environments within Manitoba's French Immersion schools, amidst a time of rapid demographic changes. Utilizing a qualitative, case-study approach based on Stake's methodology, the research draws on interviews with school principals across various educational contexts. Central to the inquiry is the question of how school leaders can effectively respond to increasing Indigenous, racialized and linguistically diverse students in French Immersion settings.
Guided by Pierre Bourdieu's theoretical framework—specifically his concepts of field, habitus, and capital—this study reveals the systemic barriers to inclusivity within French Immersion programs. The findings illustrate how the hierarchical nature of educational fields impact school leader agency and their ability to cultivate culturally proficient learning communities. By examining the habitus of educators, the research highlights the cultural and social capital that shapes their responses to diversity, as well as the phenomenon of hysteresis, where established practices lag behind changing student demographics.
Recommendations emphasize the importance of recruiting diverse professional staff, supporting teachers in developing inclusive practices, and building strong community relationships. School leaders must actively confront barriers to inclusion and foster a culture of openness among educators. By prioritizing continuous professional development and mentorship, principals can enhance staff capacity to meet the needs of all students. This research provides essential insights for developing equitable educational environments in French Immersion contexts, highlighting the critical need for culturally responsive leadership in the face of change.February 202
Community and the Black Wedding: an examination of a Jewish ritual’s historical context and development
No full textThis thesis is a study of the Black Wedding, a Jewish folk ritual which attempts to ward off an epidemic by selecting two disenfranchised people, on the periphery of the community, and marrying them to each other in a graveyard. The thesis will first analyze the Jewish context of this practice, then its broader, non-Jewish context. Finally, it will examine accounts of Black Weddings from different historical periods, concluding with a close reading of a Hasidic story.
There is a longstanding Jewish awareness of boundaries, often related to the holy and the unholy. The disenfranchised were often seen as sources of plague. By incorporating them into the community through a Black Wedding, boundaries were redrawn to preserve communal sanctity.
This Jewish practice emerged in a cross-cultural environment where marriage blurred the boundaries between life and death. The Black Wedding drew on both Jewish and non-Jewish beliefs and customs and was intended to protect non-Jewish neighbours as well.
Historical accounts of the Black Wedding show the gradual acceptance of this originally controversial practice, and its use by leaders seeking authority, especially Hasidic Rebbes. The Hasidic story which is the focus of the final chapter illustrates the range of meanings that the Black Wedding eventually took on, going as far as a sense of cosmic reconciliation.
This thesis builds on the work of Natan Meir, Jeremy Brown, and Hanna Węgrzynek who have recently contributed to the scholarship regarding the Black Wedding and the development of this Jewish folk ritual. This study also draws on the theoretical work of Mary Douglas, Arnold van Gennep, and Michael Satlow who have contributed scholarship on rituals and boundaries. In bringing these scholars together, this thesis contributes to the ongoing discourse regarding folk religious practices and how they continue to develop and evolve through historical accounts and memory.October 202
PROTOCOLS FOR MEASURING BIODIVERSITY: Phytoplankton in Freshwater
No full textThis report describes best practices for the collection and analysis of phytoplankton in freshwater
Reduction of Electron Repulsion in Highly Covalent Fe-Amido Complexes Counteracts the Impact of a Weak Ligand Field on Excited-State Ordering
No full textThe ability to access panchromatic absorption and long-lived charge-transfer (CT) excited states is critical to the pursuit of abundant-metal molecular photosensitizers. Fe(II) complexes supported by benzannulated diarylamido ligands have been reported to broadly absorb visible light with nanosecond CT excited state lifetimes, but as amido donors exert a weak ligand field, this defies conventional photosensitizer design principles. Here, we report an aerobically stable Fe(II) complex of a phenanthridine/quinoline diarylamido ligand, Fe(ClL)2, with panchromatic absorption and a 3 ns excited-state lifetime. Using X-ray absorption spectroscopy (XAS) and resonant inelastic X-ray scattering (RIXS) at the Fe L-edge and N K-edge, we experimentally validate the strong Fe-Namido orbital mixing in Fe(ClL)2 responsible for the panchromatic absorption and demonstrate a previously unreported competition between ligand-field strength and metal-ligand (Fe-Namido) covalency that stabilizes the 3CT state over the lowest energy triplet metal-centered (3MC) state in the ground-state geometry. Single-crystal X-ray diffraction (XRD) and density functional theory (DFT) suggest formation of this CT state depopulates an orbital with Fe-Namido antibonding character, causing metal-ligand bonds to contract and accentuating the geometric differences between CT and MC excited states. These effects diminish the driving force for electron transfer to metal-centered excited states and increase the intramolecular reorganization energy, critical properties for extending the lifetime of CT excited states. These findings highlight metal-ligand covalency as a novel design principle for elongating excited state lifetimes in abundant metal photosensitizers
Risk of Progression and Costs of Care for Patients with Type 2 Diabetes and Chronic Kidney Disease
No full textIntroduction: Chronic kidney disease (CKD) progression is associated with a significant incremental economic burden. Previous work has demonstrated high accuracy of the laboratory-based machine learning model, Klinrisk, in predicting the risk of CKD progression. We sought to use the Klinrisk model to evaluate the association of risk of CKD progression with healthcare resource utilization (HRU) and costs of care in adults with type 2 diabetes and CKD. Methods: This retrospective observational study included 413,177 eligible patients from Optum’s electronic health records database (1/1/2007–9/30/2022). Patients were classified into low-, medium-, and high-risk groups based on their 2-year risk of CKD progression as predicted by the Klinrisk model. All-cause HRU and medical costs during the 1 year after CKD were estimated for each group. Results: Of the 413,177 patients included, 110,399 (26.7%) were classified as low-risk of CKD progression, 253,188 (61.3%) as medium-risk, and 49,590 (12.0%) as high-risk. The observed risk of CKD progression at 2 years, 5 years, and 10 years was 18.6%, 36.5%, and 54.1% for high-risk patients, 3.7%, 11.7%, and 26.4% for medium-risk patients, and 1.5%, 5.7%, and 15.8% for low-risk patients, which were similar to the predicted risks of CKD progression. High-risk patients had higher HRU and more than 2–3 times higher costs than lower-risk patients. Inpatient costs were the major cost driver for high-risk patients. Conclusions: The Klinrisk model accurately identified patients with type 2 diabetes and CKD requiring the most healthcare resources. Such tools can support the identification and targeting of high-risk patients for interventions that may lead to a more cost-effective model of care
Exploring a design space of digital interventions for early adolescents’ disengagement from technology overuse
No full textThe widespread use of digital devices among children and teenagers has raised concerns about overuse, particularly for early adolescents (ages 11–14), who have unique developmental needs and reportedly spend more time with technology than any other age group. While numerous parental control tools exist to mitigate technology overuse, most overlook early adolescents’ perspectives. As a result, these tools often face resistance, contribute to parent-child conflicts, and may even be abandoned. Despite research on various mediation strategies, limited work has focused on designing digital interventions that actively incorporate early adolescents’ perspectives, particularly in the context of tech disengagement.
This thesis addresses this gap by investigating early adolescents’ perceptions of suitable interventions for managing their tech use and evaluating existing solutions against their preferences. Through a co-design study, we examine their conceptualization of tech disengagement and the design factors they prioritize. Building upon these insights and synthesizing prior relevant literature, we then introduce an initial design space for digital interventions tailored to this demographic. To further explore areas of alignment and divergence between early adolescents’ and their parents’ viewpoints, we conduct an elicitation study comparing their preferences within this design space. Finally, we systematically review prior research on tech disengagement interventions and analyze existing parental control applications to assess how well current solutions align with the needs and expectations of early adolescents, identified within our proposed design space.
This research contributes to the field of child-computer interaction by revealing early adolescents’ perceptions of tech disengagement, defining and exploring an initial design space for early adolescent-centric digital interventions, and systematically analyzing existing research and current solutions to identify gaps and areas for further development. These insights can be leveraged by HCI researchers and practitioners to ground future design explorations of digital interventions that better support early adolescents in managing their tech use.NSERC: https://doi.org/10.13039/501100000038October 202
Towards automatic vulnerability management in open-source software
No full textOpen-source software (OSS) is the foundation of the digital world, yet its transparent development model exposes systems to security vulnerabilities. Effective OSS vulnerability management requires two fundamental capabilities: vulnerability detection (VD), which identifies vulnerable code in repositories before exploitation, and vulnerability-fix detection (VFD), which monitors upstream commits to identify security patches. While VD prevents vulnerable code from reaching production, VFD enables downstream organizations to proactively apply critical patches before public disclosure.
However, data challenges limit current approaches. This thesis addresses four critical data-centric challenges through targeted studies. First challenge is extreme class imbalance. Vulnerability code comprising less than 5% repositories, data imbalance causing detection models to lose up to 73% effectiveness. We systematically evaluated sampling strategies on vulnerability detection models, demonstrating that random oversampling on raw code improves both VD model's performance and interpretability.
Second, long-tail vulnerability distributions leave rare but critical vulnerability types underrepresented in training data, yet these edge cases often matter most for safety-critical systems. We developed MoEVD, a mixture-of-experts architecture that routes input to specialized experts, achieving 12.8% higher F1 while improving detection of underrepresented categories.
Third, security-critical code changes are often subtle, only a small part of code changes are security related. Existing work often overlook such distinctive patterns. We created VFDelta using fine-grained delta embeddings to capture nuanced changes, improving F1 by 77.4%.
Fourth, vulnerability fixes hide within tangled commits combining security patches with unrelated changes, while crucial context resides in scattered artifacts. We introduced LLM4VFD, which uses large language models to integrate code changes with development artifacts and vulnerability history, achieving 68-145% F1 improvements while providing explanations.
This thesis provides novel techniques and practical guidance for improving automated vulnerability and vulnerability-fix detection in OSS. The presented approaches address four critical limitations in existing tools through data-centric solutions, offering both enhanced detection accuracy and explanations to bridges the gap between academic prototypes and real-world deployment.October 202
Modifiable risk factors in administrative health data: validity of smoking identification algorithms and patterns of social determinants of health documentation
No full textIntroduction: Modifiable risk factors such as smoking and adverse psychosocial circumstances contribute significantly to chronic health conditions such as hypertension. Administrative health data (AHD), including hospital discharge abstracts and physician billing claims, can be used to identify these risk factors and monitor disease burden. However, since AHD is primarily intended for healthcare management, validating algorithms that identify modifiable risk factors is crucial to ensure accuracy when repurposing for research or surveillance. This research aimed to evaluate the validity of using AHD to ascertain modifiable risk factors. The specific objectives were to: (1) synthesize evidence on the validity of electronic health data for identifying smoking status, (2) develop and assess model-based algorithms (MBAs) to identify smoking status using AHD, and (3) examine patterns in the documentation of social determinants of health (SDOH) in AHD to assess their face validity.
Methods: Four studies were conducted. The first was a systematic review and meta-analysis that assessed the validity of electronic medical records (EMR) and AHD for capturing smoking status in observational studies. The remaining three were retrospective cohort studies using AHD and clinical registry data from the Manitoba Centre for Health Policy. The second study evaluated the validity of Least Absolute Shrinkage and Selection Operator and Random Forest machine learning (ML) MBAs for identifying current smoking. The third study examined whether synthetic data-based MBA training data augmentation could enhance diagnostic algorithm performance, using three data generation techniques: Synthetic Minority Oversampling Technique (SMOTE), Conditional Tabular Generative Adversarial Networks (CTGAN), and Classification and Regression Trees. The fourth study assessed the documentation of SDOH to evaluate face validity.
Results: ML MBAs, especially those using EMR, outperformed rule-based algorithms but often had low sensitivity. ML improved sensitivity, although the positive predictive value was lower. Data augmentation, particularly with SMOTE, enhanced specificity, and CTGAN improved sensitivity for underrepresented groups. SDOH documentation varied across demographics, with shifts linked to policy changes and events like COVID-19.
Significance: This research validates smoking identification algorithms for AHD, demonstrates how ML and synthetic data can enhance performance metrics, and highlights variability in SDOH documentation, informing risk prediction models and public health surveillance.October 202
The effects of climate change on growth and thermal habitat use of Cumberland Sound Arctic charr (Salvelinus alpinus)
No full textThe Arctic is warming at an unprecedented rate and experiencing longer growing seasons, greater rainfall, and less snowfall. Cold-adapted ectotherms, such as Arctic charr, Salvelinus alpinus, are likely to experience growth changes. Anadromous Arctic charr (charr, hereafter) are of great importance for northern communities, providing income from commercial fisheries and food security from subsistence harvest. While warming may initially increase the growth of charr, temperatures exceeding the optimum for growth can lead to metabolic stress, slowed growth, and higher mortality. This thesis examines whether annual growth and thermal habitat use have changed in three charr stocks from Cumberland Sound, Nunavut, with climate change between 1984 and 2013. In Chapter 2, I examined age-specific growth using otolith radius and annulus lengths. Trend analyses indicated significant growth increases in ages 1-6 and 8. Mixed modelling revealed growing degree days (GDD=°C*Day, base 4°C) positively influenced growth, while annual precipitation (mm) had an overall negative effect. Variation in growth between stocks was most pronounced in ages 1-6, coinciding with the pre-migratory period. These results suggest that charr experienced the most growth increases prior to their first migrations. Chapter 3 explored whether growth changes resulted from longer growing seasons or warmer conditions supporting higher metabolic rates. I used otolith-derived δ18O to estimate summer experienced temperatures between 1987 and 2013. Using a mixed model, I found no significant relationships between summer experienced temperatures and the climate variables (GDD and annual precipitation). Experienced temperatures decreased with age in each stock, indicating the presence of distinct thermal habitats and thus, thermal stratification. To control for behavioural thermoregulation, I ran a linear regression to investigate changes in experienced temperatures in ovo from 1987 to 2008. Unexpectedly, the results suggest experienced temperatures in ovo have decreased over the time period. However, it is hypothesized that the cooler temperatures are likely a reflection of changing isotopic compositions of the study lakes in response to climate change. Overall, the results suggests that increases in growth are likely driven by longer growing seasons rather than rising temperatures, as evidence of thermal stratification and a disconnect between experienced temperatures and climate variables was found.Natural Sciences and Engineering Research Council of Canada - Alexander Graham Bell Canada Graduate Scholarship - Master's
Nature Manitoba Award
Manitoba Big Game Trophy Association BursaryOctober 202
Maternal metabolic health conditions and risk of stillbirth in India: evidence from a nationwide survey
No full textBackground: Stillbirth, defined by foetal death at or beyond 28 weeks of gestation, represents a significant challenge in India, contributing to approximately 500,000 foetal deaths each year. The country’s stillbirth rate of 12.2 per 1000 births underscores the imperative to address this preventable occurrence. While maternal metabolic conditions diabetes, and hypertension, are widely recognized as established risk factors for stillbirth worldwide, the extent of their impact on India’s stillbirth burden remains inadequately elucidated due to limited evidence.
Methods: This cross-sectional study utilized NFHS-5 data to examine stillbirths in the most recent pregnancy outcomes of 204,723 women aged 15–49 years, sampled from all states and union territories of India. The primary exposures assessed were diabetes, and hypertension. Descriptive analyses were conducted to determine the prevalence of diabetes, hypertension and stillbirths. Logistic regression was used to quantify the association between diabetes, hypertension and the risk of stillbirth, indicated by adjusted odds ratios (AOR) with 95% confidence intervals (CI). The study also assessed effect modification by maternal age, education, wealth quintile, and social category.
Results: The prevalence of diabetes and hypertension was 1% and 3% respectively, while the stillbirth rate was 1%. diabetes conferred a significantly higher risk of stillbirth with an increase of 74% (AOR 1.74, CI 1.14–2.67) as compared to women without diabetes. The risk was potential among mothers with hypertension with an increase of 50% (AOR 1.50, CI 1.16–1.95) on contrary to women without hypertension. The combined model (i.e. having diabetes or hypertension) also showed a significant risk of stillbirth with a higher risk of 58% (AOR 1.58, CI 1.25–1.99) indicating a synergistic interaction. Stratified analyses revealed the stillbirth risk among mothers belonging to the scheduled caste category (AOR 1.30, CI 1.10–1.53).
Conclusion: Diabetes, and hypertension, increase stillbirth risk in India, highlighting the need for better metabolic health management pre- and during pregnancy. Our research highlights the need of integrated care for diabetes and hypertension is crucial. Targeted interventions for high-risk mothers and improved screening are vital to reduce stillbirth rates. More research is needed to understand these risks better. Collaboration across medical fields is essential to save lives and improve pregnancy outcomes