Theses, Dissertations, and other Required Graduate Degree Essays

Receive updates for this collection

This collection contains digitized SFU theses except for those theses submitted within the last 12 months. If you cannot find the thesis you are looking for please search Recently Submitted Theses as it may be a recently submitted thesis and thus not yet available in Summit.

Rationalizing professional misconduct: An examination of techniques of neutralization in lawyer discipline proceedings

Author: 
Date created: 
2019-12-11
Abstract: 

This thesis investigates the use of neutralization techniques by lawyers to justify, excuse, and rationalize their behaviour in disciplinary action for misappropriation, real estate fraud, and conviction for serious financial criminal offences. In addition to assessing the nature and frequency of lawyer neutralizations, this study also considers the extent to which law society discipline hearing panels evaluate and respond to these defence and mitigation strategies in making a sanctioning determination. The dataset consists of 393 law society discipline decisions from eight of Canada’s 14 provincial and territorial law societies decided between 1990 and 2017. Content analysis addresses the characteristics of these lawyers, how they use techniques of neutralization and are disciplined by the law societies, and how hearing panels evaluate and respond to these rationalizations. The research findings have implications for neutralization theory and its application to lawyer discipline, for stakeholders and policymakers. The conclusions focus on three issues: 1) the prevalence of substance use and other mental health concerns in lawyer discipline cases, 2) the role of post-offence mitigation in the sanction determination, and 3) the suggestion that mitigating factors should be re-examined as techniques of neutralization with the goal of neutralizing some of them when imposing sanctions in disciplinary cases.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Joan Brockman
Department: 
Arts & Social Sciences: School of Criminology
Thesis type: 
(Thesis) Ph.D.

Pursuing equity: How women academics are challenging gender norms and (re)shaping university culture in Ghana

Author: 
Date created: 
2019-12-06
Abstract: 

This research focuses on the experiences of women faculty, PhD candidates and administrators, and the courses of action they employ to raise consciousness about systemic gender bias and pursue gender equity. I discuss the methodologies that participants use to overcome barriers and sensitize their social and professional environments to the experiences of women. Using nego-feminist theory and methods, participants initiate critical discourse with family, peers and institutional power holders, encouraging their respective families, colleagues, departments and universities to be gender-sensitive, reshaping their social and professional environments. Moving beyond their own departments, participants circumvent gender labour barriers and engage in collaborative co-development; building formal and informal networks, they effectively (re)create disrupted systems of support, furthering women's professional development. Women academics co-create anti-harassment policies and gender and advocacy centres, changing the institutional structures of their universities as part of their on-going pursuit of equity.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Ann Travers
Department: 
Arts & Social Sciences: Department of Sociology and Anthropology
Thesis type: 
(Thesis) M.A.

Audio-visual speech processing using deep learning techniques

Author: 
Date created: 
2019-10-02
Abstract: 

Speech separation is the task of segregating a target speech signal from background interference. To differentiate the separation of multiple speech sources from separating speech from non-speech noise, the terms Speaker Separation and Speech Enhancement (or Denoising) are commonly used, respectively. Speech separation can benefit from exploiting different modalities, i.e., audio and visual, and directional information when multiple microphones are available. A new approach is developed for subject-independent speaker separation by incorporating spectral, spatial and visual features. The audio signals have their magnitude and phase modified in the frequency domain for the speaker separation. The key idea is to estimate the target magnitudes from the audio with the Permutation Invariant Training (PIT) technique and then refine these estimates using both visual and spatial audio features. Specifically, visual features are matched to the corresponding audio, and spatial audio features are used as side information and shown to provide drastic improvement for magnitude and phase estimation in terms of output speech intelligibility, quality, and the separation performance. Visual information also provides improvements. Hence, both visual and spatial features are shown to be useful for speaker separation. A monaural speech enhancement model is also developed which incorporates both audio and visual information. In contrast to the audio-visual speaker separation model, the audio-visual speech enhancement model operates in time-domain. Hence, there is no need for a transformation and separate models for estimation of magnitude and phase spectra. According to the results of the objective evaluations, exploiting visual information for enhancement applications improves the performance in terms of both output quality and intelligibility.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Ivan Bajic
Department: 
Applied Sciences: School of Engineering Science
Thesis type: 
(Thesis) M.A.Sc.

Mathematics teaching and social media: An emergent space for resilient professional activity

Author: 
Date created: 
2019-11-19
Abstract: 

Professional activity around mathematics teaching is considered vital in the improvement of mathematics education at all levels. Research in mathematics education has identified aspects of teacher professional development that are effective, but there has been a recent push for better understanding how mathematics teacher professional development can also be sustainable. To this end, informal professional activity around mathematics teaching has become of particular interest in the field. Since many education professionals are turning to resources that are becoming increasingly available beyond the confines of institutional boundaries, such as via social media, many of the constraints of traditional forms of professional activity are being bypassed, allowing for informal professional activity to flourish. In some cases, collectives of professionals have formed in such contexts. One such collective, referred to as the Math Twitter Blogosphere (MTBoS), has remained resilient for almost ten years with ongoing activity around mathematics teaching occurring daily. Although this self-organized, bottom-up, emergent collective thrives with engagement around mathematics teaching, it has received very little empirical attention within mathematics education. As such, this study investigates the inner workings of this collective by drawing on tenets of complexity thinking to develop a more comprehensive description of its nature and how it thrives. Informed by an ethnographic journey of becoming a MTBoS participant, I select and analyze data in innovative ways to uncover both the ideational network in MTBoS and the social network that drives its existence. Analysis of these networks illuminates the influence not only of social capital, but also of ideational capital, both of which are necessary for determining ideational resilience within the collective. The results of this research indicate not only the popular topics within MTBoS, but also more importantly, features that drive ongoing and often generative activity around mathematics teaching within this online, unprompted, unfunded and unmandated professional setting.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Peter Liljedahl
Department: 
Education: Faculty of Education
Thesis type: 
(Thesis) Ph.D.

Accurate and rapid detection of circular RNA through splice-aware pseudo-alignment scheme

Author: 
Date created: 
2019-08-22
Abstract: 

The ubiquitous abundance of circular RNAs (circRNAs) has been revealed by performing high throughput sequencing in a variety of eukaryotes. circRNAs are related to some diseases such as cancer in which they act as oncogenes or tumor-suppressors, and therefore have the potential to be used as biomarkers or therapeutic targets. Accurate and rapid detection of circRNAs from short reads remains computationally challenging. This is due to the fact that identifying chimeric reads, which is essential for finding back-splice junctions, is a complex process. The sensitivity of discovery methods, to a high degree, relies on the underlying mapper that is used for finding chimeric reads. Furthermore, all the available circRNA discovery pipelines are resource intensive. We introduce CircMiner, a novel stand-alone circRNA detection method that rapidly identifies and filters out linear RNA-Seq reads and detects back-splice junctions. CircMiner employs a rapid pseudo-alignment technique to identify linear reads that originate from transcripts, genes, or the genome. CircMiner further processes the remaining reads to identify the back-splice junctions and detect circRNAs with single-nucleotide resolution. When requested, with extra time overhead, CircMiner also reports the mapping locations of the linear reads. We evaluated the efficacy of CircMiner using simulated datasets generated from known back-splice junctions and showed that CircMiner has superior accuracy and speed compared to the existing circRNA detection tools. Additionally, on two RNase R treated cell line datasets, CircMiner was able to detect more consistent, high confidence circRNAs compared to untreated samples of the same cell line.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Binay Bhattacharya
Department: 
Applied Sciences: School of Computing Science
Thesis type: 
(Thesis) M.Sc.

The evolutionary origins of amphibian extinction risk

Author: 
Date created: 
2019-09-18
Abstract: 

The rise of humanity to ecological dominance has precipitated concerted patterns of environmental change across every biome on Earth. Human activities can upend the adaptive landscapes on which species' have evolved, causing the sudden maladaptation of lineages to these novel conditions. Amphibians are amongst the most threatened vertebrates, with contemporary extinctions driven by multiple interacting stressors including habitat destruction, introduced pathogens, and climate change. Despite these looming threats, we understand little about how or why susceptibility to these stressors varies across amphibian lineages. In this thesis, I investigate the evolutionary origins of modern extinction risk in the Amphibia, by examining comparative patterns of susceptibility to various drivers of extinction. First, I show that modern extinction risk positively covaries with speciation rates across amphibian genera due to the most rapidly-diversifying clades producing numerous range-restricted and vulnerable species. Second, I demonstrate how evolutionary dynamics may influence local-scale extinction by examining amphibian species' responses to deforestation across the world. Contrary to patterns of global threat, the slowest-diversifying amphibian lineages are disproportionately lost from human-modified ecosystems - which may reflect a relationship between diversification and niche lability. Third, I examine phylogenetic and trait-based patterns of susceptibility to a human-dispersed fungal pathogen. Though species' ecology and life history consistently shape infection patterns across diverse amphibian assemblages, these traits appear to bear little weight for species' extinction risk from disease epidemics. Fourth, I test the relative effects of both dehydration and temperature on performance, and therefore climate risk, in three ecologically diverse anuran species. Performance was maintained across broad thresholds of dehydration in all species, but warmer temperatures accelerated the onset of performance decline. Species-specific biophysical modelling revealed stark differences in how dehydration is likely to limit activity in each species, suggesting that desiccation physiology may be an important driver of extinction risk from climate change in amphibians. These studies collectively illustrate that amphibian species' responses to anthropogenic environmental change have deep evolutionary roots. In turn, we can expect our continued environmental dominance to fundamentally reshape the evolutionary tree of amphibians into the future.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Wendy Palen
Arne Mooers
Department: 
Science: Biological Sciences Department
Thesis type: 
(Thesis) Ph.D.

Spatial distribution of soil class and soil pH in the Thompson-Okanagan region, British Columbia

Author: 
Date created: 
2019-09-24
Abstract: 

Soils are facing great threats from climate change and anthropogenic activities. It is essential to understand the characteristics of soils, such as class and pH, especially when it comes to the issue of evaluating soil quality. In the Thompson–Okanagan region, previous soil surveys covered most parts of the region in polygon data form; however, it would be beneficial if soil data were available at a finer resolution and with uniform soil categories. The digital soil mapping (DSM) approach has shown promising results over various landscapes with limited available data. The main objective of this study was to use an ensemble learning approach to map the spatial distribution of soil classes and soil pH at 25-meter resolution in the Thompson-Okanagan region, BC. Random Forest (RF) was used to map 16 soil subgroups. Overall prediction accuracy was 65.4% with an independent validation dataset. The study of spatial patterns of soil pH was tested with a combination of multiple base learners, which included a Multilinear Regression (GLM) learner, Stepwise Regression (STEP) learner, Lasso and Elastic-Net regularized Generalized Linear Regression (GLMNET) learner, a Kernel-based Support Vector Machine (KSVM) learner, and RF. Base learners with higher prediction accuracy were used to develop a Super Learner (SL). The fitted SL was then used to predict soil pH for three depth intervals (0 – 5cm, 5 – 15cm, and 15 – 30cm) at 25–meter resolution. For all three depth intervals, the SL proved to have the lowest MSE value and better prediction accuracy than was obtained from just using one of the base learners.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Margaret Schmidt
Department: 
Environment: Department of Geography
Thesis type: 
(Thesis) M.Sc.

Cops and robbers with speed restrictions

Date created: 
2019-09-24
Abstract: 

The game of Cops and Robbers is a pursuit-evasion game played on graphs with two players, the cops and the robber, who take turns moving on the graph. In each turn, they may move to a vertex adjacent to their current position or stay where they are. The cops’ objective is to get to the same position where the robber is, which we refer to as to capture the robber, and the robber’s goal is to evade capture indefinitely. The basic question is to find the minimum number of cops that can guarantee capturing the robber in a given graph. A very fruitful research area has been developed around the idea of modifying the way in which the cops or the robber move and analysing how these changes affect the strategies and outcome of the game. In this thesis we will study the game when we impose additional speed restrictions on the players, variants of the game popularly known as “lazy-cops and robbers” and “active cops and robbers”. In order to do so, we introduce the concept of the wide shadow, aiming to improve known results and obtain new tools and techniques which may provide further insight into other open problems in the area.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Bojan Mohar
Department: 
Science: Department of Mathematics
Thesis type: 
(Thesis) Ph.D.

Monuments and the margins: Old age and memories of Scottish Jacobitism in the epitaph of Marjory Scott, 1700-1900

Author: 
Date created: 
2019-09-25
Abstract: 

In c. 1728, the Edinburgh poet Alexander Pennecuik (1684-1730) published an epitaph on the centenarian Marjory Scott of Dunkeld. Over the next two centuries, this text was widely circulated in a variety of media. This thesis argues that the Scott epitaph functioned as a textual monument to the anti-Union and Jacobite politics of Pennecuik’s day, over a period when Jacobite memory was delimited and trivialised in the public sphere. Close study of the epitaph’s original context and its transmission reveals such sources of memory to be highly portable and also flexible, suitable to the heterogeneous identities and memories of Scottish people. As a tribute to a centenarian, the epitaph further opens up an exploration of old age as a conduit for diverse memories and as a multivalent symbol of Scottish identity.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Aaron Windel
Department: 
Arts & Social Sciences: Department of History
Thesis type: 
(Thesis) M.A.

New societies

Author: 
Date created: 
2019-09-27
Abstract: 

New Societies is an interactive theatrical experience playing a strategic board game. Seated at tables from 4 to 6 guests, players compete against or collaborate with each other in order to build their ideal society. Facilitated by a cast of eleven performers, one seated at each table and three in the centre of the room, the players propel the main movement of the performance. Echoing the politics and systemic issues we face in contemporary societies, players must navigate difficult decisions as their divisions progress. Players can either choose to build their societies as Empires or Utopias but each decision the player makes will influence how all the other guests can play. New Societies is a playful challenge to reflect on how we operate as inheritors of our current societal structures and act potential catalysts for newer and better ones.

Document type: 
Graduating extended essay / Research project
Senior supervisor: 
Kyla Gardiner
Department: 
Communication, Art & Technology: School for the Contemporary Arts
Thesis type: 
(Project) M.F.A.