Alliance One Tobacco (Malawi)
Academic Research Repository at the Institute of Developing EconomiesNot a member yet
52305 research outputs found
Sort by
Rebuilding Home and Reconstructing Community: Neighborhood Ties and Culture in post-Tubbs Fire Coffey Park, California
No full textWith climate change having an ever-increasing and direct impact on communities
across the globe, there has been an effort to better understand how these communities change following a natural disaster. In this study, Coffey Park, a neighborhood of Santa
Rosa, California, which was destroyed following the 2017 Tubbs Fire, is presented as one of these communities. I examine how the neighborhood did, or did not, change post-disaster. Previous research findings are mixed on the topic, but many present post-disaster environments that are vastly different than those before the disaster, especially when viewed through a lens of gentrification.
Using interviews with a sample of residents in Coffey Park, I found that the story is complex. While some residents felt that their neighborhood had changed, with many changes fostering a stronger community post-fire through a sense of shared identity, there were lines drawn of who and what this shared identity encompassed. Participants also had mixed feelings as to whether the neighborhood was gentrified, with many pointing to
other neighborhoods lost to the Tubbs Fire and rebuilt that they thought were gentrified in comparison. As demonstrated in quantitative data and under a conventional
understanding of gentrification, the neighborhood did not meet the definition: it lacked a
large influx of wealth and businesses, a dramatic decrease in non-White residents, or the
sense that there was a displacement of prior residents, even as home prices rose due to the
production of brand-new homes among existing, non-displaced residents. In this scenario,
while gentrification may have simply not occurred, this study suggests that gentrification can happen in other forms and that the idea of gentrification is more fluid than
conventionally defined. Participants by and large felt a stronger sense of community,
significantly higher levels of trust, and a powerful sense of shared identity that at times
precluded new residents or those that did not share their experience
Financial Regulation: Still Unsettled a Decade After the Crisis
No full textA decade after the darkest moments of the financial crisis, both the US financial system and the legal framework for its regulation are still in flux. The post-crisis regulatory framework has made systemically important banks much more resilient. They are substantially better capitalized and less dependent on runnable short-term funding. But the current regulatory framework does not deal effectively with threats to financial stability outside the perimeter of regulated banking organizations, notably from forms of shadow banking. Moreover, with the political tide having for the moment turned decisively toward deregulation, there is some question whether the resiliency improvements of the largest banks will be preserved. This article assesses the accomplishments, unfinished business, and outstanding issues in the post-crisis approach to prudential regulation.Version of Recor
On Scaling Dynamics in Deep Learning
No full textIn recent years, the field of Deep Learning has experienced remarkable growth, fueled by advancements in computational power and data availability, arguably leading to the emergence of genuine Artificial Intelligence. One critical aspect underpinning this success is the predictable scaling attributes of models' performance, adhering to power laws concerning resources invested in training, such as model size, data volume, and training duration. This thesis delves into the concept of Scaling Dynamics, systematically investigating the relationship between a model's behavior and the resources invested in it. We explore key aspects of Scaling Dynamics, including Regimes of Learning, Monotonicity, Symmetries, and Trade-offs. Our main contributions are: (1) Observing that DNNs initially learn simple concepts and progressively increase in complexity as a function of resources invested; (2) Identifying the Deep Double Descent phenomenon, wherein models exhibit unexpected behavior in the critical regime between under- and over-parameterization; (3) Introducing the pointwise perspective, enabling a fine-grained understanding of distribution shifts; (4) Demonstrating that in the online learning regime, SGD noise does not enhance optimization beyond computational considerations, and that the training dynamics do not deviate far from the path of Gradient Flow.
These findings pave the way for future exploration, raising intriguing questions about the nature of Scaling Dynamics and how they shed light on Deep Learning as a whole
Notarial Information and the Production of Knowledge in Rural Society
No full textOne of the most dramatic transformations of the lives of everyday people in medieval Europe was the sudden and widespread increase in the number of written contracts at the turn of fourteenth century. To give a sense of the magnitude of change, in the former county of Provence in southeastern France, the volume of contracts grew by several orders of magnitude from the thousands in 1300 to the millions just a century later. Everyday men and women from different socio-economic classes, professional backgrounds, and religious faiths all sought out the services of scribes known as public notaries to compose written documents. This dissertation explores how notaries managed the massive amounts of information that filled tens of thousands of registers and cartularies throughout the later Middle Ages. Drawing upon hundreds of unpublished manuscript codices and over 20,000 notarial contracts from southeastern France, the dissertation argues that an unintentional but powerful appeal of notarial contracts lay in their capacity to compress, store, and circulate the information that guaranteed a vast array of interpersonal and legal relationships. The effects of this information revolution transformed medieval social and economic life by expanding the possibilities for how individuals, families, ecclesiastical institutions, and political entities could form the relationships that constituted late medieval society
Women and Power: A Comparative Analysis of Selected Societies from the Neolithic Age into The Modern Era
No full textFour societies ranging approximately 10,000 years and occupying disparate geographic regions are analyzed for similarities stemming from the fact that each of these societies either is either overtly or suspected of being more honoring of women and mothers in positions of domestic, political, and social leadership. Utilizing various methods of determination from archaeological records to ethnographic data, these cultures are found to have various similarities which are different from patriarchal customs more familiar to many modern people. These similarities are then compared with and projected into common modern modalities of culture and suggestions for how changes could be made to potentially normalize greater social equality across gender lines
The Digital Divide: Counties Lacking Access to Broadband Internet Have Reduced Access to Mental Healthcare Resources
No full textBackground
Telemedicine has greatly improved mental healthcare access worldwide, particularly following the COVID-19 pandemic. However, the growing reliance on broadband internet-based mental healthcare raises concerns surrounding telemedicine’s accessibility in communities already facing barriers in seeking mental health information and care. This study aims to i) correspond access to broadband internet with access to several mental health resources and ii) quantify the association between social determinants of health and broadband access in the US.
Methods
For each of 3138 US counties, we collected data for the percentage of households without broadband access (2020 Federal Communications Commission Mapping Broadband Health in America), the density of various mental healthcare services (2017 National Neighborhood Data Archive), urbanization level (2013 Urban-Rural Classification Scheme for Counties), and percentage of households with an income below the poverty line (2019 American Community Survey). Two-sample t-tests and two-proportion z-tests were used to substantiate the association between broadband access and mental health resource availability, while multivariate linear regressions were performed to quantify the association between broadband internet access and mental health resource availability, while controlling for urbanicity level and poverty rate. Finally, geographical trends in broadband access and mental health services were visualized in QGIS.
Results
US counties with reduced broadband access have lower average densities of mental healthcare physicians, non-physician mental health practitioners, inpatient psychiatric and substance abuse treatment facilities, and outpatient facilities (P<0.001). Moreover, counties with reduced broadband access are nearly three times as likely to have no mental health physicians and no outpatient facilities, over twice as likely to have no non-physician mental health practitioners, and nearly twice as likely to have no psychiatric/substance abuse hospitals (P<0.001). Finally, multivariate linear regressions demonstrated that a 1% increase the proportion of a county’s population lacking broadband access is associated with a decrease of 0.0452 nonphysician mental health practitioners (P < 0.001), 0.00578 psychiatric/substance abuse hospitals (P < 0.001), 0.00835 inpatient psychiatric and substance abuse treatment facilities (P < 0.001), 0.0272 mental healthcare physicians (P < 0.001), 0.0779 pharmacies and drugstores (P < 0.001), and 0.0488 outpatient facilities (P < 0.001) per 100,000 people.
Conclusion
These results suggest that expanding access to mental health resources in rural, low-income, and medically under-resourced communities is necessary in light of their reduced access to both broadband internet and mental healthcare services.Accepted Manuscrip
Gli1+ Stem Cells Contribute to the Regeneration of Periodontal Tissues
No full textPeriodontal diseases are mainly the result of infection and inflammation of the gingiva (gums), periodontal ligament (PDL), and alveolar bone that surround and support the teeth. 20-50% of the global population suffer from periodontal diseases, which lead to gum recession, alveolar bone loss and eventually tooth loss. Timely regeneration of injured periodontal tissues is an important prevention and treatment for periodontal tissue damage and disease. The periodontal crestal region (POCR) is the tip of gingival tissue near the tooth cervical region, which can be easily damaged as it is exposed to exogenous bacterial stimuli and serves as the first line protection for deeper periodontal tissues. Recently, we established a POCR-injury model and found that injured POCR can be regenerated. The POCR contains stem cells that can regenerate the PDL, gingiva and alveolar bone. In this study, we used a genetic lineage tracing approach to identify potential stem cells that participate in POCR regeneration in response to periodontal injury. We found that Gli1+ lineage cells contain POCR stem cells, which can contribute to gingival tissue, PDL and alveolar bone during normal homeostasis and expand after injury. Further study using the diphtheria toxin subunit A (DTA)-mediated cell ablation model to deplete the Gli1+ cells indicates that the Gli1 lineage cells are required for periodontal tissue turnover during homeostasis and periodontal tissue regeneration after injury
The Enlightened Nabob
No full texthttps://www.historytoday.com/archive/feature/rehabilitating-east-india-companys-nabobsThe nabobs of the East India Company were considered violent, greedy and – worst of all in a time of Enlightenment – uneducated. Could their reputation as philistines be laundered
Does Good Governance Result in Greater Racial Equity and Racial Harmony?
No full textThis thesis aims to discover whether Hawaii’s good governance indicators of
voice and accountability (VA), control of corruption (CC), and government effectiveness
(GE) from 2020-2023 affect Hawaii citizens’ current racial harmony (RH) and racial
equity (RE). A case study approach is taken here, integrating quantitative data and
qualitative data. The quantitative data comprises publicly available data from multiple
sources that create an index for GG, RE, and RH for all U.S. states. Qualitative data for
the state of Hawaii was collected from two online newspaper sources. The qualitative
data were compared to Hawaii’s quantitative data (a component of the compiled indices)
to make conclusions about the hypotheses posed at the beginning of the research
according to the theoretical models explained in the literature review section.
Quantitative data for all of the U.S. generally showed an increase in individual
GG indicators of VA, CC, and GE correlated with an increase of RE. An increase in
composite GG was correlated with a greater increase in RE than seen for individual GG
indicators. They also revealed RE is not a mechanism between GG and RH. However, the
case of Hawaii consistently shows RH is present when RE is. The initial review of and
the deeper dive into patterns in literature both support these findings. The deeper dive
into literature revealed a new type of relationship between variables – the presence of GG
and its application to either RH or RE causes an increase in the remaining variable.
Qualitative data and examples in literature support that GG and its application to RH
could result in greater RE, and that GG and its application to RE could result in greater
RH. On the other hand, the quantitative data support the first triad relationship and not
the second.
Governance plays a direct and important role in the equity and harmony citizens
of a state experience. As such, there is a need for governments to accept more
responsibility for how their governing affects the daily lives of their citizens and to make
more genuine, concerted, and collaborative efforts in their community to address the
issues of equity and harmony their citizens experience. Similarly, citizens should hold
their governments accountable for the immediate and long-term legacies of government
action
Essays on the Economics of Education
No full textEducation can improve individual well-being and reduce inequality. This dissertation studies educational decisions made by students and their parents, in particular the role that information plays in those decisions.
The first chapter studies the role of information about prices in students’ choices of where apply to college. I present descriptive evidence that college applicants, especially low-income applicants, tend to be more focused on and responsive to sticker prices rather to net prices. I also find that among top-percentile SAT-scorers, low- income students apply to less selective colleges than higher-income students. With these two findings in mind, I build and estimate a structural model of students’ information acquisition and college application choice portfolios. The model estimates suggest a sizable friction associated with information acquisition about net college prices that if eliminated may lead lower-income students to apply to higher sticker, but lower net, price schools.
The second chapter studies the role of information about quality in parents’ choices of which childcare provider to enroll their children at. I find that parents prefer childcare providers which receive five-star ratings from the state relative to lower star ratings. The responsiveness to the star rating is larger for home-based providers than for center-based providers, likely because parents have alternative channels to learn about centers. Responsiveness to quality ratings is also larger for higher-income parents and in neighborhoods with a higher share of white residents.
The third chapter studies how grade repetition influences learning and how parents’ beliefs and educational investments shift following a student being held back a grade. We find that following grade repetition, parents revise downward their expectations, beliefs, and investments. Students are discouraged by the retention, and decrease their beliefs in the value of study effort. Overall, we find negative effects of repetition on learning, with repeating students scoring worse on math, English, and Urdu tests and more likely to drop out than their peers