49755 research outputs found
Sort by
Essays on Labor and Public Economics
No full textThis dissertation consists of three essays on labor and public economics in the context of Brazil. More specifically, these essays explore topics related to relevant challenges faced by many developing economies, such as low access to higher education, labor informality, and political clientelism. In the first chapter, I study the effect of financial aid on the career trajectory of nontraditional college students. The second chapter focuses on the impact of an income tax deduction for employers on the labor formality status of their workers. At last, the third chapter investigates whether becoming a politically connected firm for the first time affects firm dynamics.
Chapter 1 examines the effect of financial aid on nontraditional college students, and whether it leads to an improvement in their labor market trajectories. I exploit a quasi-random variation in aid recipiency of the Brazilian federal student loan program, and compare loan takers against those who applied but did not take out the loan. I find that financial aid has a positive effect on degree persistence and completion, as well as robust evidence that nontraditional loan takers are more likely to be employed in higher-skilled occupations and by better quality firms after college, with no lasting negative impacts to their employment status or earnings. These results indicate that financial aid can positively disrupt the career trajectories of nontraditional students. In particular, this career upgrade seems to be partially driven by higher education completion. I also provide suggestive evidence that the cost-effectiveness of student loan policies that aim to increase college access for nontraditional students depends on the age at the time of application.
Chapter 2 studies the labor market effects of a tax reform that implemented an income tax deduction for employers of formal domestic workers. The objective of this policy was to reduce the formalization costs of informal housekeepers, a particularly vulnerable occupation in the labor market. I leverage the introduction of such a policy in the context of Brazil, and compare targeted and non-targeted housekeepers using a weighted event study design and household survey data. I estimate that the tax reform had a positive and significant impact on labor formality and hourly wages for targeted domestic workers, as well as a negative and significant effect on their number of hours worked per week. I also provide evidence that these results are mostly driven by housekeepers that were not eligible to the Brazilian conditional cash transfer program. At last, I show that the reform had a differential effect across the unconditional distributions of monthly and hourly wages but a negative impact on the distribution of weekly hours regardless of the decile in which the domestic worker was located.
Chapter 3 evaluates the impact of a newly-established political connection on firm growth, skill composition of its workforce, and survival. I focus on clientelistic relationships that are formed through a corporate campaign contribution. Using electoral and labor market administrative data, I employ a close election regression discontinuity design to evaluate this question in the context of Brazilian mayoral elections. I find that there are no sizable and long-lasting returns to becoming politically connected for the first time. Most estimated effects, for both levels and growth rates, are either not statistically different from zero or statistically significant but transient. I also present evidence that the establishment of a political connection has a heterogeneous effect depending on the size of the municipality, the relative importance of the corporate donation to the benefited candidate, the relative size of the campaign contribution to the firm’s finances, and the total number of donations over time and across elections
Promoting employee responsibility-taking: An examination of the situational and individual antecedents of responsibility-taking behaviors
No full textThis dissertation examines the individual and situational antecedents of employee responsibility-taking behaviors. Despite their importance for work performance, little research has explored the factors that encourage employees to self-regulate these demanding behaviors, which require fulfilling existing commitments and proactively taking ownership for emerging tasks. Drawing on self-determination theory, this dissertation investigates the impact of leadership style on responsibility-taking. Additionally, drawing regulatory mode theory, the effects of goal pursuit strategies were examined, both as stable individual dispositions and as situationally-induced motivational states.
Across two studies, including a cross-sectional survey of full-time workers and a randomized experiment, we found that autonomy-supportive leadership promoted responsibility-taking, while controlling leadership did not. Furthermore, our findings showed that this effect was mediated by employees’ perceptions that their leader satisfied their core psychological needs for competence and connection. In addition, these benefits were most pronounced for individuals primed to work with urgency, who were less willing to take responsibility when leaders failed to provide a compelling rationale for the work’s importance. By contrast, leadership style did not significantly affect participants’ willingness to engage in responsibility-taking behavior among those primed for deliberative, careful work. These findings highlight key motivational mechanisms underlying responsibility-taking and clarify the conditions under which leadership support is most influential
Futures Assembled: Futures Assembled: Sociotechnical Imaginaries and Infrastructural Transformation in Western Massachusetts
No full textHigh-performance computing (HPC) infrastructure is emerging as a critical site through which economic development, technological advancement, and urban revitalization are being negotiated in post-industrial contexts. While most scholarship on data centers focuses on hyperscale, corporate-owned facilities and their environmental costs, the distinct role of computing facilities operated by academic institutions remains underexamined. This thesis addresses that gap through a case study of the Massachusetts Green High Performance Computing Center (MGHPCC) in Holyoke, MA. Drawing on interdisciplinary literature in planning, infrastructure studies, and science and technology studies, this research examines the future-oriented imaginaries that facilitated facilities development and its material entanglements in Holyoke.
Using a qualitative approach focused on interviews, supplemented through an examination of relevant literature, policies, news articles and industry documents, this project investigates how sociotechnical imaginaries—articulated by state, academic, and municipal actors—have shaped the siting, development, and symbolic positioning of the MGHPCC. These imaginaries frame the facility as an engine of regional revitalization, scientific innovation, and environmental leadership. At the state level, the MGHPCC is positioned as a strategic investment in the knowledge economy and a model for green development. Locally, however, these imaginaries are contested, as residents express aspirations to be part of these visions through meaningful engagement.
It finds that while the MGHPCC is using clean energy and contributes to scientific research, its integration into the civic and economic fabric of Holyoke has been limited by uneven access, opaque accountability mechanisms, and a narrow definition of public benefit. The data center’s location on a former mill site reflects a broader trend of spatial continuity between past industrial geographies and emerging digital infrastructures. This study offers planners and policymakers insights for understanding both infrastructural and social dimensions of HPC infrastructure in post-industrial cities. In doing so, it underscores the need for transparent accountability frameworks, equitable access strategies, and sustained institutional partnerships to ensure that the promises made in the name of scientific and technological innovation are met with tangible, inclusive benefits for local communities
Determinants of Charging Station Usage Satisfaction in Beijing's Historic Downtown
No full textAgainst the background of the global transition to electric mobility, historic urban areas face difficulties in the construction of electric vehicle charging facilities due to conflicts between cultural protection and policy development goals, limited space, and the dual contradiction between people's livelihood needs.
This study uses multiple methods such as GIS spatial analysis, field observation, questionnaire surveys and PLS-SEM modeling to systematically analyze the spatial layout of charging facilities within the second ring of Beijing, user behavior and satisfaction feedback, and policy coordination.
The study found that:
(1) the spatial distribution of charging piles in the second ring is uneven, with facilities mainly concentrated in commercial areas and transportation hubs. There is an insufficient distribution of facilities in residential areas, especially old communities, and historical and cultural areas, especially hutong-dense areas, creating a spatial mismatch of high population density and sparse facilities.
(2) Residents' satisfaction with the use of charging piles is mainly affected by the three dimensions of Built Environment, APP Usability and Charging Experience. Among them, APP Usability has the highest overall influence, and Charging Experience, as a direct explanatory variable of satisfaction, is also a key mediating variable through which Built Environment indirectly affects Satisfaction.
In addition, this study also reveals the spatial-policy-technology triple contradiction in the layout of charging piles in historical urban areas, and proposes a policy optimization path
“Lifting As We Climb”: An Examination of the Role of Black Sisterhood Networks in Supporting and Sustaining Black Women Superintendents
No full textBlack women in the superintendency currently represent less than 1.5% of all superintendents in the United States and are more likely to serve in districts that are under resourced and low performing (Miles Nash & Grogan, 2022). In addition to the systemic challenges they face as they lead to improve outcomes for the students and communities they serve, Black women superintendents also face unique and distinct personal and professional challenges both before and during their superintendencies.
Research has shown that superintendent retention can be tied to access to networks and mentorship that provide superintendents with the necessary support to sustain in the role, with access to these networks being limited for Black women superintendents as there are few Black women in the role to provide mentorship (Alston, 2000; Angel et. al, 2013; Brown, 2014; Davis & Bowers; 2019; Hibbert-Smith, 2006; Kingsberry, 2017; Tillman & Cochran, 2000). The purpose of this study was to examine the ways in which Black sisterhood networks support and sustain Black women superintendents as they navigate challenges associated with their roles, and specifically their identities as Black women, as they work to positively impact student experiences and outcomes in their districts.
Eight current and former Black women superintendents from across the United States shared their journeys and experiences through semi-structured interviews, sista circles, and document analysis. Through a qualitative lens, this study sought to examine the ways in which the support of Black sisterhood networks have provided the necessary support and guidance for the participants. The findings and analysis of the data collected demonstrated that formal and informal Black sisterhood networks helped participants to gain a deeper understanding of the nuanced experiences of leading as a Black woman and has provided them with the tools to shift policies and practices in their districts while navigating an ever-changing political climate
The Influence of Selective Voluntary Motor Control on Terminal Swing Phase in Ambulatory Children with Cerebral Palsy
No full textBackground
The ability to walk and keep up with peers is an important goal amongst children with cerebral palsy (CP) and their caregivers. Factors associated with decreased walking speed are of interest amongst clinicians and researchers. Decreased knee extension during the terminal swing phase of gait is one of a variety of factors that contributes to inadequate step length and ultimately, to walking speed. The aim of this study was to examine factors associated with the magnitude of knee extension during terminal swing phase, including a task-specific measure of selective voluntary motor control (SVMC), such as the Walking Dynamic Motor Control Index (walk-DMC). It was hypothesized that impaired SVMC of the swing limb, as measured by walk-DMC, and decreased stance phase stability, as measured by single limb support time of the stance limb, would be significant predictors of decreased magnitude of knee extension during terminal swing phase in the group with diplegia. It was hypothesized that impaired SVMC of the swing limb would be a significant predictor of decreased knee extension during terminal swing phase in the group with hemiplegia, whereby stability in stance phase of the uninvolved limb was less impaired.
Methods
The study involved a retrospective analysis of instrumented gait data, inclusive of surface electromyography (sEMG), from an accredited motion analysis laboratory between 2015 and 2024. Participants between the ages of 7-18 years with a diagnosis of diplegic or hemiplegic CP, Gross Motor Function Classification System (GMFCS) levels I and II were included. A forward stepwise multiple linear regression model was used to predict the magnitude and the timing of knee extension during terminal swing phase. Predictors of interest included SVMC, as measured by walk-DMC, knee extension and ankle dorsiflexion joint range of motion (ROM), hamstring muscle spasticity, measures of knee extensor strength, and hamstring muscle-tendon lengthening characteristics of the swing limb, as well as single limb support time of the stance limb. GMFCS level and history of prior surgery were included as covariates.
Results
The final dataset used in the analysis included 90 individuals with diplegic CP (GMFCS level I, n = 35; GMFCS level II, n = 55) and 105 individuals with hemiplegic CP (GMFCS level I, n = 59; GMFCS level II, n = 46). SVMC, as measured by the walk-DMC, demonstrated a negative correlation with the magnitude of knee extension during terminal swing phase in both the group with diplegia (r = -.39, p < .001) and the group with hemiplegia (r = -.31, p = .001), such that better SVMC was associated with increased knee extension at terminal swing phase. Stance phase stability, as measured by single limb support time of the contralateral limb, was not significantly correlated with the magnitude of knee extension during terminal swing phase in the group with diplegia (r = .01, p = .892), while a positive correlation was observed in the group with hemiplegia (r = .33, p < .001). For the group with diplegia, the stepwise multiple regression model populated with walk-DMC, knee extension ROM, ankle dorsiflexion ROM, and extensor lag in the group with diplegia was statistically significant, R2 = .324, F(4, 85) = 10.195, p < .001, adjusted R2 = .292. The addition of GMFCS level and history of prior surgery, as covariates, led to a statistically significant increase in R2 of .111, F(2, 83) = 8.163, p < .001. For the group with hemiplegia, the stepwise multiple regression model populated with walk-DMC, single limb support time of the contralateral limb, and ankle dorsiflexion ROM in the group with hemiplegia was statistically significant, R2 = .251, F(3, 101) = 11.308, p < .001, adjusted R2 = .229. The model including the covariates of GMFCS level and history of prior surgery was not statistically significant (R2 of .006, F(2, 99) = .418, p = .660).
Conclusion
Decreased knee extension in terminal swing phase in individuals with CP is multifactorial and is not simply the result of tight hamstrings or weak knee extensors. Impaired SVMC, as measured by walk-DMC, is a significant predictor of decreased knee extension during terminal swing phase. Including task-specific measures of SVMC, such as the walk-DMC, is essential when using instrumented gait analysis (IGA) in assessing gait pathology to inform clinical decision-making and predict outcomes following treatment
A Novel Game-Theoretic Framework for the Autonomous Mobility Ecosystem
No full textThis dissertation aims to answer the research question of how to optimally design decision-making processes for autonomous vehicles (AVs), focusing on their dynamic velocity control and route choice within transportation networks. Unlike traditional traffic models that abstract away individual-level control, this work explicitly models multi-agent interactions among AVs, treating each as an intelligent, decentralized agent in a shared environment.
We formulate AV coordination as a large-population differential game, which converges in the many-agent limit to a class of mean field games (MFGs). These MFGs are designed to capture decentralized decision-making under mutual interactions and network constraints. Two core formulations are developed: (i) Spatiotemporal MFGs (ST-MFGs), which model continuous-time velocity control and agent distribution across space and time, and (ii) Graph-based Dual MFGs (G-dMFGs), which address simultaneous driving and routing decisions across transportation networks. Solving these MFGs involves challenging coupled forward-backward PDEs that are often computationally intractable in large-scale, high-dimensional systems. To overcome this, we propose a suite of AI-enhanced, learning-efficient solution methods.
These include: a hybrid Reinforcement Learning and Physics-Informed Deep Learning (RL-PIDL) framework, a Pure-PIDL method, and a Physics-Informed Graph Neural Operator (PIGNO) tailored for solving networked MFGs. These algorithms integrate physical structure and domain knowledge into deep learning architectures, leading to faster convergence, improved scalability, better generalization, and greater data efficiency compared to conventional solvers. This dissertation contributes a unified, interpretable, and scalable framework that merges mean field game theory with physics-informed machine learning to support decentralized, efficient coordination among AVs. It offers new algorithmic tools and theoretical insights for applying AI to solve large-scale multi-agent control problems in intelligent transportation systems
School-based group interpersonal therapy for adolescents with depression in nepal: protocol for a phase III realist cluster-randomised controlled trial
No full textBackground
Depression is a leading cause of disability among adolescents, with the burden disproportionately affecting low- and middle-income countries (LMICs) where access to mental health care is limited. Interpersonal therapy (IPT), a structured psychological intervention, has shown promise in treating adolescent depression but there is limited evidence from LMICs and research on how it works and in which contexts it works best. This protocol describes a realist cluster-randomised controlled trial (cRCT) assessing the effectiveness, cost-utility and mechanisms of school-based group IPT for adolescents with depression in Nepal.
Methods
This superiority phase III cRCT will be conducted in 48 public secondary schools across Chitwan and Nawalpur districts, with schools randomised 1:1 to intervention or enhanced usual care. Adolescents aged 13–19 with depression (Patient Health Questionnaire modified for adolescents, PHQ-A score ≥11) will be recruited from grades 7–9. The intervention comprises two individual and ten weekly group IPT sessions delivered by trained lay facilitators. Adolescents will be surveyed pre-randomisation (baseline) and five (midline), 17 (endline) and 32 weeks (follow-up) post randomisation. The primary outcome is depression severity at 17 weeks post-randomisation assessed using the PHQ-A. Secondary outcomes include anxiety, post-traumatic stress disorder, functional impairment, school attendance and quality of life. Intermediate outcomes including hope, emotion regulation, and social support will be assessed to examine mechanisms of change. A priori hypotheses concerning IPT’s mechanisms and contextual factors influencing these (context-mechanism-outcome configurations) will be refined through analysis of qualitative process data and tested in mediation, moderation and moderated mediation analyses of trial data. Economic evaluation will estimate cost-utility and benefit-cost ratios from both provider and modified societal perspectives. The process evaluation will assess fidelity, reach, and acceptability in various school settings.
Discussion
This trial is the first to integrate realist evaluation into a cRCT of a psychological intervention for adolescents in a LMIC and has potential to advance research and practice by elucidating how IPT works in a real-world context. If IPT is effective in Nepal, it could be scaled up through the education system as a part of a comprehensive school mental health care package.
Trial registration ISRCTN52852397 (registered 21/03/2025)
Online Supplement 5 Appendix for Bowers et al. (2025) Mapping Public Open Access K-12 State Education Indicator Data Across 7 States and Washington D.C. Using the FAIR Data Principles: Ohio v1.3 Metadata Megatable
No full textThis .csv is an Online Supplement for the following report published online:
Bowers, A.J., Choi, Y., Huan, Y., Huang, Y., Jiang, J., Cibrian Lopez, G., Murdoch, A., Pu, K., Sill, M., Williams, J., Wu, Y., Xu, G., Saldaña, C., Halverson, R. (2025) Mapping Public Open Access K-12 State Education Indicator Data Across 7 States and Washington D.C. Using the FAIR Data Principles. Teachers College, Columbia University, New York, NY. https://doi.org/10.7916/c0jk-5e64
This .csv includes metadata on hundreds of Ohio public open access K-12 education datasets mapped to the National Academy of Sciences, Engineering, and Medicine's (NASEM) 16 equity indicators. The metadata catalogue (the metadata megatable) includes metadata variables including each dataset's filename, format, year, direct URL, indirect URL, and categorization to each of the 16 indicators.
For more information on the variables in the file please see the data dictionary for this project: https://doi.org/10.7916/ydwn-1s20
For more information, please see the research report
Long noncoding RNAs reveal hidden genetic circuitry that drives complex inflammatory disease susceptibility
No full textThe immune system plays a central role in protecting the body against pathogen invasion, guiding development, and maintaining physiological homeostasis. When dysregulated, however, it can contribute to a wide range of common diseases, including cancer, neurodegeneration, allergies, hypersensitivity, and autoimmunity. Most of these are complex diseases, in which environmental insults interact with multiple genetic factors to produce a quantitative distribution of phenotypes across the population.
Over the past few decades, genome-wide association studies (GWAS) have provided significant insights into the intricate architecture of polygenic risk variants underlying susceptibility to complex diseases. Notably, more than 90% of these variants lie within the noncoding genome, and their mechanisms of disease causation remain largely undefined. The prevailing model assumes that noncoding variants act primarily by modulating DNA regulatory elements, which in turn regulate the nearest protein-coding genes. However, this framework largely overlooks RNA-level mechanisms.
Although only ~2% of the genome encodes proteins, at least 60% of it is transcriptionally active. Among the resulting transcripts is a recently characterized class of noncoding RNAs—long noncoding RNAs (lncRNAs)—which are pervasively expressed across the genome. Despite their abundance, the physiological roles of most lncRNAs remain unknown.
In this thesis, I employed a bioinformatic screening strategy pioneered in Dr. Sankar Ghosh’s laboratory to identify novel lncRNA genetic risk factors for autoimmune diseases—a group of incurable disorders with largely unknown etiologies. Using an integrated approach combining human samples, molecular biology, and mouse models, we identified and characterized several novel, physiologically relevant lncRNA genetic risk factors for complex diseases, in detail. These studies not only uncover new pathways of immune regulation but also reveal several regulatory logics by which noncoding SNPs cause disease, challenging the prevailing protein- and DNA-centric paradigms of genetic disease causation