This study explores the challenges that elementary science educators face when teaching science in a time of crisis, as well as how to best provide elementary teachers with ongoing support for their science teaching during the novel COVID-19 pandemic. Using a phenomenological approach, this research focuses on elementary science teachers, educators, and formal and informal leaders to understand their experiences during the pandemic and how to best support them during remote and in-person science teaching. Using data collected from questionnaires, semi-structured interviews, and focus group discussions, findings discuss the specific experiences and challenges faced by elementary science first-year teachers, early career teachers, and leaders. Following the transactional model of stress and coping (Lazarus & Folkman, 1984) and the buffering effect of social support (Cohen & McKay, 1984), first-year and early career elementary science teachers used multiple coping mechanisms to handle the stress of science teaching during the pandemic, including problem solving and collaborating with other educators. From a distributed leadership perspective (Spillane, Halverson, & Diamond, 2001b), district-level elementary science curriculum specialists and coaches act as leaders in science education. When faced with constraints and challenges due to the pandemic, these district-level leaders used this opportunity to reimagine what their leadership work could look like, including rethinking what supports they can offer classroom teachers when they cannot easily access classrooms, how to design effective science curricula for remote teaching, and how to collaborate with other educators in new ways.

Emerging infectious diseases remain persistent threats that are challenging to predict. Humanity has faced many terrible pandemics and will face more, but to pinpoint the specific time and place of an outbreak, the type of pathogen, and the consequences is effectively impossible. This point was recently highlighted by the SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) viral pandemic, which led to global clinical and socioeconomic damage. When confronted by such a viral threat, the biomedical research community fervently responded with unprecedented haste to reveal SARS-CoV-2 clinical information, genome sequences, spike fusion protein structures, antigenic properties, antiviral therapeutics, and new vaccine platforms all within a year. As a small part of the tremendous collaborative research response, I used structural methods to study the SARS-CoV-2 spike fusion protein, specifically mechanisms of antibody-mediated viral neutralization. Viral fusion proteins are key components of virus particles that enable a virus to enter an animal host cell. Fusion proteins are the most common targets for neutralizing antibodies and serve a vital role as vaccine immunogens to elicit a protective immune response. To develop an understanding of SARS-CoV-2 antibody-mediated neutralization, one of my primary research interests was solving antibody structures in complex with the spike fusion protein using cryo-EM (cryogenic electron microscopy). With antibody structures I helped characterize spike epitopes, rationalize antigenic properties of emerging variants, and hypothesize viral neutralization mechanisms. I discovered antibody structures with multiple neutralization mechanisms including receptor blocking, conformational “locking” of the RBD (receptor binding domain), and spike disassembly. Viruses are evolving pathogens, and the Omicron sub-lineages are some of the most antibody-resistant SARS-CoV-2 variants to date. I studied mechanisms of Omicron antibody neutralization, which included traditional mechanism such as receptor blocking, as well as new mechanisms involving spike disassembly and conformational locking at SD1 (subdomain 1) epitopes. I also investigated broad antibody recognition at a conserved RBD epitope which neutralized not only SARS-CoV-2 but also SARS-CoV and other sarbecoviruses. Lastly, I had the opportunity to study other classes of viral fusion proteins, including those of alphaviruses and rabies virus, which serve as representative class members of the other varieties of viral fusion proteins, broadening my research for any type of known viral pathogen. Structural studies of antibodies highlight vulnerabilities of the spike protein when targeting SARS-CoV-2 and other fusion proteins in future vaccine design. The trials and tribulations of SARS-CoV-2 and the wealth of new research on coronaviruses offer hope of future pandemic preparedness. Understanding the structural mechanisms of viral fusion proteins and antibody neutralization gives hope of developing further therapeutic interventions. The work described in this thesis on fusion proteins SARS-CoV-2 spike (S), alphavirus envelope (E), and rabies virus glycoprotein (G) have prepared me to combat other infectious viral agents, including those already infecting humans and those at risk of spilling over into humans. When posed with such unpredictable emerging threats, we can learn from the past and position ourselves to be ready for the future.

The covid-19 pandemic forged a more intensely digital world, complicating civil society actors’ menu of options for channeling and framing their advocacy goals. As both a product and study of pandemic-era politics, this dissertation is concerned with understanding how the internet and social media shape associational life in Kazakhstan. I draw on three forms of ethnographic data collected online between October 2020 and February 2022, including semi-structured interviews, visual analysis of social media posts, and digital participant observation. I demonstrate how Kazakhstani civil society actors devise strategies to pursue reform, how they debate theories of political change, and how they exercise agency in a political system that seeks to control the public sphere. I argue that civil society groups use social media platforms to leverage power differentials across levels of administration to advance rights claims and negotiate for reform. Activists and rights defenders flock to various social media platforms because of each site’s unique technological infrastructure. They leverage different logics of visibility and bridge physical and digital forms of contentious politics to demand accountability from an authoritarian government. In addition to providing a more complete understanding of civil society dynamics in Kazakhstan, this study suggests that, in repressive contexts, civil society actors who opt for within-system engagement have not necessarily been coopted and activists do not always take dissent underground. This dissertation is an example of digital political ethnography, which stands to grow not only as a standalone method, but also a bridge to big data analysis in political science. I demonstrate the importance of an ethnographic sensibility while approaching the internet as a site of inquiry to understand political subjectivity.

Despite the fact that principals have faced exceptional challenges during the COVID-19 pandemic, and play a critical role in school building and district success, particularly in times of crisis, their voices are not being heard and their needs are not being met by district-level leadership. While the literature on the COVID-19 pandemic and its impact on education is growing, it still tends to focus on pedagogy or the challenges for students, teachers, parents, and even central/district administrators, as opposed to studies that explore the plight of school-based administrators through data elicited from the viewpoint of school principals directly. Therefore, the purpose of this qualitative collective case study was to describe the lived experiences of grade K to 8 New York City public school principals leading during the COVID-19 pandemic. Specifically, within-case and cross-case analysis of combined data from in-depth individual interviews and one focus group discussion with diverse New York City public school principals (n=5), resulted in the identification of five themes evident across all cases, with each theme representing predominant patterns within principals’ self-described lived experiences of leading during the pandemic – Response to District Policy and Governance, Community, Processing Own Trauma, Resonance of George Floyd, and Concepts of Leadership and Leadership Success. Furthermore, interpretation of these findings through the lens of the study’s conceptual framework illuminated the extent to which the experience of leading during the COVID-19 pandemic is grounded in established theories of crisis management, trauma, and culturally relevant leadership.

Chinese business leaders suffered from the crisis of COVID-19 and the Double Reduction Policy, and used various strategies and learning practices to survive the crisis. The purpose of this modified exploratory multicase study was to explore how leaders in the Chinese tutoring industry made sense of the crisis of the Pandemic and the Double Reduction Policy and learned to think strategically in a time of crisis. The study not only uncovered how Chinese business leaders used different strategies to deal with a crisis and learned to think strategically while adapting to the new environment but also brought implications and insights to business leaders about effective strategies and learning practices to cultivate strategic thinking in a complex and fast-changing world. The study addressed the following four research questions: 1.How did the business leaders in tutoring companies make sense of the complexities of the crisis of COVID-19 and the Double Reduction Policy? 2.What strategies, if any, did the business leaders develop to deal with the crisis? 3.In what ways, if at all, did the business leaders learn to think strategically while dealing with challenges? 4.What other factors helped or hindered the business leaders’ learning to think strategically in a time of crisis? Qualitative semi-structured interviews (critical incident interviews included), surveys, and a focus group discussion were used to collect data from 15 Chinese business leaders from the tutoring industry. The study generated four findings: Finding 1: The crisis negatively impacted the participants and their companies at different levels, but it also served as a valuable learning opportunity for their long-term development. Finding 2: The participants developed strategies to deal with the crisis at personal, organizational, and social levels. Finding 3: The participants learned to think strategically from direct experiences, indirect experiences, and two thinking processes—systems thinking and metaphorical thinking. Finding 4: Policies and relationships were two outstanding factors that hindered or helped their learning to think strategically. These findings indicated that (1) unprecedented crises like COVID-19 and the Double Reduction Policy can bring people benefits and valuable insights, (2) business leaders can develop critical strategies by combining their own and others' experiences and strategic insights into dealing with crises, (3) business leaders can use informal learning practices and deliberately use them to cultivate their strategic thinking, and (4) business leaders need to combine learning and action for cultivating strategic thinking capabilities. Lastly, business leaders should consider the influence of policies and relationships in their strategy formulation and learning process.

Objective. The purpose of this study is to describe the effects of a food pantry program on household food security, diet and health during COVID-19 in the greater New York City area and to understand the facilitators and barriers to accessing this vital safety-net program. Methods. This study employed a three-stage design to evaluate clinical-community food pantry program, known as the Food FARMacy program, implemented to address food insecurity in New York City. Through this program three community organizations recruited participants to receive 40 pounds of fresh produce, whole grains, beans, rice and protein on a bi-weekly basis. Analysis one was a cross-sectional analysis of baseline data to understand food security, diet, and health in those registering for the Food FARMacy program. Analysis two was a longitudinal pre-post analysis comparing baseline data with 6-month follow-up data to determine the effects of food pantry participation on food security, diet, and health. Analysis three was a qualitative case study with program participants to understand their experience participating in the program, including key facilitators and barriers to participating in a food pantry program during the COVID-19 pandemic. Data Analysis. For analysis one, descriptive statistics were used to report demographic, food security, diet and health characteristics upon program enrollment. X² tests and independent t-tests as well as multivariable regression models were used to examine predictors of very low food security status and food security score at enrollment. For analysis two, Wilcoxon signed rank and McNemar’s tests were used to identify changes in food security, diet, and health from baseline to six-months follow-up. Regression models were built to examine the association between attendance and food security status. For analysis three, a subset of 24 participants were interviewed using a semi-structured interview format to understand their lived experience with the program and barriers and facilitators to participating. Results. Through this program, 492 participants were enrolled from July 2020 to April 2021 and provided with fresh, healthy food and beverages on a twice monthly basis. The majority of the enrollees reported low (42.3%) or very low (45.5%) food security status. At 6-months follow-up, the percent of those reporting very low food security status improved significantly from 45.5% to 13.2% (p < .001). Further, fruit intake two or more times per day increased from 23.7 to 35.1%, and the percent of those reporting no fruit intake decreased from 36.6 to 15.4% (p < .001). Vegetable intake two or more times in the previous day also increased from 21.5 to 41.8%, with the percent of those consuming no vegetables in the previous day declining from 32.6 to 13.2% (p < .001). The percent drinking two or more SSBs in the previous day decreased from 23.1 to 9.5% (p < .001). The percent of participants reporting excellent, very good or good health increased from 52.3 to 60.0%, while the percent reporting fair or poor health decreased from 48 to 40% from baseline to six-months follow-up (p = .017). Qualitative analysis revealed that participants valued the fresh, high-quality food that they could prepare themselves and caring customer service provided through the program. Transportation and access to childcare were reported as intermittent barriers to accessing the pantry program. Overall, participants reported very positive experiences with the program and improvements were noted in food security, diet, and health from baseline to 6-months follow-up. Conclusions. Effective and sustainable solutions are needed to curb household food insecurity. Rapid development and implementation of an emergency food pantry program through an integrated healthcare system and community organization partnership was feasible and effectively reached high-need patients and community members. Pantry programs can be an effective mechanism for addressing disparities in food access and diet among vulnerable populations.

The outbreak of the COVID-19 pandemic found a fertile ground for the spread of online misinformation, with emphasis on social media. Avoiding misinformation spread requires rapid, engaging, and effective science communication in a clear, easy-to-understand, attractive, and entertaining format that can be readily shared online. Comics fulfill these characteristics, being a promising tool to fight misinformation on social media. The goals of this study were: 1) Develop a novel narrative comic strip to promote recognition of misinformation about the COVID-19 vaccine among adult social media users (ages 18-65) based in the United States, drawing on the existing research on the Health Belief Model and Theory of Planned Behavior; 2.) Compare the comic strip evaluation and capacity to influence misinformation identification to those of an educational text about COVID-19 vaccination. 3a) Evaluate differences in the key outcomes (misinformation identification, and attractiveness, trust, perceived usefulness, willingness to share, and acceptance of each educational tool) across participants with varying demographic characteristics, health literacy levels, COVID-19 vaccination history, and demographic characteristics. 3b) Across the entire sample, evaluate the correlation between these constructs and health literacy, digital health literacy, vaccine attitudes, trust in science and health authorities, and social media use. Participants (N = 285) were recruited via social media advertisements and randomly assigned to the comic strip group (CS) (N = 92), educational text (TX) (N = 96), or a control 4 group (CL) (N = 97), which had not read any educational material. An online survey accessed the main outcomes (misinformation about the COVID-19 vaccines, evaluation of the educational tool (attractiveness, trust, perceived usefulness, willingness to share, and acceptance of the educational material). Participants also answered demographics questionnaires, COVID-19 vaccine concerns scale, and questionnaires on Health literacy, eHealth literacy, social media use, trust in health authorities and scientists, and COVID-19 vaccination history. Group CS answered questions regarding transportation into the narrative. There were no differences in misinformation identification between groups, possibly explained by a low sensibility of the misinformation identification instrument, timing of the data collection, and sensitiveness of the vaccination topic, subject to accrued attitudes, such as believing in misinformation. Participants with lower health literacy in group TX scored less on the misinformation identification questionnaire than those with higher literacy, which was not observed in the CS group, indicating that the comic strip may benefit better individuals with low health literacy. Vaccine hesitant/ refusers’ misinformation identification scores seem to have been benefited by the comic strip. The comic strip was better evaluated for trust in its content and acceptance than the educational text. Still, misinformation identification scores were not correlated to any evaluation construct in both groups CS and TX. Transportation into the narrative was positively correlated with all comic strip evaluation constructs but not with the misinformation identification score. Future studies should focus on exploring different styles and sizes of comic strips, using more heterogenous sample and addressing different health topics.

Human decisions are simultaneously determined by economic incentives and psychological motivations. Based upon this fundamental assumption, I compose three interdisciplinary studies which analyze individual, collective and government actions at multiple levels of aggregation, and how they in turn lead to various economic and psychological outcomes. In the first study, Iexplore the key predictors of the level of compliance to social distancing and mask wearing in the United states by aggregating interdisciplinary datasets and applying multi-level analysis. I use a behavioral model to classify the determinants of compliance to COVID-19 response measures into economic incentives and psychological motivations and show that the former would have an increasing marginal effect on working hours. Empirically, I show that (a) economic vulnerability was the key predictor of failure of social distancing in 2020, even taking partisanship into account. (b) mask wearing was more politicized than social distancing, and in Fall (close to the elections), Republican partisanship was the only dominant indicator of noncompliance of mask wearing. In the second study, we use a coordination game model to discuss the dynamics of Non-Pharmaceutical Interventions (NPIs) on COVID-19 in the United States. We use atheoretical model to justify that there exist social reinforcement effects between policies in US states, i.e. the implementation of an NPI in a state would increase the possibility that others follow suit. Under certain conditions, if enough states engage in NPIs, they will tip others that have not yet done so to follow suit and thus shift the Nash equilibrium to the greatest one (allstates follow). Then, we show that there can be equilibria where states with different political leanings adopt different strategies when politics is a determinant of the interaction intensity. Empirically, we use a random utility model (RUM) to test it in reality with Probit and Logit regressions, and find robust evidence that inter-state social reinforcement is important and that equilibria can be tipped in mask wearing, and slightly weaker confirmation for social distancing. In the last study, I explore how personality traits in China are different from the traditional Five-Factor model by a large twin dataset in Yunnan Province. I find robust evidence about personality structures, formation and impacts in China and state three findings: (1) Personality traits in China seem to have a significance deviation from the well-accepted Five Factor Model. Instead, it has two general factors, relying on whether the item is positive or negative in tone. Positive factors include Social Desirability, Extraversion and Openness; negative factors include Disorderliness, Neuroticism and Introversion. (2) The genetic heritability of personality traits in China is significantly lower than that measured in the Western countries. For some traits, such as Social Desirability and Disorderliness, the genetic effect is around 0 and the shared environmental effect is much larger. This challenges previous findings in the West. (3) Using a within-twin fixed effect model, we find suggestive evidence on the causal effect on economic preferences and outcomes, including education performance, income, risk attitudes and subjective well-being.These three studies use the similar behavioral science methodology to study different levels of decision making, and all have important implications for issues of sustainable development.

My dissertation focuses on investigating the associations of neighborhood environments—namely, built and social environments—with health among community-dwelling older Americans. The first paper examines groupwide variations in social participation patterns among older adults before and during the COVID-19 pandemic, and if community social cohesion and health during the pandemic were significantly associated with social participation patterns. Using Rounds 9 and 10 longitudinal data from the National Health and Aging Trend Study, I employed latent class analysis to identify the presence of groupwide variations in social participation, before and during the pandemic. I used logistic and linear regressions to examine the associations between social participation patterns, community social cohesion, and health during the pandemic. Results suggested two participation patterns, active and selective participants. Compared to active participants, older adults who were selective in their social participation were more likely to live in less socially cohesive communities and report substantial depressive and anxiety symptoms. In the second paper, I examined cross-sectional and longitudinal relationships between neighborhood physical disorder, low social cohesion, and sleep problems among older Americans. Mediators of health behaviors (i.e., lack of physical activity and social participation) and mental health (i.e., depressive and anxiety symptoms) were also tested in these relationships. Data were derived from three rounds of panel data (Rounds 7-9) from the National Health and Aging Trends Study, involving a sample of 4,029 Americans aged 65 or older. I found statistically significant cross-sectional and longitudinal associations between physical disorder and low social cohesion, and late-life sleep problems. Only cross-sectional mediation effects of health behaviors and mental health were found in the relationship of physical disorder and sleep problems, whereas both cross-sectional and longitudinal associations between low social cohesion and sleep problems were significantly mediated by health behaviors and mental health. In the third paper, I used annual data from the 2015-16 Poverty Tracker study to examine the roles of distance to grocery stores, neighborhood disadvantage, and social cohesion in explaining food insecurity among older adults in New York City. Multiple logistic regressions were conducted to assess these relationships. Results showed that greater distance to grocery stores (0.26–0.75 miles vs. 0.00–0.25 miles) and living in more disadvantaged neighborhoods increased the odds of food insecurity. Community social cohesion was a marginally significant protective factor against food insecurity. The findings of these papers highlighted the associations between the neighborhood environment, social health, sleep quality, and food security status among older adults. These papers also emphasized the potential for environmental policy and social work program interventions to improve the well-being and quality of life among community-dwelling older adults.

This dissertation addresses three problems of applied statistics involving discrete time series and network data. The three problems are (1) finding and analyzing community structure in directed networks, (2) capturing changes in dynamic count-valued time series of COVID-19 daily deaths, and (3) inferring the edges of an implicit network given noisy observations of a multivariate point process on its nodes. We use tools of spectral clustering, state-space models, Bayesian hierarchical modeling and variational inference to address these problems. Each chapter presents and discusses statistical methods for the given problem. We apply the methods to simulated and real data to both validate them and demonstrate their limitations. In chapter 1 we consider a directed spectral method for community detection that utilizes a graph Laplacian defined for non-symmetric adjacency matrices. We give the theoretical motivation behind this directed graph Laplacian, and demonstrate its connection to an objective function that reflects a notion of how communities of nodes in directed networks should behave. Applying the method to directed networks, we compare the results to an approach using a symmetrized version of the adjacency matrices. A simulation study with a directed stochastic block model shows that directed spectral clustering can succeed where the symmetrized approach fails. And we find interesting and informative differences between the two approaches in the application to Congressional cosponsorship data. n chapter 2 we propose a generalized non-linear state-space model for count-valued time series of COVID-19 fatalities. To capture the dynamic changes in daily COVID-19 death counts, we specify a latent state process that involves second order differencing and an AR(1)-ARCH(1) model. These modeling choices are motivated by the application and validated by model assessment. We consider and fit a progression of Bayesian hierarchical models under this general framework. Using COVID-19 daily death counts from New York City's five boroughs, we evaluate and compare the considered models through predictive model assessment. Our findings justify the elements included in the proposed model. The proposed model is further applied to time series of COVID-19 deaths from the four most populous counties in Texas. These model fits illuminate dynamics associated with multiple dynamic phases and show the applicability of the framework to localities beyond New York City. In Chapter 3 we consider the task of inferring the connections between noisy observations of events. In our model-based approach, we consider a generative process incorporating latent dynamics that are directed by past events and the unobserved network structure. This process is based on a leaky integrate-and-fire (LIF) model from neuroscience for aggregating input and triggering events (spikes) in neural populations. Given observation data we estimate the model parameters with a novel variational Bayesian approach, specifying a highly structured and parsimonious approximation for the conditional posterior distribution of the process's latent dynamics. This approach allows for fully interpretable inference of both the model parameters of interest and the variational parameters. Moreover, it is computationally efficient in scenarios when the observed event times are not too sparse. We apply our methods in a simulation study and to recorded neural activity in the dorsomedial frontal cortex (DMFC) of a rhesus macaque. We assess our results based on ground truth, model diagnostics, and spike prediction for held-out nodes.