BACKGROUND. TB kills 1.5 million people each year, more than any other single infection. There is an urgent need to evaluate the impact of new interventions to strengthen TB control. Poverty is increasing globally in cities and urban areas, and is associated with factors that increase TB risk including crowding and malnutrition. Conversely, TB worsens poverty by increasing expenses and reducing income. In addition, those with TB and their families may experience stigma. Poor people have more TB and greater TB-related needs but they tend to have least access to TB care. This mismatch between need for and access to TB care undermines TB control and worsens poverty. OBJECTIVE. We will evaluate the impact of socioeconomic interventions for reducing poverty, improving access to TB care and consequently reducing the risk of future TB. SETTING. 24 peri-urban shantytowns in Northern Lima, Peru near the site of our TB control research since 2001. Peru has an acclaimed TB control programme but its levels of TB disease remain high and its rates of multidrug-resistant TB (MDR-TB) have doubled over the last decade to the highest levels in the Americas. The high TB and increasing MDR-TB rates are concentrated in "hotspots" such as poor peri-urban shantytowns surrounding Lima. It is these "hotspots" where we will work. RELATED WORK. Since 2007, our on going pilot project has been "Innovative Socio-economic Interventions Against TB (ISIAT)" which involves developing and implementing socioeconomic interventions to fight poverty and increase equitable access to TB care. Early analysis of the pilot ISIAT project showed promising results with the interventions described below increasing the number of people to a) complete TB treatment b) complete preventive therapy to prevent them getting TB c) be tested for TB and d) be tested for HIV. These results were published in 2011. The improvement in awareness, prevention and treatment of TB that our work and subsequent article showed has attracted the attention of funding bodies, like the World Bank and The Bill and Melinda Gates Foundation, and policymakers such as the World Health Organisation (WHO) and its Stop-TB department. Our on going relationship and involvement of these organizations and the published results of the pilot ISIAT project are encouraging for future work. The proposed project will rigorously assess the impact of these interventions not just on poverty and access to TB care but also on actual TB control. INTERVENTIONS AND STUDY DESIGN. The interventions will be inexpensive, involving a team of experts from different fields working with all TB-affected families. They will utilize household visit and fortnightly community meetings to implement an integrated program of social support for enhancing equitable access to TB-related healthcare and economic support to help people to afford TB care and to help them to become less poor. We will include in the study all members of a household with a new diagnosis of TB, as assessed by the Peruvian national TB program, without age limit, who provides informed written consent. TB-affected households in 12 intervention communities will be offered the socioeconomic intervention for 6-months whilst the TB patient is receiving TB treatment. TB-affected households in 12 other control communities will be offered no intervention (standard of TB care). We will then re-visit those households 2 years after recruitment to assess what happened to the people with TB and those people in the household exposed to TB, and if our interventions prevented TB and reduced poverty. BENEFITS: These socio-economic 'structural' interventions will be assessed for their capacity to reduce poverty-related TB risk factors, improve access to TB care and for reducing TB treatment failure, recurrence and transmission. This has potential importance for focusing poverty reduction on those in greatest need and who have the most to benefit and preventing TB.

International trade is of vital importance for modern economies, and governments around the world try to shape their countries' exports and imports through numerous interventions. Given the problems facing trade negotiations through the World Trade Organization (WTO), countries have increasingly turned to preferential trade agreements (PTAs) involving only one or a small number of partners. At the same time, attention has shifted from reductions of import tariffs to the role of non-tariff barriers such as differences in regulations and technical standards. Accordingly, modern PTAs contain a host of provisions besides tariff reductions, in areas as diverse as services trade, competition policy or public procurement. A key question in international trade research is how to estimate the effects of PTAs and their individual provisions on trade flows. We argue that methods from the machine learning literature can help address this challenge, and that such methods are often superior to existing approaches. We use the term 'machine learning' to refer to algorithms used for statistical prediction that are trained on subsets of the available data to make forecasts of quantifiable outcomes (here: trade flows). While such algorithms have started to be applied in economic research, they have not been used for the analysis of PTAs nor in international economics more generally. First, machine learning can help evaluate the suitability of existing methods for estimating PTA effects. Such methods evaluate PTAs by comparing the trade flows observed after the implementation of an agreement to a so-called counterfactual outcome that shows what would have happened to trade flows in the absence of a PTA. This counterfactual is invariably based on a specific statistical model. Currently, by far the most common model is the so-called gravity equation. The estimated effect does of course depend on how well the gravity equation predicts counterfactual trade flows. We will use machine learning to develop a more flexible forecast to which we can compare the gravity equation's predictive power. Machine learning can also help improve existing methods for PTA evaluation. Implicitly, approaches based on the gravity equation construct a counterfactual by using an average of the changes in trade flows between countries not involved in a PTA. Similar approaches have been applied in a range of contexts besides international trade. Recent methodological advances have shown how these approaches can be improved by applying machine learning to select more complex combinations of control units (here: countries not participating in a PTA) than simple averages. Despite their potential, these techniques have not been applied in international trade research, and we propose to adapt them to this context. Finally, machine learning can be used to determine the relative importance of individual PTA provisions. The key challenge existing research has faced is that many PTAs contain similar provisions, making it difficult to estimate their effect on trade flows separately. Thus, researchers usually aggregate provisions in some way, for example by combining them into broad groups. This limits the relevance to policymakers who need to know if they should include a given individual provision in a PTA. This problem is reminiscent of the issue of 'feature selection' in machine learning where algorithms must decide which of many potentially relevant variables to include for forecasting purposes. We plan to use a subgroup of these methods that allow to identify the subset of variables (here: provisions) with the largest effect and to accurately estimate their impact. Overall, the proposed research will deepen our understanding of how PTAs impact trade flows. This, and the empirical techniques we plan to develop, will help researchers and policymakers involved in the design and evaluation of PTAs and ultimately contribute to a better, more evidence-based trade policy.

In my fellowship I have conducted groundbreaking research and spearheaded innovative initiatives, leading a team to explore and devise technological enhancements for democratic innovation. Employing a mix of artificial intelligence, design methodologies, and social science, I've amalgamated data, and analysed and compared cases to draw out lessons for democracy. I have designed software using participatory methods to craft technologies and interventions that prioritise human experience and transparency in democratic procedures. This project is driven by the aspiration of empowering citizens, bolstering the safety of communities, and heightening confidence in societal institutions through understanding which interventions build and fortify democratic activities. It involves making space for often marginalised voices, curbing hateful speech, and ensuring diverse community participation in public discourse and in governance. My focus lies in the creation of interdisciplinary data science tools that allow for analysis of political text, speech, and video. The approach involves constructing advanced models that answer pertinent research questions and navigate the ethical and social consequences of deploying algorithms in democratic engagement. The project team is committed to designing and engineering democracy software that transcend temporary successes, with an emphasis on adapting interventions to a variety of governance, workplace, and community settings. We aim to build knowledge about how techniques can be used wherever we need to make collective decisions, and deal peacefully and fairly with disagreement. We will develop AI interventions that augment existing democratic innovations, that are pragmatic and applicable in day-to-day contexts. We aspire to understand how these innovations can be sequenced optimally for transparency, trust, and mutual oversight. This involves carrying out randomised controlled trials and employing other research methods to ascertain when and how to step in to restore trust in institutions, and consequently, enhance the governance of emerging technologies. The fellowship supports me to lead a multidisciplinary research and innovation hub in the UK, one that is fervently devoted to preserving and enriching democracy. This project aims to deliver novel research involving many scientific disciplines contributing to a flourishing economy and society in the UK and beyond.

Worldwide, there has been growing interest in understanding the nature of quality education. A major key to this quest lies in what goes on inside classrooms, where children derive the bulk of their daily experiences in academic and social learning. While factors like the physical condition of the school building, textbooks, and teacher degrees play a role in children's learning and life outcomes, they are small and indirect. Teacher instructional practices and classroom processes, in terms of supportiveness and organization, play considerable roles in children's learning and well being outcomes. Yet, the focus of many attempts to improve (and evaluate) educational programs has been based on classic, though simple, input-output model. In other words, an intervention takes place, and then the change in child academic or social outcomes are measured. Studies of this type can be viewed as "black box" studies; they tell us little more than whether the program worked or not. They fail to provide us with insights on how to more effectively facilitate deeper learning. To do this, we first need to be able to effectively measure instructional practices and classroom processes. The most accurate way of measuring instructional practices and classroom processes is with the use of observational methods. To date, available methods have been too labor-intensive and costly for large-scale evaluation studies or for use in daily practice. Reliable, valid, cost-effective, and practically useful tools are needed. Nowhere is this truer than in low-income and fragile states. This is the goal of the proposed investigation. To achieve these ends, we capitalize on a large-scale experimental school and classroom-based intervention program undertaken in Ugandan public secondary schools by the World Bank (WB), in partnership with the Ministry of Education and Sports (MoES). In a second phase of this project, the WB enlisted New York University (NYU) to supplement the impact evaluation by examining the instructional practices and classroom processes with live observations using an innovative tool, known as TIPPS, before, in the middle, and at the end of the intervention year. Samples of these classrooms are also videotaped for more intensive analysis. This data provides a unique opportunity to further develop and validate an innovative, affordable, scalable, and practically useful tool for assessing teacher practices and classroom processes. It also has the potential to provide feedback to teachers, especially when used in tandem with mentoring and reflected practice for improved teacher performance. We conduct a series of scientific studies to assure the viability, validity, and utility of this instrument. In addition to the development and validation of an effective classroom observational instrument, we want to assure its use in policy and practice in Uganda and eventually in other low-income and fragile states. Thus, we begin the project year by working closely with the various stakeholder groups - ministry, union officials, school administrators, teachers, and World Bank Africa Region staff - to facilitate buy-in and ownership. We will engage them in interviews and focus groups to both inform them about the instrument and to gain their assistance in structuring the end of the year workshops for maximum effectiveness. The goals of these workshops are to explain our findings with regard to the intervention and the tool, and more importantly, so that the tool can be implemented at policy levels by the ministry, with the aide of the unions. In this manner, this tool could then be put into practical use in secondary schools around the country, and eventually primary schools as well.

The COVID-19 pandemic and related social and economic crises are undermining children's education in low- and middle-income countries through school closures, unequal access to remote-learning activities, and increased household food insecurity and poverty. Groups at greater risk, including girls and children from the poorest families, are likely being disproportionately affected, amplifying existing inequalities in child education, health and broader development. We embed in an ongoing longitudinal project, Quality Preschool for Ghana, a study of the pandemic's repercussions on children's education and broader development for a representative sample of urban Ghanaian boys and girls aged 10-12 years (N=~2,000), their households, and teachers (N=~400). We have four main goals. First, we investigate household and child vulnerability and resilience to the crisis, with three phone surveys with parents and one phone survey with children starting in late summer, followed by already-funded child and parent direct assessments later in the 2020-2021 school-year. Second, with three additional phone surveys with teachers, we generate new data on how children, parents and teachers are faring with the remote-learning implemented during school closures and with re-entry into in-person schooling should that happen in the 2020-21 school year. Third, by piggy-backing on already-funded data collection activities planned for later in the Fall 2020 and Spring 2021, and combined with four prior rounds of data on these children starting in preschool, we examine inequalities in the effects of the crisis on learning and broader child development domains (health, psycho-social outcomes). Fourth, we monitor changes in poverty and food security and examine their associations with later-in-life children's educational outcomes. The proposed study provides the Ghanaian government with unique, real-time data to inform remote-learning, school-reentry, how children, families and teachers are coping with the crisis, and social-protection efforts. Results will provide timely and much-needed academic and policy insights for Ghana and broader global educational efforts to protect children from the long-term effects of the pandemic on their learning and development.