Experience in foster care is common among youth in the United States and high-quality foster care is critical to ensure their safety and long-term wellbeing. This paper studies the effectiveness of higher payments to foster care providers as a quality improvement tool. I leverage a federal court case that induced an over 20% increase in monthly payments to foster families and relatives recruited and managed by California counties, but did not extend to those recruited and managed by private agencies. Additionally, some relatives who established guardianship over a child in California were retroactively eligible for the pay raise based on the date their guardianship was established. I use a regression discontinuity design and two decades of administrative data to identify the short term effects of the court case on the safety of children in foster care, as well as its long-term effects on the safety of children in a related guardian's care. I find that higher monthly payments did not immediately affect the safety of children in the care of foster families or relatives recruited by the county, but the long-term receipt of higher payments by related guardians reduced the likelihood that a child was maltreated within five years of their guardianship start by 55% (8 percentage points). Further, despite no change in pay, the announcement of the raise for county-recruited homes immediately decreased the monthly likelihood of maltreatment while placed in private agency homes by 30% (0.16 percentage points). These findings are consistent with changes in provider effort and investments on the intensive margin and contrast with the scarce evidence to-date that suggests wages are an ineffective quality improvement tool for foster care.

Empowering adolescent girls through education has become a priority goal among multiple stakeholders, including aid agencies, non-governmental organizations, foundations, and corporations. However, recent evidence suggests that education alone may not be sufficient if women remain in a low-empowerment equilibrium and face internal constraints as they relate to aspirations, self-efficacy, leadership, and other life (soft) skills. We study the long-term impacts of a school-based upper-secondary intervention, the Educate! Experience program, designed to enhance adolescents’ leadership and social entrepreneurship skills in Uganda. The program was implemented as a cluster randomized controlled trial (RCT) in 48 schools. Four years post-intervention, we document important and lasting impacts on a wide array of leadership and soft skills.  Overall, Educate! graduates developed skills that are traditionally associated with greater focus on long-term goals; they reported being more in control of aspects of their lives (self-efficacy and grit) and more empowered to implement actions towards their plans. Young women in the treatment group are also more likely to complete secondary education, delay family formation, enroll in tertiary education, and pursue STEM and Business majors relative to their counterparts in the control group. The program yielded socially desirable gender-relevant spillovers, including expansions in women’s agency. Both male and female Educate! graduates had more (gender) egalitarian views concerning women’s standing in the society and women’s ability to exercise their agency to engage in the labor market and refuse sex. Educate! graduates’ attitudes toward and incidence of intimate partner violence also improved.

The U.S. Census Bureau’s Person Identification Validation System facilitates anonymous linkages between survey and administrative records by assigning Protected Identification Keys (PIKs) to individuals. However, previous research demonstrates that the likelihood of being assigned a PIK varies by several socioeconomic and demographic characteristics. While the potentially non-random nature of PIK assignment is generally acknowledged by users of PIK-linked data, less is known about the systematic drivers of selection into PIK assignment, the magnitude of the bias induced by this selection, and the efficacy of methods to mitigate the bias. Using the American Community Survey, we document differences in PIK assignment over nearly two decades and investigate differences in the characteristics of respondents who can and cannot be linked. We then quantify the magnitude of linkage bias in a measure of income among a sample of ACS respondents linked to IRS W-2s and evaluate the performance of statistical methods in mitigating this bias. We evaluate these methods for several demographic groups with varying PIK rates to provide practical guidance for researchers and improve inference from linked data.

Physical education (PE) is used to promote physical activity but state policies encouraging PE have demonstrated limited success in affecting health behaviors and health outcomes among youth. We study the effectiveness of a state-level policy that increased the intensity of PE by requiring at least 50% of high school PE time to be moderate-to-vigorous physical activity. Using a synthetic difference-in-differences design and eight survey waves of the Youth Risk Behavior Surveillance System, we find this policy had no overall effect on students’ physical activity levels or obesity, and reduced PE participation. The selection out of PE is larger among older, non-White students and in settings where PE is not required, suggesting that intensified PE curricula may inadvertently discourage the very students who would benefit most from such policies. We conclude that a reduction in participation is one mechanism to explain why state-level PE policies fail to improve student health when enrollment is voluntary.