|
Ph.D. Research
Dissertation summaries from Harris School doctoral candidates Carolyn Hill, Brian Jacob, and Len Lopoo appear below. What Accounts for Differences in Performance Across Welfare-to-Work
Offices? Carolyn J. Hill's dissertation research examines variations in performance across welfare-to-work offices using administrative and staff survey data from three randomized experiments of welfare-to-work programs conducted by the Manpower Demonstration Research Corporation (MDRC). Her research examines why certain programs or offices are more successful than others, what specific practices of frontline workers are associated with greater effectiveness, and how managers influence program success. To date, Hill finds that organization and management can influence program outcomes, even after controlling for observable characteristics of the clients that welfare-to-work offices serve, and for characteristics of the local economic environment. Three Essays on Education and Urban School Reform Brian A. Jacob's dissertation consists of three essays that examine current issues within education and urban school reform. The first essay examines the gender gap in college attendance (nearly 60 percent of college students today are women). Jacob uses a nationally representative cohort of eighth grade students in 1988 and finds that higher non-cognitive skills and college premiums among women account for nearly 80 percent of the gap. Jacob's second essay examines the impact of high-rise public housing on student outcomes, using a plausibly exogenous source of housing assistance variation generated by Chicago's recent public housing demolitions. Children in households affected by the demolitions are found to do no better or worse educationally than their peers and that, more specifically, the majority of families either do not take advantage of city-sponsored relocation opportunities or move to neighborhoods closely resembling those that they left. The third essay in this dissertation utilizes detailed administrative
data on student achievement in Chicago to examine the impact of a high-stakes
accountability policy implemented in 1996. Jacob finds that student achievement
on the ITBS math and reading exams increased sharply following the introduction
of high-stakes testing, but that there was little if any effect on a comparable,
low-stakes exam. He also finds that high-stakes increased retention rates
among first and second graders and decreased relative achievement in science
and social studies. Do Child Care Prices Affect Fertility? Rudimentary economic theory suggests that when something becomes less expensive, people will demand more of it. Since child care constitutes between 4 and 31 percent of the total cost of a child, depending on the opportunity cost and income of the mother, reductions in the cost of child care can dramatically decrease the total cost of a child. Len Lopoo hypothesizes in his dissertation that as the price of child care falls, therefore, people will demand more children in the contemporaenous period. Lopoo separates the child care market into two groups: family and non-family child care, then tests his hypothesis on both groups. Using the National Education Longitudinal Study of 1988, Lopoo finds that never-married teens with family members who could provide child care are between 4 and 13 percentage points more likely to have a child than those without family child care providers available. Using the Panel Study of Income Dynamics, he finds that as the availability of relative child care increases, the annual hazard rate of birth for never-married, non-teen women increases by 0.7 to 2.5 percentage points. In a state level analysis, Lopoo measures the effect of changes in the average price of child care on number of births. Using an instrumental variables model, he estimates that a 1 percent decrease in the average price of child care is associated with a 0.14 percent increase in the number of first births, a 0.19 percent increase in the number of second births, and a .11 percent increase in the number of third births. Estimates for first and second births proved to be robust to a variety of specifications.
|
