Research Interests:
My research interests are in economics of education and labor economics as well as other applied microeconomics fields. My education research interests are in school choice and market-based education reforms, policies affecting the transition from high school to college, and how education policies affect non-academic outcomes. Please read the abstracts and linked papers below for more information.
Publications:
“K-12 and Postsecondary Alignment: Racial/Ethnic Differences in Freshmen Course-taking and Performance at California’s Community Colleges,” (with Michal Kurlaender), Education Policy Analysis Archives, North America, 21(16), February 2013
(Publication Link) featured in InsideHigherEd.com, February 2013
Abstract: In this paper we focus on California high school students' transition to community colleges. Our unique dataset tracks five cohorts of California high school juniors into their freshman year at in-state community colleges. We evaluate the extent to which high school achievement tests (currently not utilized by community colleges in course placement decisions) are useful for predicting academic success at community college. In addition, given persistent disparities in college attainment by race, we explore whether this fundamental relationship in high school achievement, college course-taking, and performance differs for students from different racial/ethnic backgrounds.
"Experiments and Quasi-Experiments," (with Douglas N. Harris), in Brewer, D.J. & Picus, L.O., ed. Encyclopedia of Education Economics and Finance, Forthcoming 2014
Working Papers
"Does Closing Schools Close Doors? The Effect of High School Closures on Achievement and Attainment," Job Market Paper
(PDF Copy)
Abstract: The idea that low-performing schools should be closed, either through market competition or government intervention, is now a central tenet of state and federal school reform efforts, yet little is known about the impacts of these closures. Most previous studies examine the effects of elementary school closings on test scores. This study furthers the literature by focusing on high school closures and examining several measures of both achievement and attainment. I utilize student level data from the Milwaukee Public School district and follow five freshman cohorts (2005-06 to 2009-10) as they progress through high school. During this time there were 33 school closures affecting these cohorts. I find that, on average, school closings cause a negative shock to students, lowering both their GPA and attendance. There is evidence that students bounce back over time, however, for many students, high school ends before the effects can completely be reversed. These closures also have long-run consequence both lowering the probability of high school graduation and college attendance.
“High Bars or Behind Bars? The Effect of Graduation Requirements on Arrest Rates,” Revisions Requested at Economics of Education Review
(PDF Copy)
Abstract: This paper investigates the effect of high school graduation requirements on arrest rates. Many states have altered their high school curriculum in an attempt to improve the human capital acquired by their high school graduates, but these changes may lead to externalities that have not been appreciated. I examine changes in the arrest rates of young people following state level increases to the minimum number of required courses and the implementation of exit exams. Identifying variation comes from changes in state laws governing high school graduation requirements from 1980 to 2000. By utilizing repeated cross section arrest data I estimate the effects of across state-cohort differences in graduation requirements on arrest rates. While pinpointing the exact mechanism is not possible, there are signs that point to both human capital effects that lower arrest rates and dropout effects that increase arrests due to the increased rigor of the requirements. I find evidence that using less difficult exit exams can reduce the arrest rate by 4.4% suggesting that the human capital effect may dominate the dropout effect on average. This effect is strongest on property crimes and in the lowest-income counties.
“War and Marriage: Assortative Mating and the World War II GI Bill,” (with TJ McCarthy, Jeremy Moulton, Marianne Page, and Ankur Patel), Revisions Requested at Demography
(PDF Copy)
Abstract: World War II and its subsequent G.I. Bill have been widely credited with playing a transformative role in American society, but there have been few quantitative analyses of these historical events’ broad social effects. We exploit between-cohort variation in the probability of military service to investigate how WWII and the G.I. Bill altered the structure of marriage, and find that it had important spillover effects beyond its direct effect on men’s educational attainment. Our results suggest that the additional education received by returning veterans caused them to “sort” into wives with significantly higher levels of education. This suggests an important mechanism by which socio-economic status may be passed on to the next generation.
“What Schools Do Parents Want (and Why)? Academics, Extracurricular Activities, and Indirect Costs in New Orleans Post-Katrina School Reforms,” joint with Douglas N. Harris and Jill Zimmerman
(PDF Copy - Coming Soon!)
Abstract: Policymakers have been trying for decades to improve schools through market-based policies. While these have met with mixed results, no city had ever adopted a comprehensive market system until New Orleans did in the wake of Hurricane Katrina. Attendance zones were eliminated, parents were given the opportunity to make school choices, and almost all schools were turned into charter schools that offered differentiated services. We leverage these changes to understand parental preferences and how they may have changed as a result of the market reforms. Contrary to the large literature based on stated preferences, our results suggest that the role of extracurricular activities and indirect costs such as distance are at least as large as academics. Also, while all New Orleans families gained greater access to schools with higher test scores, the lowest-income families have weaker preferences for academic outcomes and are more constrained in their ability to choose the highest-performing schools. We also examine how measured preferences have changed over time in somewhat predictable ways. In addition to helping understand the demand for schools generally, the somewhat limited influence of academics in family decisions helps explain several puzzles from prior research, including why the competitive effects of market-based school reforms on student test scores have been mixed.
“Testing the Validity of Across Cohort Quasi-Experiments Using a Cluster-Randomized Control Trial of an Early College Scholarship,” joint with Doug Harris
(Copy not Publicly Available)
Abstract: Randomization is often considered the surest way to identify unbiased estimates of treatment effects, but often only quasi-experimental methods are feasible and there is debate about whether and when they are likely to yield biased estimates. A growing number of “within-study comparisons” suggest that propensity score matching often does not yield the same conclusions as randomized control trials (RCTs), while regression discontinuity designs do generally yield the same conclusions. In this paper, we study an early college scholarship and compare the RCT identification with difference-in-differences (DD) of the same intervention. While there are conflicting conclusions for about half of the outcomes, the exogeneity assumption is rejected for all of them, so we would not generally trust these DD estimates. This implies that, by the usual standards, we would reach the same general conclusions regardless of whether we used an RCT or a DD for this intervention. The similarity in conclusions occurs despite the over-rejection of the null when the DD involves a small number of groups by Donald and Lang (2007).
“The Effect of High School Graduation Requirements on Teenage Birth Rates and Infant Health”
(PDF Copy)
Abstract: This paper analyzes possible spillover effects of raising high school graduation requirements. Using Natality data, I compare the birth rates and other teen pregnancy outcomes for state/cohorts with more difficult graduation requirements to those with less difficult requirements. I find evidence that more requirements reduces teenage birth rates, where each additional course required decreases the birth rate by 1% and 0.7% for white and black mothers respectively. this effect is focused in low income counties, which may be the counties most in need of assistance. I find no effects of more difficult graduation requirement on a variety of other teen pregnancy outcomes including birthweight and premature births. Results are favorable for increasing graduation requirements, given that I find no negative effects of these requirements.
Works in Progress:
"Heterogenous Preferences in School Choice," (with Douglas N. Harris)
Abstract: Following Hurricane Katrina, New Orleans eliminated school attendance zones and almost all schools were converted to charter schools. Harris, Larsen, Zimmerman (2014) examines what school characteristics the average parents care most about when choosing a school. In this paper, we examine how those choices differ across families based on racial, socioeconomic, and prior academic differences. Do parents with special needs children look for schools with different services than those without special needs children? Do children who have had discipline issues in the past choose schools with a different disciplinary philosophy? Do students with low test scores choose schools with high or low average test scores or high? High or low Value-Added? We use a unique dataset which connects students' prior academic history to their ranking of schools on a centralized district school application to answer these questions.
"From One to Many: The Effect of Requiring Course Specific Graduation Exams"
Abstract: Previous research has shown that the use of graduation exams can reduce the probability of graduation. Comprehensive graduation exams are often taken early in high school and yield several re-test opportunities for those students who fail. These tests often taken in the Spring, test for comprehensive knowledge of the core topics taken early in high school. Recently, many states have opted to switch to "End of Course" exams, which are taken at the end of each course and only test on the material within that course. Beginning with the freshmen cohort of 2010-11, Louisiana moved from requiring a comprehensive graduation exam to multiple end of course tests. While students now will take exams that are more focused on certain subjects, they are required to pass 3 of 6 exams some of which are taken as late as 11th grade. I use regression discontinuity analysis to estimate what effect each regime has on the graduation rate of students at the margin of passing. I also describe how the average characteristics of the marginal students differ under each regime.
“The Effect of School Investment on Juvenile Crime,” (with Danielle Sandler)
Abstract: It has long been argued that investment in education and schools can help reduce crime. We seek to estimate if that relationship holds true for school facility investments. Updating and renovating schools can potentially improve education and morale of students. Investments in gyms and sports courts may provide an outlet for neighborhood kids who may turn to crime and mischief otherwise. Finally, the simple act of investing in schools may demonstrate that a neighborhood is invested in itself and deter crime from occurring. Following Cellini, Ferreira, and Rothstein (2010) we use a regression discontinuity in voting thresholds yielding approval for school facility investments in California to estimate how those investments affect local crime rates.
My research interests are in economics of education and labor economics as well as other applied microeconomics fields. My education research interests are in school choice and market-based education reforms, policies affecting the transition from high school to college, and how education policies affect non-academic outcomes. Please read the abstracts and linked papers below for more information.
Publications:
“K-12 and Postsecondary Alignment: Racial/Ethnic Differences in Freshmen Course-taking and Performance at California’s Community Colleges,” (with Michal Kurlaender), Education Policy Analysis Archives, North America, 21(16), February 2013
(Publication Link) featured in InsideHigherEd.com, February 2013
Abstract: In this paper we focus on California high school students' transition to community colleges. Our unique dataset tracks five cohorts of California high school juniors into their freshman year at in-state community colleges. We evaluate the extent to which high school achievement tests (currently not utilized by community colleges in course placement decisions) are useful for predicting academic success at community college. In addition, given persistent disparities in college attainment by race, we explore whether this fundamental relationship in high school achievement, college course-taking, and performance differs for students from different racial/ethnic backgrounds.
"Experiments and Quasi-Experiments," (with Douglas N. Harris), in Brewer, D.J. & Picus, L.O., ed. Encyclopedia of Education Economics and Finance, Forthcoming 2014
Working Papers
"Does Closing Schools Close Doors? The Effect of High School Closures on Achievement and Attainment," Job Market Paper
(PDF Copy)
Abstract: The idea that low-performing schools should be closed, either through market competition or government intervention, is now a central tenet of state and federal school reform efforts, yet little is known about the impacts of these closures. Most previous studies examine the effects of elementary school closings on test scores. This study furthers the literature by focusing on high school closures and examining several measures of both achievement and attainment. I utilize student level data from the Milwaukee Public School district and follow five freshman cohorts (2005-06 to 2009-10) as they progress through high school. During this time there were 33 school closures affecting these cohorts. I find that, on average, school closings cause a negative shock to students, lowering both their GPA and attendance. There is evidence that students bounce back over time, however, for many students, high school ends before the effects can completely be reversed. These closures also have long-run consequence both lowering the probability of high school graduation and college attendance.
“High Bars or Behind Bars? The Effect of Graduation Requirements on Arrest Rates,” Revisions Requested at Economics of Education Review
(PDF Copy)
Abstract: This paper investigates the effect of high school graduation requirements on arrest rates. Many states have altered their high school curriculum in an attempt to improve the human capital acquired by their high school graduates, but these changes may lead to externalities that have not been appreciated. I examine changes in the arrest rates of young people following state level increases to the minimum number of required courses and the implementation of exit exams. Identifying variation comes from changes in state laws governing high school graduation requirements from 1980 to 2000. By utilizing repeated cross section arrest data I estimate the effects of across state-cohort differences in graduation requirements on arrest rates. While pinpointing the exact mechanism is not possible, there are signs that point to both human capital effects that lower arrest rates and dropout effects that increase arrests due to the increased rigor of the requirements. I find evidence that using less difficult exit exams can reduce the arrest rate by 4.4% suggesting that the human capital effect may dominate the dropout effect on average. This effect is strongest on property crimes and in the lowest-income counties.
“War and Marriage: Assortative Mating and the World War II GI Bill,” (with TJ McCarthy, Jeremy Moulton, Marianne Page, and Ankur Patel), Revisions Requested at Demography
(PDF Copy)
Abstract: World War II and its subsequent G.I. Bill have been widely credited with playing a transformative role in American society, but there have been few quantitative analyses of these historical events’ broad social effects. We exploit between-cohort variation in the probability of military service to investigate how WWII and the G.I. Bill altered the structure of marriage, and find that it had important spillover effects beyond its direct effect on men’s educational attainment. Our results suggest that the additional education received by returning veterans caused them to “sort” into wives with significantly higher levels of education. This suggests an important mechanism by which socio-economic status may be passed on to the next generation.
“What Schools Do Parents Want (and Why)? Academics, Extracurricular Activities, and Indirect Costs in New Orleans Post-Katrina School Reforms,” joint with Douglas N. Harris and Jill Zimmerman
(PDF Copy - Coming Soon!)
Abstract: Policymakers have been trying for decades to improve schools through market-based policies. While these have met with mixed results, no city had ever adopted a comprehensive market system until New Orleans did in the wake of Hurricane Katrina. Attendance zones were eliminated, parents were given the opportunity to make school choices, and almost all schools were turned into charter schools that offered differentiated services. We leverage these changes to understand parental preferences and how they may have changed as a result of the market reforms. Contrary to the large literature based on stated preferences, our results suggest that the role of extracurricular activities and indirect costs such as distance are at least as large as academics. Also, while all New Orleans families gained greater access to schools with higher test scores, the lowest-income families have weaker preferences for academic outcomes and are more constrained in their ability to choose the highest-performing schools. We also examine how measured preferences have changed over time in somewhat predictable ways. In addition to helping understand the demand for schools generally, the somewhat limited influence of academics in family decisions helps explain several puzzles from prior research, including why the competitive effects of market-based school reforms on student test scores have been mixed.
“Testing the Validity of Across Cohort Quasi-Experiments Using a Cluster-Randomized Control Trial of an Early College Scholarship,” joint with Doug Harris
(Copy not Publicly Available)
Abstract: Randomization is often considered the surest way to identify unbiased estimates of treatment effects, but often only quasi-experimental methods are feasible and there is debate about whether and when they are likely to yield biased estimates. A growing number of “within-study comparisons” suggest that propensity score matching often does not yield the same conclusions as randomized control trials (RCTs), while regression discontinuity designs do generally yield the same conclusions. In this paper, we study an early college scholarship and compare the RCT identification with difference-in-differences (DD) of the same intervention. While there are conflicting conclusions for about half of the outcomes, the exogeneity assumption is rejected for all of them, so we would not generally trust these DD estimates. This implies that, by the usual standards, we would reach the same general conclusions regardless of whether we used an RCT or a DD for this intervention. The similarity in conclusions occurs despite the over-rejection of the null when the DD involves a small number of groups by Donald and Lang (2007).
“The Effect of High School Graduation Requirements on Teenage Birth Rates and Infant Health”
(PDF Copy)
Abstract: This paper analyzes possible spillover effects of raising high school graduation requirements. Using Natality data, I compare the birth rates and other teen pregnancy outcomes for state/cohorts with more difficult graduation requirements to those with less difficult requirements. I find evidence that more requirements reduces teenage birth rates, where each additional course required decreases the birth rate by 1% and 0.7% for white and black mothers respectively. this effect is focused in low income counties, which may be the counties most in need of assistance. I find no effects of more difficult graduation requirement on a variety of other teen pregnancy outcomes including birthweight and premature births. Results are favorable for increasing graduation requirements, given that I find no negative effects of these requirements.
Works in Progress:
"Heterogenous Preferences in School Choice," (with Douglas N. Harris)
Abstract: Following Hurricane Katrina, New Orleans eliminated school attendance zones and almost all schools were converted to charter schools. Harris, Larsen, Zimmerman (2014) examines what school characteristics the average parents care most about when choosing a school. In this paper, we examine how those choices differ across families based on racial, socioeconomic, and prior academic differences. Do parents with special needs children look for schools with different services than those without special needs children? Do children who have had discipline issues in the past choose schools with a different disciplinary philosophy? Do students with low test scores choose schools with high or low average test scores or high? High or low Value-Added? We use a unique dataset which connects students' prior academic history to their ranking of schools on a centralized district school application to answer these questions.
"From One to Many: The Effect of Requiring Course Specific Graduation Exams"
Abstract: Previous research has shown that the use of graduation exams can reduce the probability of graduation. Comprehensive graduation exams are often taken early in high school and yield several re-test opportunities for those students who fail. These tests often taken in the Spring, test for comprehensive knowledge of the core topics taken early in high school. Recently, many states have opted to switch to "End of Course" exams, which are taken at the end of each course and only test on the material within that course. Beginning with the freshmen cohort of 2010-11, Louisiana moved from requiring a comprehensive graduation exam to multiple end of course tests. While students now will take exams that are more focused on certain subjects, they are required to pass 3 of 6 exams some of which are taken as late as 11th grade. I use regression discontinuity analysis to estimate what effect each regime has on the graduation rate of students at the margin of passing. I also describe how the average characteristics of the marginal students differ under each regime.
“The Effect of School Investment on Juvenile Crime,” (with Danielle Sandler)
Abstract: It has long been argued that investment in education and schools can help reduce crime. We seek to estimate if that relationship holds true for school facility investments. Updating and renovating schools can potentially improve education and morale of students. Investments in gyms and sports courts may provide an outlet for neighborhood kids who may turn to crime and mischief otherwise. Finally, the simple act of investing in schools may demonstrate that a neighborhood is invested in itself and deter crime from occurring. Following Cellini, Ferreira, and Rothstein (2010) we use a regression discontinuity in voting thresholds yielding approval for school facility investments in California to estimate how those investments affect local crime rates.