Search

Randomized controlled trials (RCTs) and regression discontinuity (RD) studies both provide estimates of causal effects. A major difference between the two is that RD only estimates local average treatment effects (LATE) near the cutoff point of the forcing variable. This has been cited as a drawback to RD designs (Cook & Wong, 2008). Comparisons of RCT estimates of average treatment effect (ATE) and RD estimates of LATE are rare because few studies have both randomized assignment and a forcing variable. Using a study which is amenable to analysis both as an RCT and an RD, the authors compare the RCT estimate of ATE to the RD estimate of LATE. They also explore how the RD estimate of average treatment effect depends on econometric specifications, including for parametric and non-parametric estimators, and for various bandwidths. Finally, they use Monte Carlo simulation based on observed data to explore the performance of RD estimators with a forcing variable which is standardized within schools by choosing a different cutoff in each school. Findings can be used to inform decisions about which estimators to use in doing regression discontinuity, as well as providing evidence of how well RD estimators approximate RCT estimators. Tables are appended.

Parental involvement programs aim to increase school-and-parent communication and support children's overall learning environment. This paper examines the effects of low-cost, group-based parental involvement interventions in Mexico using data from two randomized controlled trials. The first experiment provided financial resources to parent associations. The second experiment provided information to parents about how to support their children's learning. Overall, the interventions induced different types of parental engagement in schools. The information intervention changed parenting behavior at home -- with large effects among indigenous parents who have historically been discriminated and socially excluded -- and improved student behavior in school. The grants did not impact parent or student behaviors. Notably, we do not find impacts of either intervention on educational achievement. To understand these null effects, we explore how social ties between parents and teachers evolved over the course of the two interventions. Parental involvement interventions led to significant changes in perceived trustworthiness between teachers and parents. The results suggest that parental involvement interventions can backfire if institutional rules are unclear about the expectations of parents and teachers as parents increase their involvement in schools. [Funding for this research was provided by the Spanish Impact Evaluation Fund and the Bank-Netherlands Partnership Program.]

We randomly assign applicants to over-subscribed programs to study the effects of teaching hard and soft skills in vocational training and examine their impacts on skills acquisition and labor market outcomes using both survey and administrative data. We find that providing vocational training that either emphasizes social or technical skills increases formal employment for both men and women. We also find that admission to a vocational program that emphasizes technical relative to social skills increases overall employment and also days and hours worked in the short term. Yet, emphasis on soft-skills training helps applicants increase employment and monthly wages over the longer term and allows them to catch up with those learning hard skills. Further, through a second round of randomization, we find that offering financial support for transportation and food increases the effectiveness of the program, indicating that resource constraints may be an obstacle for individuals considering vocational training. [Carvajal Foundation partnered with the authors to run and implement the program.]

This study estimates the causal effects of a public per-student subsidy program targeted at low-cost private schools in Pakistan on student enrollment and schooling inputs. Program entry is ultimately conditional on achieving a minimum stipulated student pass rate (cutoff) in a standardized academic test. This mechanism for treatment assignment allows the application of regression-discontinuity (RD) methods to estimate program impacts at the cutoff. Data on two rounds of entry test takers (phase 3 and phase 4) are used. Modeling the entry process of phase-4 test takers as a sharp RD design, the authors find evidence of large positive impacts on the number of students, teachers, classrooms, and blackboards. Modeling the entry process of phase-3 test takers as a partially-fuzzy RD design given treatment crossovers, they do not find evidence of significant program impacts on outcomes of interest. The latter finding is likely due to weak identification arising from a small jump in the probability of treatment at the cutoff. (Contains 13 tables, 7 figures and 29 footnotes.)

The Oaxaca-Blinder technique was originally used in labor economics to decompose earnings gaps and to estimate the level of discrimination. It has been applied since in other social issues, including education, where it can be used to assess how much of a gap is due to differences in characteristics (explained variation) and how much is due to policy or system changes (unexplained variation). The authors apply the decomposition technique in an effort to analyze the increase in Indonesia's score in PISA mathematics. Between 2003 and 2006, Indonesia's score increased by 30 points, or 0.3 of a standard deviation. The test score increase is assessed in relation to family, student, school and institutional characteristics. The gap over time is decomposed into its constituent components based on the estimation of cognitive achievement production functions. The decomposition results suggest that almost the entire test score increase is explained by the returns to characteristics, mostly related to student age. However, the authors find that the adequate supply of teachers also plays a role in test score changes. (Contains 3 figures and 9 tables.)

Low student learning is a common finding in much of the developing world. This paper uses a relatively unique dataset of five semiannual rounds of standardized test data to characterize and explain the short-term changes in student learning. The data are collected as part of the quality assurance system for a public-private partnership program that offers public subsidies conditional on minimum learning levels to low-cost private schools in Pakistan. Apart from a large positive distributional shift in learning between the first two test rounds, the learning distributions over test rounds show little progress. Schools are ejected from the program if they fail to achieve a minimum pass rate in the test in two consecutive attempts, making the test high stakes. Sharp regression discontinuity estimates show that the threat of program exit on schools that barely failed the test for the first time induces large learning gains. The large change in learning between the first two test rounds is likely attributable to this accountability pressure given that a large share of new program entrants failed in the first test round. Schools also qualify for substantial annual teacher bonuses if they achieve a minimum score in a composite measure of student test participation and mean test score. Sharp regression discontinuity estimates do not show that the prospect of future teacher bonus rewards induces learning gains for schools that barely did not qualify for the bonus. (Contains 13 tables, 16 figures and 17 footnotes.) [This paper is a product of the Education Unit, South Asia Region; and the Education Unit, Human Development Network of the World Bank.]