Working Papers
Teacher mobility to high-cost urban areas [Job Market Paper]
with Pedro Freitas and Luís Catela Nunes
Abstract: The design of new policies to attract and retain new teachers has been a top issue in many educational systems. In this work, we focus on which incentives work best in attracting teachers to hard-to-staff high-cost urban areas. We conducted a discrete-choice experiment with approximately 800 teachers from 170 school clusters in Portugal, a country where teacher shortages are expected to worsen in the coming years. Alternatives differ in terms of school contract attributes related to location, salary, and stability. We find that substantial salary increases are needed to incentivize teachers to move to Lisbon, a high-cost urban area with severe teacher shortages. The results also provide insight into how to design cost-efficient incentives by taking into account the observed heterogeneity in teacher preferences in terms of several of their characteristics such as gender, experience, and personal and material ties. We suggest that bonuses varying with distance and targeted to regions with higher scarcity may be an effective way to address shortages. Finally, we estimate a supply curve to quantify the additional expenditure in salary increases needed to attract teachers to Lisbon.
Measuring school quality in Portugal: Drift, Bias and Heterogeneity by Baseline Achievement
with Pedro Freitas, Luís Catela Nunes and Ana Balcão Reis
Abstract: The value-added methodology is used in many countries to rank schools for both accountability and school choice purposes. We estimated value-added for Portuguese schools, using administrative data from 2007 to 2018. In particular, we analysed the effect of the school on the students’ scores at the end of upper secondary school, controlling for baseline scores and other student characteristics. To account for noise, these estimates were adjusted by applying an empirical Bayes approach. We find that a one standard deviation (SD) improvement in school value-added corresponds to an increase in standardised test scores by about 0.13 SD in Portuguese and 0.2 SD in Mathematics. We then allowed for school quality to vary over time, and estimated the bias in forecasting schools' impacts on student achievement from omitting a set of parental characteristics which are only available for public schools. A small bias was found for Portuguese scores and a moderate one for Mathematics. We also investigate heterogeneity in value-added by students’ baseline scores, and find that some schools perform better for initially low-achieving students and others for initially high-achieving students, and that there is more variation in value-added for lower levels of initial achievement.
The role of management practices on educational outcomes in Vietnam
with Pedro Carneiro, Anusha Guha and Sonya Krutikova
Abstract: We study whether school management and leadership are related to student and teacher outcomes in Vietnam, a high-performing, lower-middle-income country with a centralised education system. We collect data on management practices in 141 primary schools, and link these to a broad range of educational outcomes: standardised test scores for over 10,000 students over four years, and measures of higher-order cognitive skills, non-cognitive skills, teaching quality, and teacher burnout. To account for composition and sorting, we include a rich set of covariates regarding school, student, teacher and parental characteristics, as well as prior achievement. Then, we attribute the change in the estimated relationship between management practices and outcomes to different sets of covariates, identifying how each group contributes to the overall variation.
Publications
Understanding the private-public school performance gap in PISA: Evidence from Portugal
with Pedro Freitas, Luís Catela Nunes and Ana Balcão Reis
Abstract: We analyse the PISA-reported convergence in the performance of private and public schools in Portugal. When PISA sampling weights are used, the number of students enrolled in those types of schools and specific grades/tracks of study differs significantly from official population figures. To account for those differences, we apply a post-stratification adjustment; however, sample sizes are small, resulting in estimates with low precision for several subgroups. We propose recommendations for improving the handling of these issues in future PISA cycles. In an additional analysis, we also account for changes in the composition of the student population. When all factors are considered, the convergence in scores is far less impressive than reported. For instance, in Science, after adjusting the sampling weights and removing population composition effects, the reported convergence of 46 points between private and public schools from 2015 to 2018 amounts to only 9 points. The decomposition and sample adjustment methods used in this paper can be easily adapted to other contexts.
Policy Reports
Trends in student achievement in Portugal: What does PISA tell us
with Pedro Freitas and Luís Catela Nunes
Abstract: This report reflects on the evolution of student performance in Portugal, specifically among 15-year-olds, over the past 20 years, based on an analysis of PISA results. It highlights the influence of parental education and socioeconomic context on students' performance in the PISA tests.