School zoning policy in Bandung city: A map based approach

Abstraksi

As a consequence of the school zoning policy implementation, acceptance to public school is now partially determined by the distance of student’s residency to school. Despite the good intention in reducing school trips distance, the policy potentially raises new problems, especially on the inequality of access to public education because students living close to good-quality school vicinity are the one benefiting most from the policy (Hamnett & Butler, 2013). Through this study, we provide an initial picture of school quality distribution in Bandung city prior to the implementation of school zoning policy. With a clearer understanding of the situation, we expect to provide recommendations for better implementation of the school zoning system and support the policy to achieve its objective of equal access to quality education that is crucial for inclusive growth. Background Inequality in education is often a result of inadequate academic resources distribution such as teachers, schools, books, and technology. However, there is another factor that contributes to the problem: distance. When the allocation of students depends on the distance to school, the policy reinforced social segregation, leading to the exclusion of disadvantaged students from strong reputation schools. Until 2016, the system to allocate students in public schools mainly relies on the final exam scores from their previous school level. The applicants' rank determined their acceptance, adjusting on the school's student capacity. In 2017, MoEC released a school zoning policy which aims to provide equal distribution of the quality of education where school enrollment is also condition on distance. Research objective Our study springs from previous literature that shows zoning policy exacerbates the existing residential social disaggregation. We propose two research questions: 1. What is the picture of school quality distribution in Bandung city before the introduction of school zoning, do they show a similar level of achievement or there is a hierarchy of performance? 2. What is the association between school learning performance and residential social division before the zoning policy in Bandung city? Data We select Bandung city for this study because the city fully implemented the policy in 2019 and the feasibility in gaining information on how the city government implement the policy is accessible. To identify the possible school hierarchy based on their quality, we utilise the national examination score in 2017. We also use various poverty indices as a proxy of residential social division. The first is SMERU’s poverty map that provides poverty measures for each village in terms of headcount, poverty gap and poverty severity index against the national poverty level. The second is the number of individuals who fall into the bottom 40 percent of income in each sub-district from TNP2K data. We extract their percentage by comparing the figures with the total number of individuals living in the respective sub-district from BPS data. Methods To answer the first research question, we perform cluster analysis for public schools based on their national exam performance. The results using several clustering methods for robustness check are consistent, where we show a dendrogram from an agglomerative hierarchical cluster that intuitively displays the clustering stages. For the second question, we examine the linear relationship between national exam scores and various poverty indices using simple linear regression. Additional insight is derived from data visualisation. Findings We find that four clusters best exhibit the grouping of public high school quality in Bandung. The average silhouette width is 0.61 indicates a reasonable structure of school clusters based on national exam performance. This evidence suggests that inequality of educational quality is present in public education prior to zoning policy commencement. From a spatial dimension perspective, we found that public high schools in Bandung are not evenly distributed spatially. Most public schools are located in the west part of the city, which means that households in the east part have less opportunity to register to the public school that is known to have relatively good quality. Findings show inequality in education access and quality as a problem faced by Bandung residents. To what extent they interact with each other? Regression analysis shows that the increase of one percent of the bottom 40 percent income earners is associated with a decrease of 0.64 points in average national exam score (p-value 0.009). This is an early indication of inequality on quality education that is biased towards higher income areas. Adding raw distance as a constraint to access public education threats (i) the relatively fair enrollment system that centred on effort-based and (ii) the existing inequality on access and quality education such that it encourages the residential social division even more. Policy recommendation We propose an alternative to the current straight distance-based zoning policy, which is the multiple Voronoi layers cell-based zoning policy. Voronoi cell has public school as the centre of gravity, while the layers represent the clusters. It has a distinct advantage over the radius approach, gives an equal opportunity for a student to reserve public school in each cluster. We argue that this approach is fairer and more inclusive. The top cluster consists of one school, so this layer has only one cell that covers the whole Bandung city. Therefore all residents in Bandung city can apply regardless the distance. The second cluster consists of nine schools, making the second layer divided into nine zones. Students living in a certain zone are fully eligible to apply to the respective public school. Since there are four clusters, each student has the right to apply to four schools that is within her proximity unconditionally. We encourage similar methods are applied to other cities in order to obtain a bigger picture on the level of inequality in access and quality of public education. These insights are useful to inform public policy and tailor it to specific needs to reduce inequality and improve inclusive education.