Inequality of opportunity for educational achievement in Latin America: Evidence from PISA 2006–2009

We evaluate how far away six Latin American countries stand from a normative goal of equality of opportunity for educational achievement in PISA 2006–2009. We work with alternative characterizations of types: gender, school type (public or private), parental education, and their combinations. Following Checchi–Peragine's (2010) non-parametric method, we find that inequality of opportunity for educational achievement in Latin America ranges from less than 1% to up to 25%, depending on the year, the country, the subject and the specification of circumstances. These magnitudes are substantial with respect to what is found in comparator countries. Parental education and school type prove to be important sources of inequality of opportunity, contrary to gender. By means of sensitivity analyses, while most results show small to moderate variation in terms of magnitudes, in ordinal terms (rankings) they remain quite stable. Brazil stands out as the most opportunity-unequal country of the sample.     

What does it take to achieve equality of opportunity in education?: An empirical investigation based on Brazilian data

Roemer's [Roemer, J. (1998). Equality of opportunity. Cambridge, MA: Harvard University Press.] seminal work on equality of opportunity has contributed to the emergence of a theory of justice that is modern, conceptually clear and easy to mobilize in policy design. Inspired by Roemer's theory, this paper is fundamentally a policy-modeling exercise coupled to a micro data analysis. In a pure allocation setting, we first analyze the reallocations of educational expenditure required to equalize opportunities (taken to be test scores close to the end of compulsory education). Using Brazilian data, we find that implementing an equal-opportunity policy across pupils of different socio-economic background, by using per-pupil spending as the instrument requires multiplying by 6.8 on average the current level of spending on the lowest achieving pupils. This result is driven by the extremely low elasticity of scores to per-pupil spending. We then show that the simultaneous redistribution of monetary and non-monetary inputs, like peer group quality and school effectiveness, considerably reduces—by around 23%—the magnitude of financial redistribution needed.     

Measuring Inequality of Opportunity in Education by Combining Information on Coverage and Achievement in PISA

Measuring the distance between countries and the goal of equality of opportunity in education has been the focus of recent contributions in the economic literature, which have concentrated either on intergroup gaps in access to a given level of studies or on intergroup gaps in educational achievement. We argue that both aspects are important and interrelated and, as a consequence, should be measured simultaneously and combined in a composite index of equality of opportunity in education. Employing recent PISA data for a selected set of countries, rank reversals are observed with respect to orderings based upon a single dimension. The index can be generalized to include further dimensions, such as average scores or dissimilarity in access.