Three empirical essays on the economics of education
- Autores: Antonio Di Paolo
- Directores de la Tesis: José Luis Raymond Bara (dir. tes.), Jorge Calero Martínez (codir. tes.)
- Lectura: En la Universitat Autònoma de Barcelona ( España ) en 2011
- Idioma: español
- Tribunal Calificador de la Tesis: Jaume García Villar (presid.), Lorenzo Cappellari (secret.), Regina T. Riphahn (voc.)
- Materias:
- Texto completo no disponible (Saber más ...)
- Resumen
This PhD Thesis is composed by three essays on the Economics of Education, with a marked empirical orientation. The common idea behind these essays consists in providing additional evidences on the role of parental education as an input of the schooling process. Indeed, as usually reported in the literature, parental background represents the main determinant of individuals' schooling attainment. There exists a large amount of research concerning the direct effect of parental schooling background. Several papers attempted to rule out the genetic transmission component from the relationship between parents' and child's education. Other studies tried to quantify such relationship in terms of causal effects, exploiting exogenous variations in schooling among the parents' generation.
However, it is also worth studying with more detail other potential mechanisms through which parental education may operate as an indirect input. Therefore, during the last years, I have focused my research on giving an answer to three main empirical questions: 1) to which extent children of better educated parents achieve more schooling because higher parental education is also associated with a better endowment of other schooling determinants? 2) Which part of the statistical relationship between parental and individual's education is mediated by the effect of other family characteristics? 3) Which is the effect of an increase in the school-average parental education on students' competences acquisition? With respect to the first question, in the first essay I compare the evolution of educational opportunities for individuals of different parental background for two concrete countries, namely Italy and Spain. I believe that this study is relevant because both countries maintained a highly stratified education system during an important part of the last century. This means that, to some extent, the disparities in educational attainments might have persisted across generation, which might be one of the causes of the significant socio-economic inequalities that we observe nowadays. Therefore, the analysis of temporal changes might be informative about whether institutional and economic development, as well as other social dynamics, have reduced or not such disparities. Moreover, after the analysis of temporal patterns, I also try to answer the main empirical question of this work. That is, using a variant of the well-know Oaxaca-Blinder decomposition, I isolate the part of the disparities in educational opportunity that is explained by the better endowment of other family characteristics that children of more educated parents may enjoy.
For the second point, I apply a novel index for measuring the degree of educational mobility to examine the changes in intergenerational persistence across time for twelve European countries. This topic is interesting in its own right, given that educational mobility is considered to be one of the main ingredients of social and economic inequality. With this work, apart from the methodological contribution based on the proposal of this new index, I also provide some evidence about the role of both parents in the intergenerational transmission of education to their offspring. Moreover, the properties of the proposed mobility index enable investigating the extent to which the relationship between parental education and other family characteristics mediates the educational persistence from one generation to the next, which corresponds to the second main research question of these PhD thesis. With the analysis for different countries during a long time period, I am also able to describe how this mediating effect of family characteristics evolves over time and place.
Finally, the third empirical question that I investigate in my thesis is more specifically addressed to a target population. In fact, I use the Spanish data from the Programme for the International Students Assessment (PISA), which contains information on the competences' acquisition of students enrolled in lower secondary school, their parental background and other characteristics of the schools they are attending. This makes possible to consider the effect of the socio-economic environment at the school level (which I define in terms of the school-average parental education) on students' test score. In other words, I consider how students' performance is affected by the parental background of their peers in the same school. As explained in what follows, the semi-parametric methodology used in this essay enables capturing the whole effect of these externalities (i.e. not only changes in the average test score). Moreover, I also deal with the endogenous sorting process that allocates more able students into better schools, which may account for a significant part of these potential spillovers of parental education. As recognised by several authors, these kinds of evidences are extremely useful for education policies. In fact, they help to design more equitable and even efficient school-district policies (in terms of segregation or integration of students from different backgrounds), as well as ability-streaming policies within each school.
