An international comparison of equity in education systems

Abstract

This paper uses pupil responses to the PISA study in 2000 for all EU countries. Using indicators of the pupil intakes to schools and their outcomes it computes segregation indices for 15 countries, and then tries to explain the resulting patterns in terms of the characteristics of national school systems. Segregation by sex in each country is explicable by its provision of single‐sex schools, religious schools, and the use of academic selection in allocating school places. Segregation by outcome is largely explicable by the use of academic (and other forms of) selection. Segregation by parental occupation or country of birth is lower in countries allocating places at school through elements of choice or with relatively little governmental control of schools rather than use of rigid catchment areas or selection. In all countries there are small gaps between the performance of boys and girls in reading, in favour of girls. This gap is generally smaller in countries with the highest overall scores. Overall, the Scandinavian countries of Sweden, Finland and Denmark show less segregation on all indicators, while Germany, Greece and Belgium show the most. The UK has below average segregation in terms of all indicators except sex, despite a commonly held but unfounded view that segregation in the UK is among the worst in the world.

Notes

* Corresponding author: Cardiff University School of Social Sciences, Glamorgan Building, King Edward VII Avenue, Cardiff, CF10 3WT, UK. Email: [email protected]

• The Dissimilarity index

• A_i and B_i are the number of pupils in the minority (10%) and majority (90%) categories respectively within each school. X and Y are the total number of pupils in either category within each country.

• The Gini coefficient

• Where y_I is the cumulative proportion of pupils in the lowest 10% group (or those who were born outside the test country or were female, depending on indicator being measured) and x_i is the cumulative proportion of whole population. The Gini Coefficient is based on the Lorenz curve. This is a cumulative frequency curve that compares the ‘distribution of a specific variable with the uniform distribution that represents equality’ (Castillo‐Salgado et al., 2002, p. 1). This ‘equality distribution’ is represented by a diagonal line, the greater the deviation of the Lorenz curve from this line, the greater the inequality. The Gini Coefficient ranges from 0 to 1, with 0 representing perfect equality and 1 total inequality.

Additional information

Notes on contributors

Stephen Gorard *

* Corresponding author: Cardiff University School of Social Sciences, Glamorgan Building, King Edward VII Avenue, Cardiff, CF10 3WT, UK. Email: [email protected]