This paper focuses on the direct and indirect effects of family background and school quality on the educational achievement scores of high school students in developed countries. Not only does the educational, occupational, and economic characteristics of one's family have a direct impact on how well one does on high school achievement tests, but it can also have an indirect effect through the school that one attends. The quality of schools thus may serve both as a mediator of family background and an independent source of variation in student achievement scores. Therefore, the structure of a country's educational system—particularly the variation in school quality and the process of student assignment to schools—can impact the relationship between family background and achievement scores and serve as a source of inequality in educational achievement.

I used data for twenty-one countries from the first round of the Programme for International Student Assessment (PISA 2000), which assessed students' ability in the domain of reading literacy. The data were analyzed at the individual country-level data and were grouped by similarities in the national educational systems. Educational systems were grouped by their level of standardization, which is based on the level of centralized control over schools, and stratification, which is determined by the type of student tracking that occurs at the secondary level. These data were then analyzed using descriptive analyses, regression analysis and multiple variance decomposition techniques.

Results of these analyses show a number of interesting findings. First, there are significant differences in reading achievement, family socioeconomic status, and school characteristics across different types of education systems, most notably across levels of stratification. Second, the socioeconomic composition of a school's student body has the largest, most consistent effect of all school variables and is the strongest predictor in a majority of countries. In addition, the school variables as a group explain more variance in student achievement than the family background variables in half of the countries, especially those with stratified education systems. Third, the indirect effects of family background through school characteristics on educational achievement are larger in countries where school factors explain more achievement variation than family factors and in countries with highly stratified education systems. Fourth, simulations indicate that changes to the effects and variances of school variables have a much larger impact on achievement variance than changes to family background variables. Lastly, as reflected in many of the findings above, the stratification of a country's education system is an important factor in determining the impact of families and schools on a child's educational achievement. Students in non-stratified systems perform, on average, better than those in stratified education systems, and their educational performance in high school is less dependent on family background factors.