Social Inequalities in the Choice of Secondary School

ABSTRACT

The main aim of this work is to examine the long-term trends in the association between social class of origin, enrolment in upper secondary education and the choice of high school track. In the first part, we describe the Italian education system and the main educational reforms which occurred in the second half of the twentieth century. We also discuss several theories which can help to make predictions on the expected trends in vertical and horizontal inequalities in secondary education. We used binomial and multinomial logistic regression models on data from the Italian Households Longitudinal Survey to test our hypotheses. In line with the maximally maintained inequality argument, we found that absolute inequalities in the probability of enrolling in upper secondary education declined but relative inequality persisted. As predicted by the ‘effectively maintained inequality’ thesis, the association between social class of origin and the choice of the academic track grew over time (both in absolute and in relative terms). This is because children from the upper classes became increasingly likely to attend the academic track, while those from the working class were more likely to attend the technical and vocational schools. The educational reforms played a negligible role in reducing social inequalities in school prosecution after lower secondary education, while they had a side effect of boosting horizontal inequality in track choice.

Key words:

• tracking
• inequality of educational opportunities
• upper secondary education
• social stratification
• Italy
• educational reforms

Notes

¹ This suggests that it should be lower class children with higher academic ability who mostly benefited from these changes.

² Despite the decline in the average occupational prestige attained by technical diploma holders in this period, a degree from a technical school effectively protected from the risk of unemployment and gave access to medium-level occupations, which potentially allowed middle-lower class children to avoid social demotion.

³ More detailed information on the sample design can be found in Pisati and Schizzerotto (2004).

⁴ Nevertheless, the number of individuals excluded is modest and does not substantially change the results.

⁵ Studies on tracking in the Italian system usually include these humanistic curricula into the academic track because in recent years their students have rapidly increased their transition rate to university. Since we focus on the twentieth century, we decided to aggregate these curricula with the technical ones because they share more similar trends of enrolment at university.

⁶ The Akaike Information Criterion (AIC) is 17,036 for the ordinal logit, 15,880 for the generalized ordered logit and 15,870 for the multinomial logit.

⁷ One potential problem of such model is that it may violate the independence of the irrelevant alternatives (IIA assumption) at the basis of the multinomial logit model. We performed the Hausman test to check this issue. The results indicate that there is variability according to the category considered: the null hypothesis that a category of the dependent variable is independent from the others cannot be rejected for all the categories, with the exception of the technical education. Nevertheless, since negative test statistics are very common with this test, following Hausman and McFadden (1984: 1226), we do not consider this as a strong violation of the IIA assumption and thus we employ such model in the analysis.

⁸ The average partial effects on the outcome ‘not enrolled’ are not reported, since the results are specular to those extracted from the binomial logistic regression which analyses the probability of enrolment.

⁹ Therefore, future analyses on larger sample sizes are needed to provide a definite test of such hypothesis; unfortunately, at the moment such a data-set is not available in Italy.

Additional information

Notes on contributors

Nazareno Panichella

Nazareno Panichella, Ph.D., is a post-doc research fellow in the Department of Social and Political Science at the University of Milan. His research interests include migrations, education and social stratification.

Moris Triventi

Moris Triventi, Ph.D., is a research fellow in the Department of Political and Social Sciences at the European University Institute. His research interests include education, social stratification and graduates' transition in the labour market.