This paper analyses the complex relationship between social classes, scholastic stratification and educational choices in Italy, paying due attention to the role teachers may play in enhancing working-class students' chances to attend university.

We follow a perspective that considers the school choices as made on the basis of a complex interweaving between individual and institutional factors.

About the individual factors, for years rational choice (RC) scholars (Boudon, 1973; Gambetta, 1987) have been stating the factors that would affect the students’ expectations and their scholastic choices are just the material resources. In such a framework, students from the upper class are impelled to choose high level of education in order to maintain their social position; while those of lower class get the low or middle level of education, i.e. the level they consider sufficient enough to maintain their social position (Goldthorpe, 2000).

Following the distinction proposed by Girard and Bastide (1963), RC scholars think “secondary effects” of social class are more important than “primary ones”.

From a completely different point of view, Bourdieu (1979) showed how the primary effects are more relevant than the secondary ones: on the one hand the cultural capital of the working class students is inadequate to scholastic requests; on the other hand the school itself incorporates and reproduces high classes’ culture (Bourdieu, 1984).

We analyse the Italian education system through the lenses of the bourdesian perspective (Bourdieu, 1979, 1984), but we also consider the students’ expectations, as RC scholars suggest.

This strategy lead us to examine the role that school plays in contrasting inequalities in educational attainment through the pedagogical relationship, as New Sociology of Education highlights (Young, 1971; Giroux, 2005; Apple, 2012).

We think that on the one hand the educational system affects students’ motivation, acting on the definition of risks and benefits (Young, 2007), on the other hand class subculture affects the evaluation of risks, costs and benefits, that in turn may be reduced by educational system (Jackson, 2013).

To recap, our research investigates the bivalent nature of the educational system that on the one hand legitimizes the social order, on the other promises emancipation and universalism (Collins, 1979; Brint, 2006).

Given the above, we want to test four hypotheses:

Hypothesis I – the choice to go to university depends on social class and parental education, but the last one is more relevant since the evaluation of risks and benefits is influenced by parental education, in turn associated to social class;

Hypothesis II – the direct effect exercised on educational choices by parental education is more important than its indirect effect through scholastic performance;

Hypothesis III – the working class students can reduce their social distance from school thanks to a good relationship with teachers, who can give a motivational boost to go to university, also when parental education is low;

Hypothesis IV – the dynamics above is less important for middle and upper class students: they tend to have a middle-high parental education and to choose university, also when their scholastic performance is low and the pedagogical relationship is not positive, given the apprehension to loose the social position inherited.

About this, we test if a positive pedagogical relationship (Giroux, McLaren, 2014), namely the enactment of an inclusive teaching (Apple, 2013), increases working class students’ likelihood to choose university.

Consistently with our theoretical frame, we built a multilevel research design. By means of the data collected by Istat in 2011, among secondary school graduates in 2007 (our sample is composed of 8,334 interviewees), we analysed the interviewees' situation in 2011, i.e. four years after they acquired the upper-secondary school diploma. Therefore, in order to test the first two hypotheses (see introduction) we point at the unequal likelihood of going to university between the upper-secondary school graduates that come from different social classes. At this aim, we evaluate the influence of principal factors that reproduce social inequalities regarding scholastic attendance in university, by means of a binary logit model. It is a model where the decision to go to university (68.8% of interviewees) or not (38.5%) represents the dependent variable, whereas social class and parental education are the independent variables. Further, we selected three mediating variables. The first one is the type of upper-secondary school students attend: high (52,2%), technical (27,7%) and vocational schools (20,1%). The second one - we choose because it represents a good proxy of scholastic performance - is the exit exam result, whose score is on 100-point scale. We divided marks in four modalities: 60-70 (35,3%); 71-80 (27,4%); 81-90 (18,7%); 91-100 (18,6%). Eventually, we built an index that reproduces 63% of variance of three variables through the principal component analysis (PCA). These last ones concern students' satisfaction about: teachers' expertise (+. 454), relationship with teachers (+. 437), and lesson content (+. 366). The values in brackets represent the factor score coefficients (Di Franco & Marradi, 2003). Numeric index was converted into a categorical variable with the following modalities: negative when the score is less than zero (22,5%); neutral when it is zero (56,1%); positive when it is more than zero (20,4%). Therefore, this index is used as a third mediating variable, since it allows us to measure the relationship between students and teachers in upper-secondary schools: the higher its value, the more positive the pedagogical relationship becomes. As said above, in order to test the second two hypotheses (see introduction) - and by resorting to the previously examined panel of variables - we adopted MCA joined to cluster analysis. In effect, these techniques allow us to focus on the role that a good pedagogical relationship may play in reducing social distance between working class students and school, motivating this kind of students in choosing university.

Our findings underline the university choice especially depends on the social distance by parental education and school culture. In this framework, middle and upper class students have a wealth of knowledge that allows them to follow a positive scholastic trajectory. These students can rely on a good amount of parental education that motivates them to choose university also when the scholastic performance and (or) the relationship with teachers are negative (Ball, 2006). Working-class students, for their part, have a low cultural capital, so they are disadvantaged in their scholastic learning. On the one hand their social dispositions lead them to feel detached from school. On the other hand, they rely on the teachers' evaluations. Our results suggest that working-class students in many cases attribute a sort of social superiority to teachers, as Bourdieu had already noted (Bourdieu, 1979). Therefore, working-class students can react to an asymmetric social relationship with their teachers, and further distance themselves from school. In this last case the students did not show interest in education nor did they have enough confidence in their scholastic career. In this case, their performance, therefore, was not very high. However, when the interactions with teachers are positive, working-class students tend to improve their performance, become particularly motivated to study, may choose to attend university. So, working-class students can enhance an interest in studying when they encounter teachers who inspire them through a positive relationship, good expertise and interesting lessons. These findings show the importance of the pedagogical relationship as transformative factor (Gramsci 1975; Giroux, 2005), that can counteract the social reproduction, by changing the relevance attached to education by the working-class students. Unfortunately, the effect of pedagogical relationship is much weaker than that exercised by structural factors connected to scholastic stratification.

Apple, Michael W. (2012). Can education change society?. London and New York: Routledge. Apple, Michael W. (2013). Education and power. London and New York: Routledge. Ball, Stephen J. (2006). Education Policy and Social Class. London: Routledge. Boudon, R. (1973). L'inegalité des chances. Paris: Armand Colin. Bourdieu, P. (1979). La Distinction. Critique sociale du jugement. Paris: Editions de Minuit. Bourdieu, P. (1984).Homo academicus. Paris: Editions de Minuit. Brint, S. 2006. Schools and Societies. Stanford: Stanford University Press. Giroux, H.A. (2005). Schooling and the struggle for public life. Democracy's promise and education's Challenge. Boulder: Paradigm Publishers. Giroux, H.A. & McLaren, P. (Eds). 2014. Between borders: Pedagogy and the politics of cultural studies. London and New York: Routledge. Goldthorpe, J.H. (2000). On sociology: Numbers, narratives, and the integration of research and theory. Oxford: Oxford University Press. Gramsci, A. (1975). Quaderni dal carcere (IV Volumi), edited by V. Gerratana. Torino: Einaudi. Jackson, M. (2013). Determined to succeed?: performance versus choice in educational attainment. Palo Alto: Stanford University Press. Willis, P. (1977). Learning to Labour. New York: Columbia University Press. Young, M.F.D.1971. Knowledge and Control. London: Collier-Macmillan. Young, Michael. 2007. Bringing knowledge back in: From social constructivism to social realism in the sociology of education. London and New York: Routledge.