Worldwide most secondary schools apply some form of tracking. This educational stratification consists of grouping students according to their ability. The research into the consequences of tracking occupies a significant place within the sociology of education. From the end of the 1960s it has been demonstrated repeatedly that students in lower tracks develop an anti-school culture, to overcome the status deprivation resulting from being in a lower track (Ball, 1981; Hargreaves, 1967; Lacey, 1970; Rosenbaum, 1976). Usually in quantitative research this anti-school culture is assessed by study engagement or academic attitudes (Berends, 1995; Catsambis, Mulkey, & Crain, 1999). This polarisation into an anti-school-culture and a pro-school-culture in answer to the differentiation into distinct tracks, has usually been demonstrated within schools. It is the case that in the British and North-American context, where the bulk of this kind of research has been carried out, tracking is mainly ability grouping within a comprehensive schooling structure. In Flanders—that is, the Dutch-speaking, northern part of Belgium—as in other European countries, there are not only tracks within schools, but schools themselves can be distinguished according to the curriculum they offer. Usually a broad distinction is made between academic tracks, which prepare students for higher education, and technical and vocational tracks. Schools tend to specialize in one of those curricula, giving rise to academic schools on the one hand and technical/vocational and a few pure vocational schools on the other hand. Besides these ‘categorical’ schools, though, there are also so-called ‘multilateral’ schools, incorporating academic, technical and vocational tracks. But even within these multilateral schools the distinction between tracks remains, so there is no question of education in heterogeneous groups. The existence of these different types of school allows to test whether or not the segregation of academic and non-academic tracks into different schools sharpens the difference between students. Conforming the differentiation-polarisation theory, previous research has demonstrated that technical/vocational schools are characterised by a less study-oriented student culture than academic schools are (Van Houtte, 2004, 2006). It is assumed (Van Houtte 2006) that the organization of different tracks in distinct schools, as is the case in Flanders, makes differentiation between students in various tracks still more pronounced inasmuch as potential interaction with colleagues in other tracks is even more limited (Karweit and Hansell 1983). This contribution aims to examine whether inter-school segregation accentuates the difference in study-involvement between academic students and vocational students, or not. Therefore it questions more specifically how school type—multilateral or categorical—relates to academic students’ and vocational students’ study involvement, and whether the relation between track position and study involvement varies according to school type. We do not know of any studies dealing with this question. But based on the research into the differences between tracked and untracked systems, it can be assumed that vocational students in a multilateral school benefit from positive peer effects that might be gleaned from coming in regular contact with academic students. It is reasoned than that inter-school tracking will hinder this positive spill-over effect (Figlio and Page 2002; Zimmer 2003). Furthermore, inter-school tracking might be expected to encourage stigmatising of vocational students because of the firm distinction between academic and vocational students, arousing anti-school attitudes in the vocational students (Kulik and Kulik 1982). On the other hand, it is also possible that vocational students in multilateral schools take the academic students as a comparative reference group, leading to relative deprivation (Richer 1976). It can be assumed than that vocational students in multilateral schools get more aware of status deprivation—i.e. Davis’ frog pond effect (1966)—resulting in anti-school attitudes.

The analyses are based on a subsample of the Flemish Educational Assessment (FlEA), which are data gathered in 2004–2005 from 11.872 third and fifth grade students in a sample of 85 secondary schools in Flanders. To answer the questions at hand we selected four school types: there are 10 multilateral schools (offering academic, technical and vocational education) and 56 categorical schools of which 22 academic schools, 4 vocational schools and 30 technical/vocational schools. We limited our analyses to the 3758 academic students and the 2152 vocational students enrolled in these schools. By means of multilevel analyses (HLM6, Bryk and Raudenbush (1992)) we examined first whether vocational students’ and academic students’ study involvement varies according to the school type enrolled in, differentiating between multilateral schools and categorical schools. For these analyses vocational and academic students are dealt with separately. Secondly, taken vocational and academic students together we investigated by means of multilevel analysis whether the difference in study involvement between these two groups varies according to school type, again differentiating between multilateral and categorical schools. In each analysis we also take into account the students’ gender, age, socio-economic status, ethnicity and parental involvement.

The analyses show that vocational students enrolled in multilateral schools have a slightly lower study involvement than those enrolled in categorical schools. As for the academic students’ study involvement, it does not seem to make any difference whether they are enrolled in multilateral or categorical schools. As expected, attending an academic track or a vocational track makes a significant difference with respect to study involvement, academic students having a higher study involvement than vocational students. In addition our analyses show that the difference between tracks is larger in multilateral schools than in categorical schools: the interaction-effect between school type and track attended is statistically significant (p < 0.05). Our findings suggest that in multilateral schools vocational students compare themselves with the academic track students, leading to a higher awareness of status deprivation, resulting in even stronger anti-school attitudes.