A diploma in higher education has a significant impact on an individual’s life course. Higher educated individuals may have a higher income, a better job, and a lower risk of poverty than lower educated individuals. They also score higher on many non-economic outcomes, such as overall health. However, one of the most consistent findings in the sociology of education is social inequality in educational attainment. A student from an upper-class family, for example, has a remarkably higher likelihood to attend higher education than a student from a working-class family. Therefore, education plays a crucial role not only in social mobility, but also in the reproduction of social inequality. To understand the process of educational attainment, sociologists investigate the educational expectations that students express in secondary education. Students have certain expectations about what they will do after secondary education, such as attend university or enter the labor market. Students’ expectations are theoretically postulated and empirically confirmed as a central link between ascribed and achieved characteristics of the student on the one hand, and status attainment as an adult on the other hand. Expectations are believed to guide behavior. High educational expectations increase the odds to attend higher education, as these expectations as they positively influence motivation and efforts in secondary education. However, students’ expectations are influenced not only by family background and school performance, but also by both the school they attend and the broader education system. The plethora of research on educational expectations for higher education focuses solely on the individual level, such as looking at socialization processes within families. This focus on the individual level is attributable to “the Wisconsin Model” of the 1960s, which is still the (explicit or implicit) departure point for most current research. The model posits that students’ expectations are the result of socialization by significant others. The three most important groups of significant others are parents, peers, and teachers, although current research mainly focuses on parents. The Wisconsin Model attaches importance specifically to students’ perceptions of the significant others’ expectations. Although other research traditions, such as Sociological Rational Choice theory and theories on cultural and social capital, were advanced to explain the formation of expectations, they have all been preoccupied with mechanisms within the family. While this body of earlier research has enriched our understanding of individual-level processes, it individualizes and de-contextualizes the educational decision-making of students. However, as students form their expectations of postsecondary pathways while attending secondary education, it is important that the context of a students’ secondary school and the features of the education system are also considered. Educational expectations not only differ according to the social background of students, but also vary according to the characteristics of the school and of education systems. Although the secondary education system is central in the process of sorting and allocating students to positions in society, recent research pays little attention to how school and system characteristics influence students’ expectations. Nevertheless, school features matter for several student outcomes, as shown by the School Effects Research tradition. Moreover, in the sociology of education a substantial amount of research has identified what the consequences are for students’ outcomes of attending a particular organization within the secondary education system, particularly with regard to an organization with a rigid division of pupils into different types of curricula, which is called “tracking”. Hence, the main objective of this research is to gain a better understanding of the social inequality in expectations by transcending the existing (explanations for) individual effects and by considering the meso level – the school – as well as the macro level – the education system. To this end, the empirical part of this dissertation comprises four quantitative studies that fruitfully combine different research traditions, integrating School Effects Research and research on tracking into the Wisconsin Model. Our research uses primary data from the unique, longitudinal International Study of City Youth (ISCY), which follows 10th grade students from multiple cities around the world during their transition from secondary education to higher education or the labor market. One of the four studies is a cross-national study, comparing 10th grade students’ expectations in four European cities. This study contributes to the research field on expectations, as very few cross-national studies – almost none of which pay attention to the school level – still characterize the field. The other three studies, one of which is longitudinal, focus on the data of more than 2000 youngsters in the city of Ghent in Flanders (Belgium). The Flemish education system is an interesting case study to examine educational expectations, as it combines a highly tracked and segregated secondary education system with an open-access higher education system. These three studies demonstrate how the composition of the schools’ student body influences students’ expectations and students’ enrollment in higher education. A student attending a school whose students are predominantly from a high SES background will have more ambitious expectations and will be more likely to be enrolled in higher education than a student from a school with mainly low SES students. Moreover, by looking at explanatory school processes, this dissertation’s studies transcend other research that offers only a mechanical understanding of the association between school composition and expectations. The Wisconsin Model, which attributes a central role to the expectations of significant others such as peers and teachers, provides the inspiration for relevant school processes. This dissertation transcends this individual-level approach of teacher and peer effects by looking at the culture of teacher expectations in one study and the culture of peer expectations in another. In contrast to mainstream research, our measurement of significant others’ expectations are based on objective data retrieved from the teachers and peers themselves and not on subjective student perceptions of their teachers’ and peers’ expectations. The study on teachers’ culture surpasses the traditional conceptions of “the invisible teacher” or “the biased teacher” and shows how the shared expectations of teachers from the same school can compensate for the detrimental effects of low SES composition on expectations. Moreover, the longitudinal study confirms that students’ expectations play a role in eventual higher education attendance five years later. In addition to the expectations of the 10th graders, the shared expectations of their peers at school also matter. Students attending schools with high peer expectation cultures are more likely to attend higher education and to prefer university over institutes that are regarded as less prestigious compared to students in schools with low expectation cultures. However, ambitious peers also have negative effects on some groups of students because of comparative group processes. The cross-national study highlights the fact that the school SES composition effects is not found in all secondary education systems, only in highly differentiated systems. Consequently, this school composition effect is an additional source of inequality in education systems that are already unequal. We therefore question the assumption that tracked systems effectively divide students into homogeneous groups to prepare and allocate them to distinct postsecondary pathways. Generally, the assumption is that these homogeneous groups would benefit effective and efficient teaching. The question of whether students’ expectations align with the intended objectives of the system in these tracked systems needs to be addressed. This dissertation shows that in Flanders, which has a stringent tracked system, the vocational track has the lowest homogeneity of expectations, the intended efficiency is therefore questionable. This unmet goal of tracking can be attributed to the unequal selection of students to tracks and the cascade system, as well as to the information deficits about the importance of secondary education for postsecondary pathways. As national and supranational policymakers endeavor to increase higher education attendance to strengthen the knowledge society, one easy solution would seem to be to raise expectations. However, this strategy has been criticized as having no – and potentially detrimental – effect when students are not also offered the means to turn their expectations into reality. Moreover, the strategy individualizes and decontextualizes unequal access to higher education. Instead, based on results of the cross-national study, raising student engagement, which is constructed in the interaction between the student and the school context, has been proposed. This dissertation also strongly recommends acknowledging the positive role that teachers can play. The results show the overall, long-lasting effects of school cultures and suggests that schools could create school cultures to make the transition to higher education desirable and feasible (‘college going cultures’). Finally, a critical look at the supposed benefits of tracking at an early age is recommended, especially regarding the adverse effects for students and teachers in vocational education. In summary, this quantitative study examines the effects of school composition and explanatory school processes and considers the effect of tracking on the formation and realization of educational expectations. This dissertation emphasizes that students’ expectations are not formed in a vacuum; schools and education systems also play a substantial role in shaping them.