Background: Research on individual learning approaches (or learning styles) is split in two traditions, one of which is biased towards academic learning, and the other towards learning from direct experience. Aims: In the reported study, the two traditions are linked by investigating the relationships between school-based (academic) and work-based (experiential) learning approaches of students in vocational education programs. Samples: Participants were 899 students of a Dutch school for secondary vocational education; 758 provided data on school-based learning, and 407 provided data on work-based learning, resulting in an overlap of 266 students from whom data were obtained on learning in both settings. Methods: Learning approaches in school and work settings were measured with questionnaires. Using factor analysis and cluster analysis, items and students were grouped, both with respect to school- and work-based learning. Results: The study identified two academic learning dimensions (constructive learning and reproductive learning), and three experiential learning dimensions (analysis, initiative, and immersion). Construction and analysis were correlated positively, and reproduction and initiative negatively. Cluster analysis resulted in the identification of three school-based learning orientations and three work-based learning orientations. The relation between the two types of learning orientations, expressed in Cramer's V, appeared to be weak. Conclusions: It is concluded that learning approaches are relatively context specific, which implies that neither theoretical tradition can claim general applicability.