During lectures, some students are continuously focused and attentive, whereas others tend to be bored, jittery, or inattentive. The same might happen when students are given student-activating assignments. Some students simply love one type of instruction, whereas others tend to resent it. Moreover, it is not the context itself, but the context as it is perceived by the student that affects learning. This study sets out to discover how students view instruction, based on their experiences with a particular setting, and aims to investigate the effects of ''context-by-student'' characteristics, such as students' course experiences, and the effects of ''student-by-context'' features, such as approaches to learning. The research has a quasi-experimental design and compares a lecture-based setting with student-activating teaching methods (i.e. self-discovery learning by means of authentic tasks with the help of a scaffolding approach); followed by one of four assessment methods: a multiple-choice test, a case-based examination, a peer assessment or a portfolio assessment. Data (N ¼ 579) were collected by means of validated questionnaires, comprising the ASSIST and LSQ instruments and the course experience questionnaire. Results demonstrate that students' appraisal of the assessment method, their course experiences and the teaching methods that are dominant in the setting contribute positively to students' appraisal of instruction. Interestingly, this conclusion applies to both the lecture-based and the student-activating groups and to both conventional assessment methods and new modes of assessment. Moreover, deep approaches to learning, monitoring studying and/or organized studying, as well as preferences for courses that support understanding, positively affect students' appraisal of student-activating teaching methods. In short, both student-by-context and context-by-student features prove to be strong predictors of students' appraisal of instruction and, as a consequence, student learning.