Academic self-concepts of ability are key factors in promoting education and learning throughout students’ school career. Yet we know little about their structure and structural change when students leave high school to face novel academic tasks. The present study investigated the structure and structural change of first-year students’ study-related self-concepts of ability. Data stems from a longitudinal study with two measurement points covering the initial study phase (t1: N = 341; age: M = 21.6; SD = 3.56; 57.5% female). Self-concepts were assessed regarding the participants’ study program and four of its subordinate subjects. Confirmatory factor analyses and structural equation models were used to compare structural models and to investigate structural stability and directional effects. Results support the assumption of multidimensionality (i.e., distinct self-concepts for different subjects) and hierarchy with a generic field-of-study-specific self-concept at the apex. Specifying generic field-of-study-specific self-concept as a method factor (i.e., indicated by both subject-specific and field-of-study-specific items) was most consistent with theoretical assumptions. The structural model was invariant over the first months at university. Generic field-of-study-specific self-concept and subject-specific self-concepts largely developed independently from one another. The results emphasize the recently suggested conceptualization of generic self-concept as a method factor to reflect self-concept hierarchy. Self-concepts were structurally stable over time. Several significant horizontal effects (i.e., stability within subjects) suggest that students align their self-concepts closely to the curriculum they encounter in educational contexts and, therefore, may benefit from ample feedback on their performance to develop appropriate subject-specific self-concepts.