Score: 0

Differential Filtering in a Common Basic Cycle: Multi-Major Trajectories and Structural Bottlenecks in Exact Sciences and Engineering Degrees

Published: December 3, 2025 | arXiv ID: 2512.04285v1

By: H. R. Paz

Potential Business Impact:

Makes college math harder for some students.

Business Areas:
STEM Education Education, Science and Engineering

Universities often present the Common Basic Cycle (CBC) as a neutral levelling stage shared by several degree programmes. Using twenty years of longitudinal administrative records from a Faculty of Engineering and Exact Sciences, this study tests whether the CBC actually operates as a uniform gateway or as a differential filter across majors. We reconstruct student trajectories for 24,017 entrants, identifying CBC subjects (year level <= 1), destination major, time to exit from the CBC, and final outcome (progression to upper cycle, drop-out, or right-censoring). The analysis combines transition matrices, Kaplan-Meier survival curves, stratified Cox models and subject-level logistic models of drop-out after failure, extended with multi-major enrolment data and a pre/post 2006 curriculum reform comparison. Results show that the CBC functions as a strongly differential filter. Post-reform, the probability of progressing to the upper cycle in the same major ranges from about 0.20 to 0.70 across programmes, while overall drop-out in the CBC exceeds 60%. Early Mathematics modules (introductory calculus and algebra) emerge as structural bottlenecks, combining low pass rates with a two- to three-fold increase in the hazard of leaving the system after failure, with markedly different severity by destination major. Multi-major enrolment, often treated administratively as indecision, is instead associated with lower drop-out, suggesting an adaptive exploration of feasible trajectories. The findings portray the CBC not as a neutral academic foyer, but as a structured sorting device whose impact depends sharply on the targeted degree and on the opportunity to explore alternative majors.

Country of Origin
🇦🇷 Argentina

Page Count
24 pages

Category
Computer Science:
Computers and Society