The transition from secondary to higher education represents a critical point in academic trajectories, particularly in programmes with a strong emphasis on basic sciences. Across different higher education systems, introductory Mathematics and Physics courses consistently concentrate high rates of early failure and attrition, yet most available evidence relies on cross-sectional analyses or limited time spans. This study presents a longitudinal evaluation of pre-university preparation based on early academic outcomes in Mathematics and Physics, conceptualised as "sensor" courses of initial academic demands. Using complete administrative records from a large public university in Argentina, the analysis covers entry cohorts from 1980 to 2019 with census-level coverage and a population-based approach. Pre-university preparation is operationally defined as cohort-level compatibility between students' prior educational background and the functional demands of introductory university coursework, observed through first-attempt outcomes. For each cohort and by type of secondary school (public or private), we estimate the probability of course approval, the probability of non-attempt (enrolment without evaluative participation), and the public-private success gap. The results reveal consistent long-term patterns: a gradual decline in early approval probabilities, a sustained increase in non-attempt behaviour, and the persistence of moderate but stable public-private gaps. These findings point to structural changes in the articulation between secondary education and higher education rather than short-term fluctuations or individual-level effects. The study contributes to the international literature on educational evaluation by providing rare long-horizon longitudinal evidence from an Ibero-American context.

翻译：从中学到高等教育的过渡是学术轨迹中的关键节点，尤其在强调基础科学的专业中。在不同高等教育体系中，数学与物理入门课程持续呈现较高的早期不及格率和辍学率，但现有证据大多依赖于横截面分析或有限的时间跨度。本研究基于数学与物理早期学业成果——这些课程被概念化为初始学术要求的“传感器”——对大学前准备进行了纵向评估。利用阿根廷一所大型公立大学的完整行政记录，分析覆盖了1980年至2019年的入学队列，采用普查级覆盖范围和基于人群的研究方法。大学前准备在操作上定义为学生先前教育背景与大学入门课程功能需求之间的队列层面匹配度，通过首次尝试的学业成果进行观测。针对每个队列并按中学类型（公立或私立），我们估算了课程通过概率、未尝试概率（注册但未参与评估）以及公私校成功率差距。结果揭示了长期一致的模式：早期通过概率逐渐下降，未尝试行为持续增加，以及中等但稳定的公私校差距持续存在。这些发现指向中等教育与高等教育衔接的结构性变化，而非短期波动或个体层面效应。本研究通过提供伊比利亚-美洲背景下罕见的长期纵向证据，为国际教育评估文献作出了贡献。