Student-Level Analytical Extract for Comparative Evaluation of an Adaptive Pedagogical Support Agent in an Operating Systems Course (Experimental vs. Control; N=182)

This dataset is a student-level tabular analytical extract used to support a comparative evaluation of the implementation of an adaptive pedagogical support agent (AIP-OS) in an Operating Systems course, under two conditions (experimental and control). The dataset contains 182 records (91 per condition) and aggregated performance variables: domain-level metrics on the [0,1] scale (interpretative, argumentative, and propositional), an aggregated average (AvgAcc) on [0,1], and the final grade (Final Grade) on the [0,5] scale. It also includes a deterministically derived high-performance indicator based on an operational threshold (G ≥ 3.8) and, for the experimental condition, a categorical trajectory-profile (TrajectoryProfile) variable reported descriptively. Records use coded identifiers and contain no direct personally identifiable information (e.g., names, emails, or addresses).
The dataset is intended to replicate the summaries (tables and figures) reported in the manuscript. However, it does not include the inputs required to perform an end-to-end audit of upstream operational logs or the full extraction/transformation pipeline used to build the analytical extract; therefore, between-condition contrasts should be interpreted as descriptive associations within the scope of the analytical package.