LPR 2026 v1.0: Data and Replication Package for the Lexicographic Prize Ranking

This repository contains the data and replication materials for LPR 2026 v1.0 (Lexicographic Prize Ranking), a transparent, non-compensatory, outcome-based framework for measuring elite prize-recognized scientific distinction at the university level. LPR ranks institutions lexicographically using three tiers: (1) Gold — unique qualifying Nobel laureates in Physics, Chemistry, Physiology or Medicine, and Economic Sciences, credited according to institutional affiliation at the time of the prize announcement; (2) Silver — Fields Medal affiliations at the time of the award; and (3) Bronze — Nature Index Count for articles published in Nature and Science during calendar year 2025. The framework does not calculate a weighted composite score. Institutions are ordered first by Gold, then by Silver when Gold is tied, and finally by Bronze when both Gold and Silver are tied. This makes the hierarchy explicit, non-compensatory, transparent, and auditable. The replication package contains the frozen institution-level inputs, laureate-level Nobel audit, complete Fields Medal audit, institutional crosswalk, candidate-universe data, validation statistics, sensitivity analyses, and executable Python code reproducing the frozen Top 100. Comparative validation uses the frozen matched samples for QS World University Rankings 2026 (n=29, Spearman ρ=0.633), Times Higher Education World University Rankings 2026 (n=40, ρ=0.766), and ARWU 2025 (n=45, ρ=0.830). The empirical ranking is frozen as LPR 2026 v1.0 (17 August 2026). Post-freeze corrections are restricted to documented data or institutional-normalization errors and require a new version. Associated manuscript: “Beyond Composite University Rankings: A Non-Compensatory, Outcome-Based Framework for Measuring Scientific Excellence.”