Doctoral education in library and information science (LIS) is a workforce-development mechanism through which the profession builds research capacity, academic leadership, and evidence-informed institutional practice. Yet little is known about how structural opportunity, professional capital, and institutional access conditions shape expressed interest in doctoral LIS study within national librarian populations. This article develops an interpretable predictive segmentation framework using the Philippine Librarians Census to classify expressed doctoral study interest and translate empirically derived workforce segments into ethically bounded recruitment personas. The study uses a cross-sectional, sequential predictive segmentation design combining descriptive workforce profiling, interpretable machine-learning classification, calibration and subgroup diagnostics, segmentation analysis, and persona translation safeguards. The findings show that expressed doctoral study interest is patterned across educational, professional, institutional, and geographic contexts, supporting the use of predictive analytics as a planning tool rather than as an admissions, ranking, or individual forecasting system. The article contributes to LIS workforce research by demonstrating how national professional census data can support doctoral pipeline planning while preserving inferential restraint, interpretability, fairness awareness, and non-exclusionary governance.

Aquest document presenta una anàlisi visual de la posició ideològica de diversos partits italians a partir de dades CHES, elaborada amb R.