HRM.Source:1 - Grounds and their limits

This framework is a conceptual and methodological construction from the following different contributions. A professional prescription, a theory, a reported association and a constructed example retain different evidential roles. Dates identify editions and observations, not the modernity of every underlying model.

Source	Contribution used and limit
Boxall and Purcell, Strategy and Human Resource Management, 5th ed., 2022, publisher preview, selected introductory/chapter-one pp. 5–12 and 14–30	Work organization and employment practices, plural interests, viability and legitimacy ground the connected field account. The selection is not a full-book synthesis and supplies no current AI capability limit.
CIPD, Profession Map, talent management and employee relations	Professional breadth and differentiated operations; the field map is not an executable Method or proof that each specialism is a separate discipline.
O’Connor et al., Integrated Talent Management Framework, 2025	A conceptual alternative concerning access, two-sided fit and some trajectories outside the employer. Predicted benefits and proposed practices are not established effects.
van Zelderen et al., Paradox of Inclusion, online 2024, issue 2025	Experiments and a cross-sectional field study qualify simple assumptions about talent-pool size and experienced inclusion. They supply neither a universal access rule nor a secrecy prescription.
SIOP, Principles for the Validation and Use of Personnel Selection Procedures, 2018, selected pp. 5–8, 33 and 38–39; OPM, Job Analysis and Work Samples and Simulations	Work analysis, warranted use of assessment evidence and conditional choice of work samples. Neither source establishes local validity or one universal instrument. OPM’s federal setting does not create appointment authority elsewhere.
SIOP, AI-based assessments for employee selection, January 2023, introduction and selected operational considerations, pp. 20–24	AI scoring still needs evidence for its particular use; changes and data sources matter. The document excludes broader employee–AI collaboration. Its upload path does not make it a 2024 or 2026 capability study.
ILO, Towards pay equity, 2026, chapter 5, pp. 30–34; CIPD, Reward: an introduction, 1 October 2026	Job-demand evaluation and the broader reward combination supply distinct operations. Local institutional, market and implementation grounds remain necessary.
Acas, Code of Practice on disciplinary and grievance procedures, effective 2015 edition, foreword and paragraphs 1–47, read October 2026	Great Britain procedural comparator with stated exclusions. The 2026 consultation draft is not used as effective law. The receiving act requires the applicable current process.
ICO, employment records, selected purpose, lawful-basis and consent sections	A jurisdiction-bound distinction between collecting information and another permitted use. Guidance under legislative review does not settle a current local decision.
Li, Fu, Chadwick and Harney, HRM and employee wellbeing, online 2023, issue 2024	The study uses Ireland’s 2009 employee survey, 4,823 complete cases, and distinguishes wellbeing dimensions. Cross-sectional, single-source evidence limits causal inference. It challenges automatic mutual-gain claims, not every training or reward practice.
WHO, Mental health at work, 15 September 2026; 2022 guideline overview	Work-condition prevention, individual support, accommodation and return are distinct. The factsheet’s main text and the guideline overview do not supply the complete guideline, all occupational hazards or a clinical protocol.
Endocrine Society, Adrenal fatigue and Adrenal insufficiency, January 2022 patient resources	Corrects an unsupported diagnostic label and separates a real clinical condition. The framework uses no disease-specific diagnosis or treatment recommendation.