Agustini
Doctoral Program in Educational Management, Graduate School, Universitas Negeri Jakarta, Jakarta, Indonesia
Teguh Trianung DS
Lecturer in Educational Management Graduate School, Universitas Negeri Jakarta, Jakarta, Indonesia
Background: Professional Certification Bodies (Lembaga Sertifikasi Profesi, LSP) conduct competency assessments within Indonesia’s national certification system under the National Professional Certification Agency (BNSP), so the credibility of certificates depends on the quality of their governance. Studies that systematically evaluate LSP governance remain limited. This study evaluates the implementation of LSP governance programs and constructs an evaluation model integrating the Context–Input–Process–Product (CIPP) model with Good Corporate Governance (GCG) principles.
Methods: A qualitative evaluative multi-site design was applied to three LSPs in Jakarta: a first-party LSP (Universitas Mercu Buana), a second-party LSP (BRI), and a third-party LSP (Microfinance Indonesia). Data were collected through in-depth interviews with 18 informants (July–October 2026), observation, and document study, and analysed in NVivo 14 using deductive and inductive coding, within-site analysis, and cross-site analysis.
Results: All three LSPs had policy foundations, resources, procedures, and competency recognition mechanisms aligned with BNSP regulations. Gaps were found in the currency of national competency standards (SKKNI) and assessment materials, equitable distribution of assessor assignments, consistency of procedure implementation, conflict-of-interest control, affordability, accommodation for participants with disabilities, and post-certification evaluation. Cross-site synthesis produced a model mapping four CIPP dimensions against five GCG principles into 20 evaluation aspects, complemented by a CMR (Achievements, Problems, Recommendations) approach.
Conclusions: LSP governance quality cannot be judged by administrative compliance and certificate counts alone, but by policy relevance, resource management, assessment integrity, and the credibility of results. The resulting model is conceptual and operational and requires testing in more diverse LSPs.