Educational assessment and indicator-based governance in Brazil: a comparative analysis of SAEB 2025, SINAES 2023, and the CAPES quadrennial assessment

Abstract

This article comparatively analyzes public policies for educational evaluation in Brazil, investigating how the Basic Education Assessment System (Saeb), the National System for the Evaluation of Higher Education (Sinaes), and the Capes Quadrennial Evaluation operationalize the concept of "governance by indicators." The study aims to analyze the architecture of these policies, identifying epistemological convergences and regulatory specificities regarding their purposes, objects, instruments, indicators, institutional uses, and implications for management. Methodologically, it employs a bibliographic, documentary, and comparative research design with a qualitative approach, supported by a concise descriptive quantitative component. The empirical scope examines aggregated secondary data from Saeb 2025 (traditional upper secondary education), Sinaes 2023 (Preliminary Course Concept - CPC band distribution), and Capes 2021–2024 (program grades distribution). The findings reveal that the three systems converge by translating educational complexity into standardized metrics that function as devices for state regulation, political coordination, and behavior induction across schools, undergraduate courses, and postgraduate programs. However, their specificities lie in their scale, purpose, and the nature of the institutional constraints they produce: Saeb acts primarily as a network diagnosis tool; Sinaes is tightly coupled with supervision and accreditation; and Capes is structured around academic productivism and prestige stratification. The study concludes that, while essential for government planning, indicators do not capture the full multidimensionality of educational quality, demanding a contextualized interpretation that considers the social and institutional inequalities of the Brazilian educational landscape.