A ponte entre a lei e o código: uma metodologia de modelagem multinível para a efetividade da transparência governamental

Abstract

Context: Effective social control and the oversight of public spending in Brazil are critical processes hampered by the inconsistency and lack of standardization of the data provided on government transparency portals. This absence of a technical standard results in interoperability barriers, generates inefficiencies, and creates loopholes for irregularities. Although legislation (LAI - Access to Information Law) requires the disclosure of data, its nationwide application is hindered by the fundamental gap between the legal framework, written in legal language, and the absence of an information architecture capable of translating these requirements into technical specifications in a scalable, robust, and replicable manner. Objective: This work aims to overcome this barrier on two fronts: first, to characterize the state of the art in the analysis of transparency portals, identifying the challenges and existing solutions in the literature; second, to present and evaluate a multi-level modeling methodology, inspired by the openEHR standard from the healthcare domain, materialized in a functional prototype, the Archetype Generator, designed as a 'bridge' for the standardization of Electronic Government Data Records (REDG). Method: The methodology adopted a two-phase approach. First, a Systematic Literature Mapping (SLM) consolidated the state of the art and empirically validated the research gap. Subsequently, these findings guided the design of the multi-level modeling methodology and the development of the prototype. Its effectiveness and applicability were investigated through an exploratory case study with experts (researchers and market professionals), which evaluated the feasibility, clarity, and usability of the approach in a practical scenario of government data modeling. Results: The systematic mapping analyzed 609 articles and selected 52 primary studies, revealing that no research addressed the three pillars of legal compliance, usability, and technical standardization in an integrated manner. The application of the methodology in the case study demonstrated high perceived usability (mean 4.36), a large effect size (Hedges’ g=1.25), and unanimous clarity (100%) in the generation of technical artifacts (JSON/SQL), with 80% of participants completing the task without assistance. Conclusion: The research presents exploratory evidence that the multi-level modeling methodology constitutes a promising and viable approach for public institutions to overcome the data standardization challenge. By offering a practical guide and a tool whose initial effectiveness was demonstrated, this work empowers organizations to align their technical implementations with legal requirements, establishing the foundation for greater interoperability, consistency, and, consequently, more transparency and effectiveness in social control.