tional Assessment and School Improvement in Indonesia: From Accountability to Diagnostic Use

Abstract

Indonesia replaced its high-stakes school-leaving examination with a national assessment designed on deliberately different principles: it samples learners rather than examining all of them, reports results at the level of the school rather than the individual, measures literacy and numeracy competence rather than curricular coverage, and surveys character and the learning environment alongside cognitive outcomes. The design is diagnostic by construction, yet the uses to which its results are put are determined downstream by local authorities and schools, and those uses may reintroduce the incentives the reform was intended to remove. This article aims to analyse the design logic of the national assessment, to identify the mechanisms through which a diagnostic instrument may acquire accountability stakes, and to assess what conditions are required for its results to inform improvement. The study is a qualitative policy analysis combining document analysis of the regulatory framework with critical synthesis of the assessment and accountability literature; no field data were collected. The analysis finds that the design successfully removes the individual-stakes mechanism that drove examination-oriented teaching, that the residual risk lies in institutional stakes attached to publicly reported school results, and that the capacity to interpret and act upon diagnostic data is the binding constraint on improvement. It recommends protection of the low-stakes character of reporting, investment in school-level data interpretation capability, and explicit separation of diagnostic reporting from any subsequent selection instrument