With the shift towards learner-centred assessment, students are increasingly positioned as active participants in assessment as learning. Although student assessment has been widely shown to enhance learning outcomes, there remains limited conceptual clarity regarding the specific assessment-related knowledge, skills, and dispositions required for students to engage effectively in assessment. To address this gap, the present study developed a framework of student assessment literacy through an integrated methodological approach involving literature analysis, interviews, and expert consultation. The framework comprises four structural dimensions, namely assessment knowledge, assessment skills, assessment dispositions, and assessment metacognition, and is constituted by 12 second-level elements and 46 third-level sub-elements. These domains informed the construction of an integrative conceptual framework that captures the multifaceted nature of student assessment literacy. The study advances a refined conceptualisation by providing a comprehensive definition, clearly delineating core dimensions and elements, and identifying key components of students' assessment literacy for effective assessment engagement.
Artificial Intelligence (AI) is increasingly integrated into higher education admissions systems. This scoping review maps and critically synthesises peer-reviewed research published between 2020 and 2026 on AI in higher education admissions. Following PRISMA-informed search procedures across ERIC, Scopus, and Web of Science, 673 records were screened, and 25 studies met inclusion criteria. Descriptive analysis indicates a rapidly expanding body of research that is predominantly quantitative and geographically concentrated in the United States. Reflexive thematic analysis identifies three interconnected themes: (1) AI used by institutions in admissions systems; (2) AI used by applicants in performance-based assessments; and (3) AI, fairness, and the governance of admissions. The review argues that AI in admissions extends beyond efficiency gains, reshaping how merit is defined, how performance is interpreted, and how fairness is perceived and justified within high-stakes higher education decision-making systems.
Chile has undergone transformations in its higher education admission system, transitioning from an elitist model to a framework where over 90% of those who take the admission tests are eligible to apply to university. A key milestone is the implementation of the Higher Education Admission Test (PAES), which was designed to reduce academic barriers associated with socioeconomic gaps and historical inequities. Alongside tuition free policies and other equity in access reforms, Chile diversified its student population by incorporating alternative admission pathways. Nevertheless, all applicants must participate in the testing process to be considered for admission. Thus, it is essential to evaluate whether the tests maintain their predictive power despite these compositional shifts. This paper provides the first evidence on the predictive capacity of the PAES. By contrasting the regular admission group against all centralised admission pathways, the results show that predictive relationships remain stable across the entire admitted population.
