International Journal of Artificial Intelligence and Education Technology

All people must have access to educational opportunities that meet their needs through Sustainable Development Goal 4. The goal requires education systems to obtain sufficient funds and use their resources properly. The relationship between public education spending and student academic performance remains disputed because countries achieve different results from their spending levels. This study uses country-level PISA 2022 scores, the first international assessment data published after COVID-19, to analyze public education expenditure as a share of GDP together with the pupil–teacher ratio and GDP per capita in relation to student academic performance across three subjects. Public education funding as a percentage of GDP is not statistically associated with PISA scores across 35 countries (r = −0.095, p = 0.586). The pupil–teacher ratio is a strong negative predictor of student performance (βˆ = −4.097, R2 = 0.312, p < 0.001). A three-variable regression model combining expenditure share, pupil–teacher ratio, and GDP per capita explains 59% of cross-country score variance (R2 = 0.592). High-income economies dominate the upper achievement tier, but several systems, notably Estonia and Poland, substantially outperform their GDP-predicted scores. The results indicate that education systems should focus resources on strengthening instructional capacity.

Student-success analytics has moved from experimental prediction toward an institutional capability for reduc-ing attrition, allocating support resources, and improving digital learning governance. This paper develops a business-oriented early-warning model for education technology environments in which predictive performance, interpretability, intervention priority, and governance are treated as joint design requirements. The study uses a public student-success dataset from a higher education institution and evaluates decisive outcome prediction for dropout and graduation, while preserving a wider discussion of the enrolled group as an unresolved operational state. The proposed model combines a transparent predictive layer, a risk-to-action prioritization layer, and a governance layer that restricts how predictions are translated into student support decisions. The results show that a parsimonious logistic specification can provide competitive performance compared with more complex tree-based models, while producing clearer accountability for academic advising and digital student-success units. The discussion argues that student-success technology should not be judged by accuracy alone, but by whether the analytics pipeline produces timely, explainable, privacy-aware, and operationally usable support signals.

Early identification of students at risk of academic failure is a persistent challenge in educational technology, with direct implications for student retention, institutional equity, and the allocation of support resources. Although supervised machine learning has been widely applied to student outcome prediction, the relative merit of competing algorithm classes and the degree to which demographic and behavioural features contribute predictive power beyond prior academic assessments remain incompletely resolved in the secondary school context. This paper presents a structured comparative evaluation of five supervised classifiers trained on a rich combination of periodic grades, attendance records, socio-demographic characteristics, and lifestyle indicators drawn from secondary school students. A dual importance analysis—combining impurity-based measures with held-out permutation importance—disentangles the distinct predictive roles of grade trajectories, absenteeism, parental background, and lifestyle variables. Ensemble methods demonstrate consistent superiority across all evaluation criteria, with prior periodic assessments and attendance emerging as the dominant predictors. Parental education level introduces a socioeconomic gradient that operates independently of student-controlled factors, generating structural inequities that standard grade-monitoring systems are unlikely to address. These findings provide both a methodological benchmark for secondary school prediction tasks and practical guidance for institutions designing equitable and evidence-based early warning interventions.