Maria Eugénia Ferrão

Value-added (VA) models are widely used to evaluate educational effectiveness, yet their conceptual and methodological development remains insufficiently systematized. This study conducts a scoping review, complemented by an exploratory bibliometric analysis of studies published between 1971 and 2024, focusing on the seminal contributions of Hanushek (1971) in Education economics and Bryk and Weisberg (1976) in educational statistics. The findings reveal persistent gaps in the literature, notably the limited evidence from developing countries and the underexplored roles of school leadership and causal inference within VA frameworks. The review highlights a growing diversification of methodological approaches alongside an increasing alignment with accountability policies. From a policy perspective, the results underscore the need for context-sensitive VA applications, stronger school leadership, improved longitudinal data systems, and closer collaboration between researchers and policymakers to support more effective and equitable educational interventions.

This scoping study synthesizes recent developments in artificial intelligence in education (AIEd), addressing knowledge structures, research priorities, learning theories, ethical considerations, and impact evaluation. A bibliometric analysis of 31 review articles (2019-2023) indexed in Scopus (Social Sciences) was conducted using VOSviewer and complementary statistical methods. The corpus is concentrated in the United Kingdom, Hong Kong, China, Germany, the United States, and Taiwan, with Asian countries accounting for 48.4%. Multidisciplinary journals attract 4.4 times more citations than education-only outlets. Findings indicate a marked rise in AIEd research, primarily led by education-affiliated scholars, yet often lacking robust pedagogical grounding and systematic impact assessment. The prevalence of small samples, limited quantitative rigor, and inconsistent contextual reporting constrain generalizability and inference. To enhance educational relevance and fairness, the field should be anchored in pedagogical frameworks and advanced through collaborative efforts that build stronger theories, methods, and practices responsive to diverse educational needs.

Composite multidimensional indices are broadly used to measure child poverty and social exclusion. Many of such indices are based on EU-SILC data or similar large scale complex sampling surveys, with the household as unit of analysis. Indicators related to households with or without children may quantify the intended attribute differently depending on the household structure and characteristics of individuals, potentially compromising the assessment. We conducted statistical modelling and hypotheses tests using a two-parameter logistic item response model (IRM) and the likelihood-ratio test for DIF verification. Methods were applied to 2020 EU-SILC Portuguese data comprising 11,367 households representing a population of 4,099,052. Statistical analysis have allowed for the survey sampling design. Our findings demonstrate differential item functioning in the assessment material deprivation in households with or without children.

This article aims at a better understanding of the Brazilian education system’s performance concerning the quality and equity over the decade 2007–2017. It examines the extent to which students’ sociodemographic characteristics are related to schooling trajectory without failure in primary education and how such relationships have changed over time. Multilevel logistic models are applied to cross-sectional student assessment data (Prova Brasil), considering the hierarchical structure of four levels. The total number of students covered is 12.4 million. The results suggest a pattern of educational inequity marked by socioeconomically disadvantaged status, gender, and self-declared race/ethnicity. The analyses show that the gender gap increased and the differentials by socioeconomic status and race/skin colour decreased over the decade. The estimates also suggest that the effect of school socioeconomic composition was reduced, but the effect of the proportion of repeating students per school on the individual probability of success was reinforced. In addition, the probability of students’ success varies randomly across schools, municipalities, and states, and such educational disparities across states are increasing over time.

Os valores omissos representam um problema frequente no processo de análise de dados. Neste artigo foram comparados seis métodos distintos de imputação, disponíveis no software R e avaliado o seu desempenho em conjuntos de da­dos relacionados com a área da educação. Foi estudada uma amostra de 20408 estudantes para testar os seis algoritmos em quatro conjuntos de dados gerados por simulação com diferentes percentagens de valores omissos, considerando 5%, 10%, 15% e 20% nas variá­veis de interesse. Foram explorados métodos de imputação simples (Média, Mediana e Moda), métodos baseados em aprendizagem automática (kNN e bPCA) e um método de imputação múltipla (MICE). Foi avaliado o desempenho de cada método calculando os respetivos erros de imputação através as métricas RMSE e MAE. Os resultados obtidos mostram que a imputação pela Moda forneceu quase de forma constante menores valores de erro.

Agradecemos aos professores e pesquisadores que contribuíram para o número 121 com suas avaliações, dentro de suas áreas de especialidade. Agradecemos também aos editores associados que funcionaram como editores de seção, contribuindo no processo de triagem, na avaliação dos artigos em inglês e espanhol, e com seus votos de minerva quando os pareceres foram díspares.

In this study, clustering is conceived as an auxiliary tool to identify groups of special interest. This approach was applied to a real dataset concerning an entire Portuguese cohort of higher education Law students. Several anonymized clustering scenarios were compared against the original cluster solution. The clustering techniques were explored as data utility models in the context of data anonymization, using k-anonymity and (ε, δ)-differential as privacy models. The purpose was to assess anonymized data utility by standard metrics, by the characteristics of the groups obtained, and the relative risk (a relevant metric in social sciences research). For a matter of self-containment, we present an overview of anonymization and clustering methods. We used a partitional clustering algorithm and analyzed several clustering validity indices to understand to what extent the data structure is preserved, or not, after data anonymization. The results suggest that for low dimensionality/cardinality datasets the anonymization procedure easily jeopardizes the clustering endeavor. In addition, there is evidence that relevant field-of-study estimates obtained from anonymized data are biased.