This paper provides findings of research on school effectiveness and discusses implications for evaluation in Brazil and Portugal. Most findings reported over the last decade have been published in Brazilian or Portuguese refereed journals. Thus, a brief literature review of such studies enables that knowledge to reach international scholars and researchers. The magnitude of school effects obtained from longitudinal and cross-sectional data modelling is presented and discussed. In particular, a value-added approach based on multilevel models is used to explore consistency across models with different controlling variables and across different curricular contents, and stability over time. The results show a great deal of regional disparities regarding educational outcomes that are related to pupils' socioeconomic and prior achievement heterogeneity. In addition, evidence is given for stronger school effects in primary than in lower secondary education, larger amplitude of the variance partition coefficient in primary than in lower secondary education, stronger consistency of value-added estimates across models with different controlling variables and moderate consistency across different curricular contents. Weak-to-moderate stability of value-added estimates is also shown when yearly measured compared to when measured by cycles of education, and moderate-to-strong stability when measured in different curricular contents. Recommendations are outlined in terms of how the results could be used to mentoring evaluation in Portuguese speaking countries and enhance school improvement.