The development of basic education has been guided by mechanisms that encourage improvements in the quality and efficiency of educational institutions. The implementation of public policies, restructuring the responsibilities of each administrative sphere and implementing evaluation and inspection systems are some examples of how to enhance educational performance. In Brazil, the National Education Plan (PNE) is one of the main instruments for conducting education, based on guidelines and goals. PNE monitoring reports highlight the deficiency in achieving goal seven, related to educational quality. This goal essentially aims to increase the Basic Education Development Index (Ideb), which must reach intermediate targets (every two years). Due to the difficulties in aligning with the PNE objectives within the expected deadline (2024), public and school managers must, therefore, seek new actions that help achieve improvements. Measuring school performance stands out as a means to foster these changes, as it allows performance standards to be assessed, not only academically, but also organizationally. Therefore, this work aims to evaluate the efficiency of Brazilian primary schools, as well as suggest approaches that facilitate school management and progress towards educational quality. With the exception of the introduction and conclusion, the chapters of this work were developed independently and structured as scientific articles, each one guided by a specific objective. Chapter 2 aims to discuss the issue of school efficiency based on a systematic review of academic literature, exploring information such as predominant techniques and variables. Based on a sample of 75 articles, the results demonstrated the use of Data Envelopment Analysis (DEA) as the main tool for evaluating school efficiency, with the socioeconomic context and the student-teacher relationship often associated with the models. Chapter 3 aims to calculate and compare the efficiency of Brazilian primary schools, defining reference points for inefficient institutions. Four DEA models are proposed with different combinations of variables and periods. Context-dependent DEA will also be used so that schools further away from the original efficiency frontier obtain practical goals, in stages. In general, it was found that efficiency becomes lower when using Ideb as an output, and can be reduced by up to 21.7% in relation to other models that consider students' grades in Portuguese and Mathematics as outputs. Although the initially projected goals were high, changes in the evaluation context resulted in lower intermediate values and, consequently, easier to achieve, maintaining the incentive for inefficient institutions. In turn, using regression with panel data, chapter 4 aims to verify the association between the type of administrative dependence (state or municipal) and school efficiency, measured by a dynamic DEA model. A positive relationship was identified when schools are under state control, as well as when located in urban areas. Finally, it was concluded that the approaches suggested in the aforementioned chapters can facilitate school management (at an organizational and public level) and allow for progress towards educational quality.