Avaliação de desempenho da educação superior brasileira por análise envoltória de dados e conceito de porte relativo

• Ano de defesa: 2018
• Autor(a) principal: Leal, Marina Reis Deusdará
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Rio Grande do Norte; Brasil; UFRN; PROGRAMA DE PÓS-GRADUAÇÃO EM ENGENHARIA DE PRODUÇÃO
• Palavras-chave em Português: Avaliação do ensino superior; Análise de desempenho; DDEA; Índice Malmquist
• Link de acesso: https://repositorio.ufrn.br/handle/123456789/47541

This study aims to propose new approaches to performance evaluation based on Data Envelopment Analysis (DEA) so that government agencies can evaluate Brazilian higher education efficiency. Data from the Federal Institutions of Higher Education that obtained ENADE grades in every year from 2013 to 2016 were used to illustrate the application, which totaled 70 institutions distributed in all macroregions of the country. Recent public policies promoted a significant campuses expansion over the country. Since public resources are involved, it is important to verify if those institutions can provide a quality service corresponding to the investments made. The Relative Size concept is a distance-based analysis that comes from the evolution of the Superefficiency DEA with the common weights model to indicate which DMU has the largest combination of inputs and outputs. In addition to these models, the Dynamic DEA (DDEA) and classical models are applied, all considering both constant returns to scale and variable returns to scale. Malmquist Index is also used to observe the evolution between periods completing the analysis. The educational process is evaluated through the interaction between three input variables (quantities of teachers, administrative staff and enrollments), an intermediate product (new admissions) and two outputs (number of students awarded certificates, and the average grades of the National Student Performance Exam - ENADE). The analysis covered two perspectives: annual and global performances. The second ones can be obtained by the DDEA model and it covers two triennial periods: from 2013 to 2015, and from 2014 to 2016. The global perspective allows us to observe DMUs performance in relation to ENADE’s triennial cycle, which each year encompasses one of the three knowledge areas: Humanities, Health and Technology. Thus, the dynamic modeling fitting is verified due to this specificity of the Brazilian higher education evaluation system. In addition, goals to be achieved by each institution are indicated through benchmark implementation, helping to promote a culture of continuous improvement. This study broadens the discussion about the best educational evaluation method, since it has two distinct aspects, the relative size and the triennial evaluation, which is necessary in the Brazilian case. However, these models have the potential to be applied in other educational systems, even in other countries. New variables can also be included in the proposed model.