Métodos matriciais e tensoriais como suporte à gestão nos Cursos de Engenharia e Sistemas e Mídias digitais na Universidade Federal do Ceará (UFC)

Abstract

The training offered by higher education institutions in Brazil is expanding, both in quantity and in quality. One way that quality control is performed is by external indicators, such as, e.g.: ENEM (National Secondary Education Examination), ENADE (National Examination of Student Performance), CPC (Preliminary Course’s Concept), CC (Concept of Course). However, it is important that institutions also have their internal assessment strategies. In this context, the educational evaluation may contribute to the quality of courses, in order to diagnose the improvement and correction needs about the support activities of graduate students from the course management actions. This study is based on the development of an instrument and its application for analysis and diagnosis. The instrument is a survey consisting of 15 statements related to aspects of higher education management, in which the respondent marks his level of agreement with statements. The study had the participation of teachers of six of the engineering course and the Systems and Digital Media course, of the Federal University of Ceará(UFC). The answers to the instrument were quantified, which resulted in a database, that was analysed by Matrix methods, Principal Components Analysis (PCA) and Tensor Tucker. The results show the feasibility of using the proposed methods to analyze aspects of academic management of higher education. The methods were efficient due to the extraction capacity and latent information classification about academic and administrative management contained in the data of the courses. This study indicates that the instrument and the proposed methods are able to contribute to the educational evaluation in such a way to produce results that can generate actions in order to improve learning processes of students.