Desengajamento acadêmico: um estudo dos motivos na perspectiva de discentes de graduação da UFBA

Abstract

Academic disengagement is a complex and multifactorial phenomenon that affects performance, retention, and completion of higher education. This study aimed to identify the reasons that influence the disengagement of students in the Exact Sciences programs at the Federal University of Bahia (UFBA), specifically Physics, Chemistry, Mathematics, and Computer Science. An exploratory and descriptive study with a quantitative approach was administered to 416 undergraduate students. The questionnaire consisted of 79 items, including 26 items related to disengagement and 53 items related to the reasons for disengagement, in addition to 33 sociodemographic variables. The analysis used the following techniques: exploratory factor analysis to define the latent factors of disengagement and the reasons; bi-factor analysis to define the general disengagement factor (GDF) and factor scores; multiple linear regression to assess the influence of the reasons on the general disengagement factor; and t-tests and ANOVA to assess differences in disengagement scores according to sociodemographic variables. The results revealed four predictors of disengagement: academic performance-related motives (APM), with a positive association with GFD; support and interpersonal relationships motives (SPM), with a negative association, representing a protective effect against disengagement; and vocational and career motives (SPM) and institutional motives (SPM), both with a positive association. The first two, APM and SPM, showed greater explanatory power for disengagement, and the model explained 18.4% of the variance in the overall disengagement factor, which is acceptable for applied research in complex contexts and phenomena. Furthermore, from the analyses to verify the influence of sociodemographic variables, the variables course, shift, gender, age, type of school where high school was completed and the condition of being a quota student, revealed, even though presenting a small effect size, that students of the Computer Science course (in relation to Physics and Mathematics), being a graduate of a public high school, of the night course, being a quota student, of the female gender, being younger (< 24 years old) and having a dysfunctional family life represent a greater risk of disengagement. Based on these findings, recommendations are presented to support the improvement of academic management, formulated based on five axes of structural actions, aiming to develop more targeted and effective retention policies through specific institutional actions, sensitive to the identified vulnerabilities, such as "implementing a tutoring and monitoring program targeted at courses with high failure rates, especially Calculus and General Physics I" and "implementing an individual disengagement risk assessment system to map and monitor students most prone to disengagement." Thus, the study demonstrates that academic disengagement cannot be explained by isolated variables, but rather by contextual, structural, and subjective factors that are interdependently linked. Therefore, it requires a system of identification, monitoring, and ongoing actions open to active listening of students, aiming to ensure satisfaction and success in the academic experience.